From 48d35ef2036a1802ac0c8e86bac60359cc263283 Mon Sep 17 00:00:00 2001 From: Fernando Alarid-Escudero Date: Fri, 13 Nov 2026 17:01:39 -0800 Subject: [PATCH 01/12] CI: also run the minimum-dependency check on the minimum R version in DESCRIPTION DESCRIPTION declares R (>= 4.1.0), but no job ran R 4.1: oldrel-1 and the min-deps job use current R. The min-deps job now has a matrix entry that reads the R bound from DESCRIPTION and checks the package on that version with the minimum package versions. --- .github/workflows/check-min-deps.yaml | 20 +++++++++++++++++++- 1 file changed, 19 insertions(+), 1 deletion(-) diff --git a/.github/workflows/check-min-deps.yaml b/.github/workflows/check-min-deps.yaml index ccb5581..08b438e 100644 --- a/.github/workflows/check-min-deps.yaml +++ b/.github/workflows/check-min-deps.yaml @@ -3,6 +3,8 @@ # Runs R CMD check with the *minimum* versions of the dependencies declared in DESCRIPTION # (for example ggplot2 (>= 3.5.0) installs ggplot2 3.5.0), so that a lower bound that is too low # fails CI instead of failing for users with older packages (as with #198). +# It runs twice: on the current R release, and on the minimum R version in DESCRIPTION +# (`R (>= x.y)`), so that the R bound is tested too. on: push: branches: [main] @@ -20,7 +22,11 @@ concurrency: jobs: min-deps: runs-on: ubuntu-latest - name: minimum dependency versions + name: minimum dependency versions (R ${{ matrix.r }}) + strategy: + fail-fast: false + matrix: + r: ['release', 'minimum'] env: GITHUB_PAT: ${{ secrets.GITHUB_TOKEN }} R_KEEP_PKG_SOURCE: yes @@ -29,8 +35,20 @@ jobs: - uses: r-lib/actions/setup-pandoc@v2 + - name: Resolve the R version + id: rver + run: | + if [ "${{ matrix.r }}" = "minimum" ]; then + v=$(grep -oE 'R \(>= *[0-9.]+\)' DESCRIPTION | grep -oE '[0-9.]+') + else + v=release + fi + echo "R version: $v" + echo "version=$v" >> "$GITHUB_OUTPUT" + - uses: r-lib/actions/setup-r@v2 with: + r-version: ${{ steps.rver.outputs.version }} use-public-rspm: true - uses: r-lib/actions/setup-r-dependencies@v2 From fa407d8f1444b731e6299ed89f762729f3a07a9d Mon Sep 17 00:00:00 2001 From: Fernando Alarid-Escudero Date: Fri, 13 Nov 2026 17:12:57 -0800 Subject: [PATCH 02/12] CI min-deps: run the minimum-R job on Ubuntu 24.04 without Bioconductor repositories R 4.1 on ubuntu-latest (Ubuntu 26) failed while resolving dependencies: the Bioconductor 3.14 repositories pak adds for R 4.1 are unreachable, and the package manager has no R 4.1 binaries for Ubuntu 26. dampack uses no Bioconductor packages. --- .github/workflows/check-min-deps.yaml | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/.github/workflows/check-min-deps.yaml b/.github/workflows/check-min-deps.yaml index 08b438e..f095174 100644 --- a/.github/workflows/check-min-deps.yaml +++ b/.github/workflows/check-min-deps.yaml @@ -21,15 +21,20 @@ concurrency: jobs: min-deps: - runs-on: ubuntu-latest + runs-on: ${{ matrix.os }} name: minimum dependency versions (R ${{ matrix.r }}) strategy: fail-fast: false matrix: - r: ['release', 'minimum'] + include: + - {r: 'release', os: ubuntu-latest} + # An old R needs an Ubuntu release that the package manager still builds binaries + # for, and the Bioconductor mirrors of that era are gone (dampack uses none). + - {r: 'minimum', os: ubuntu-24.04} env: GITHUB_PAT: ${{ secrets.GITHUB_TOKEN }} R_KEEP_PKG_SOURCE: yes + PKG_USE_BIOCONDUCTOR: false steps: - uses: actions/checkout@v6 From 0bb2e245557123d4f7c447e31f605f6dd30cfb00 Mon Sep 17 00:00:00 2001 From: Fernando Alarid-Escudero Date: Fri, 13 Nov 2026 17:35:47 -0800 Subject: [PATCH 03/12] Require R (>= 4.1.3): current triangle needs it The new minimum-R CI job could not install the dependencies on R 4.1.0: 'triangle: Needs R >= 4.1.3'. Raise the declared bound to what installs from CRAN, and let the job confirm that nothing else needs a newer R. --- DESCRIPTION | 2 +- NEWS.md | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/DESCRIPTION b/DESCRIPTION index bd0dad3..c984ac6 100644 --- a/DESCRIPTION +++ b/DESCRIPTION @@ -47,7 +47,7 @@ BugReports: https://github.com/DARTH-git/dampack/issues Encoding: UTF-8 Language: en-US Depends: - R (>= 4.1.0), + R (>= 4.1.3), ggplot2 (>= 3.5.0) Imports: ellipse, diff --git a/NEWS.md b/NEWS.md index 8098708..842fe60 100644 --- a/NEWS.md +++ b/NEWS.md @@ -17,7 +17,7 @@ - Strategy names are now kept exactly as provided in `make_psa_obj()`, `create_dsa_oneway()`, `create_dsa_twoway()`, and the objects built from them, so names such as "Standard of care" appear unchanged in tables and plot legends. Previously they were converted with `make.names()` (to `Standard.of.care`) with a warning. Code that refers to the converted names must use the original ones, for example `psa$cost[["Standard of care"]]` instead of `psa$cost$Standard.of.care`. Strategy names must be unique and not missing or empty, and a wrong number of names now gives an informative error (#208). - `summary.ceac()` renames its third output column from `cost_eff_strat` to `optimal_strategy`. Code that refers to the old column name with `$cost_eff_strat` must be updated (#173). -- dampack now requires R 4.1.0 or later, the minimum required by current versions of ggplot2. +- dampack now requires R 4.1.3 or later, the minimum required by current versions of its dependencies (triangle needs R 4.1.3, ggplot2 needs R 4.1). - dampack now requires ggplot2 3.5.0 or later and scales 1.3.0 or later, for ggplot2's `transform` argument and `scales::new_transform()` (#211). ## Bug fixes From 477a390346ef24a1dab7792f614f9b1189b1d05b Mon Sep 17 00:00:00 2001 From: Fernando Alarid-Escudero Date: Fri, 13 Nov 2026 17:45:24 -0800 Subject: [PATCH 04/12] Require R (>= 4.2.0) instead of 4.1.3 R CMD check warns 'Dependence on R version 4.1.3 not with patchlevel 0', which CRAN does not accept. On R 4.1.3 every other check passed. --- DESCRIPTION | 2 +- NEWS.md | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/DESCRIPTION b/DESCRIPTION index c984ac6..36657e8 100644 --- a/DESCRIPTION +++ b/DESCRIPTION @@ -47,7 +47,7 @@ BugReports: https://github.com/DARTH-git/dampack/issues Encoding: UTF-8 Language: en-US Depends: - R (>= 4.1.3), + R (>= 4.2.0), ggplot2 (>= 3.5.0) Imports: ellipse, diff --git a/NEWS.md b/NEWS.md index 842fe60..dc7d086 100644 --- a/NEWS.md +++ b/NEWS.md @@ -17,7 +17,7 @@ - Strategy names are now kept exactly as provided in `make_psa_obj()`, `create_dsa_oneway()`, `create_dsa_twoway()`, and the objects built from them, so names such as "Standard of care" appear unchanged in tables and plot legends. Previously they were converted with `make.names()` (to `Standard.of.care`) with a warning. Code that refers to the converted names must use the original ones, for example `psa$cost[["Standard of care"]]` instead of `psa$cost$Standard.of.care`. Strategy names must be unique and not missing or empty, and a wrong number of names now gives an informative error (#208). - `summary.ceac()` renames its third output column from `cost_eff_strat` to `optimal_strategy`. Code that refers to the old column name with `$cost_eff_strat` must be updated (#173). -- dampack now requires R 4.1.3 or later, the minimum required by current versions of its dependencies (triangle needs R 4.1.3, ggplot2 needs R 4.1). +- dampack now requires R 4.2.0 or later. Current versions of its dependencies need at least R 4.1.3 (triangle), and CRAN only accepts R bounds of the form x.y.0. - dampack now requires ggplot2 3.5.0 or later and scales 1.3.0 or later, for ggplot2's `transform` argument and `scales::new_transform()` (#211). ## Bug fixes From ab1bfdb50185177f069b4d345791f859e945fdb5 Mon Sep 17 00:00:00 2001 From: Fernando Alarid-Escudero Date: Fri, 13 Nov 2026 18:01:38 -0800 Subject: [PATCH 05/12] Bring the installed size under CRAN's 5 MB: xz lazy data, ragg figures in vignettes (#219) R CMD check reported an installed size of 6.0 MB (data 3.5 MB, doc 2.0 MB). CRAN flags packages over 5 MB. - LazyDataCompression: xz. The same PSA samples are stored three times in the lazy-load database (cancer_psa_data, and the superseded example_psa and example_psa_obj rebuilt from it at install), so the compression matters: data 3.5 -> 2.8 MB. Loaded objects identical. - Vignette figures use ragg_png (ragg in Suggests; falls back to png without it): same resolution, about 22% smaller PNGs. doc 2.0 -> 1.6 MB. Installed size 6.0 -> 4.8 MB ("checking installed package size ... OK"); tarball 2.83 -> 2.51 MB. --- DESCRIPTION | 4 +++- vignettes/basic_cea.Rmd | 4 +++- vignettes/dsa_generation.Rmd | 4 +++- vignettes/psa_analysis.Rmd | 4 +++- vignettes/psa_generation.Rmd | 4 +++- vignettes/voi.Rmd | 4 +++- 6 files changed, 18 insertions(+), 6 deletions(-) diff --git a/DESCRIPTION b/DESCRIPTION index 36657e8..03eb033 100644 --- a/DESCRIPTION +++ b/DESCRIPTION @@ -66,8 +66,10 @@ Suggests: testthat, knitr, rmarkdown, - kableExtra + kableExtra, + ragg LazyData: true +LazyDataCompression: xz BuildResaveData: no RoxygenNote: 7.3.2 VignetteBuilder: knitr diff --git a/vignettes/basic_cea.Rmd b/vignettes/basic_cea.Rmd index 238ebcd..60a03f9 100644 --- a/vignettes/basic_cea.Rmd +++ b/vignettes/basic_cea.Rmd @@ -15,7 +15,9 @@ knitr::opts_chunk$set( comment = "#>", fig.width = 7, fig.height = 5, - fig.align = "center" + fig.align = "center", + # ragg writes smaller PNGs than the default device, which keeps the installed size down + dev = if (requireNamespace("ragg", quietly = TRUE)) "ragg_png" else "png" ) ``` diff --git a/vignettes/dsa_generation.Rmd b/vignettes/dsa_generation.Rmd index db7a162..1087826 100644 --- a/vignettes/dsa_generation.Rmd +++ b/vignettes/dsa_generation.Rmd @@ -15,7 +15,9 @@ knitr::opts_chunk$set( comment = "#>", fig.width = 7, fig.height = 5, - fig.align = "center" + fig.align = "center", + # ragg writes smaller PNGs than the default device, which keeps the installed size down + dev = if (requireNamespace("ragg", quietly = TRUE)) "ragg_png" else "png" ) ``` diff --git a/vignettes/psa_analysis.Rmd b/vignettes/psa_analysis.Rmd index 0921ab3..a429712 100644 --- a/vignettes/psa_analysis.Rmd +++ b/vignettes/psa_analysis.Rmd @@ -15,7 +15,9 @@ knitr::opts_chunk$set( comment = "#>", fig.width = 7, fig.height = 5, - fig.align = "center" + fig.align = "center", + # ragg writes smaller PNGs than the default device, which keeps the installed size down + dev = if (requireNamespace("ragg", quietly = TRUE)) "ragg_png" else "png" ) ``` diff --git a/vignettes/psa_generation.Rmd b/vignettes/psa_generation.Rmd index a870dcb..0daa617 100644 --- a/vignettes/psa_generation.Rmd +++ b/vignettes/psa_generation.Rmd @@ -15,7 +15,9 @@ knitr::opts_chunk$set( comment = "#>", fig.width = 7, fig.height = 5, - fig.align = "center" + fig.align = "center", + # ragg writes smaller PNGs than the default device, which keeps the installed size down + dev = if (requireNamespace("ragg", quietly = TRUE)) "ragg_png" else "png" ) ``` diff --git a/vignettes/voi.Rmd b/vignettes/voi.Rmd index 0bc40a2..5b8cf2d 100644 --- a/vignettes/voi.Rmd +++ b/vignettes/voi.Rmd @@ -15,7 +15,9 @@ knitr::opts_chunk$set( comment = "#>", fig.width = 7, fig.height = 5, - fig.align = "center" + fig.align = "center", + # ragg writes smaller PNGs than the default device, which keeps the installed size down + dev = if (requireNamespace("ragg", quietly = TRUE)) "ragg_png" else "png" ) ``` From 944c7072e90c8c474877f88caf952b4b23d3dbae Mon Sep 17 00:00:00 2001 From: Fernando Alarid-Escudero Date: Fri, 13 Nov 2026 18:22:57 -0800 Subject: [PATCH 06/12] README.Rmd with a statement of need, feature overview, minimal example and links (#217) - README.md is now generated from README.Rmd (devtools::build_readme()); README.Rmd is build-ignored, README.md and its one figure (man/figures/README-ceac-1.png, ragg, 48 KB) are kept for GitHub, CRAN and pkgdown. - New sections: why dampack (statement of need), what it does, a runnable example with cancer_psa_data (calculate_icers_psa() output and a CEAC figure), learn more (website, reference, articles in reading order, news), citation, getting help and contributing (issues, CONTRIBUTING, code of conduct). CRAN status badge added. - Kept: installation (now with remotes), the macOS locale tip, funding, the DARTH team. - The team section listed David Garibay as CRAN maintainer; Fernando is maintainer since #245. --- .Rbuildignore | 1 + README.Rmd | 131 ++++++++++++++++++++++++ README.md | 182 ++++++++++++++++++++++++++++------ man/figures/README-ceac-1.png | Bin 0 -> 48131 bytes 4 files changed, 284 insertions(+), 30 deletions(-) create mode 100644 README.Rmd create mode 100644 man/figures/README-ceac-1.png diff --git a/.Rbuildignore b/.Rbuildignore index c67a5e5..f599bbe 100644 --- a/.Rbuildignore +++ b/.Rbuildignore @@ -14,3 +14,4 @@ ^codecov\.yml$ ^LICENSE\.md$ ^CODE_OF_CONDUCT\.md$ +^README\.Rmd$ diff --git a/README.Rmd b/README.Rmd new file mode 100644 index 0000000..ade58d9 --- /dev/null +++ b/README.Rmd @@ -0,0 +1,131 @@ +--- +output: github_document +--- + + + +```{r, include = FALSE} +knitr::opts_chunk$set( + collapse = TRUE, + comment = "#>", + fig.path = "man/figures/README-", + fig.width = 7, + fig.height = 4.5, + dpi = 96, + dev = if (requireNamespace("ragg", quietly = TRUE)) "ragg_png" else "png", + out.width = "100%" +) +``` + +# dampack + + +[![CRAN status](https://www.r-pkg.org/badges/version/dampack)](https://CRAN.R-project.org/package=dampack) +[![R-CMD-check](https://github.com/DARTH-git/dampack/actions/workflows/check-standard.yaml/badge.svg)](https://github.com/DARTH-git/dampack/actions/workflows/check-standard.yaml) +[![Codecov test coverage](https://codecov.io/gh/DARTH-git/dampack/branch/main/graph/badge.svg)](https://app.codecov.io/gh/DARTH-git/dampack?branch=main) + + +The Decision Analytic Modeling Package (dampack) is a suite of functions for analyzing and visualizing the health economic outputs of mathematical models. + +## Why dampack + +Decision-analytic models, such as decision trees, Markov cohort models, microsimulations and discrete-event simulations, estimate the costs and health outcomes of a set of strategies. Turning those outputs into the results that cost-effectiveness studies and health technology assessments report takes the same analyses every time: finding the efficient frontier, computing incremental cost-effectiveness ratios (ICERs), summarizing the probabilistic sensitivity analysis (PSA), and estimating the value of collecting more information. + +dampack does these analyses in R for any model. It only needs the model's outputs, so it works the same whether the model was written in R or elsewhere. It is developed by the [DARTH workgroup](https://www.darthworkgroup.com/), alongside the DARTH coding framework for decision models ([Alarid-Escudero et al., 2019](https://doi.org/10.1007/s40273-019-00837-x)), for health decision scientists, health economists, and students. + +## What it does + +- **Cost-effectiveness analysis.** `calculate_icers()` identifies dominated and extendedly dominated strategies and computes ICERs along the efficient frontier; `plot()` draws the frontier. +- **Probabilistic sensitivity analysis.** `make_psa_obj()` stores the PSA output of any model. Its `plot()` and `summary()` methods show the cost-effectiveness plane, `ceac()` computes cost-effectiveness acceptability curves and the acceptability frontier, and `calc_exp_loss()` computes expected loss curves. +- **Value of information.** `calc_evpi()` computes the expected value of perfect information. `calc_evppi()` and `calc_evsi()` estimate the expected value of partial perfect information and of sample information from regression metamodels of the PSA, so no further model runs are needed. +- **Deterministic sensitivity analysis.** One-way and two-way analyses come either from the PSA, through metamodels (`owsa()`, `twsa()`), or from running your model directly (`run_owsa_det()`, `run_twsa_det()`), with tornado diagrams and optimal-strategy plots. +- **PSA inputs.** `gen_psa_samp()` draws parameter samples from common distributions, helpers such as `beta_params()` and `gamma_params()` convert means and standard deviations into distribution parameters, and `run_psa()` runs your model over the samples. + +## Installation + +Install the released version from CRAN: + +``` r +install.packages("dampack") +``` + +Or the development version from GitHub: + +``` r +# install.packages("remotes") +remotes::install_github("DARTH-git/dampack") +``` + +Installing from GitHub on macOS with a non-US locale may fail with `Error: (converted from warning) Setting LC_CTYPE failed, using "C"`. Running `defaults write org.R-project.R force.LANG en_US.UTF-8` in the terminal fixes it. + +## Example + +`cancer_psa_data` holds the PSA output of an illustrative model that compares three cancer treatments over 10,000 samples. + +```{r example, message = FALSE} +library(dampack) +data("cancer_psa_data") +l_psa <- make_psa_obj(cost = cancer_psa_data$cost, + effectiveness = cancer_psa_data$effectiveness, + parameters = cancer_psa_data$parameters, + strategies = cancer_psa_data$strategies) + +# ICERs from the mean costs and QALYs of each strategy +calculate_icers_psa(l_psa) +``` + +How likely is each strategy to be cost-effective across willingness-to-pay thresholds? The acceptability frontier (squares) marks the strategy with the highest expected net benefit at each threshold. + +```{r ceac, fig.alt = "Cost-effectiveness acceptability curves for chemotherapy, radiation and surgery, with the acceptability frontier."} +v_wtp <- seq(0, 200000, by = 10000) +df_ceac <- ceac(l_psa, wtp = v_wtp) +plot(df_ceac) +``` + +## Learn more + +The [package website](https://darth-git.github.io/dampack/) has the full [function reference](https://darth-git.github.io/dampack/reference/) and these articles, best read in this order: + +1. [Basic Cost Effectiveness Analysis](https://darth-git.github.io/dampack/articles/basic_cea.html) +2. [Probabilistic Sensitivity Analysis](https://darth-git.github.io/dampack/articles/psa_analysis.html) +3. [Value of Information Analysis](https://darth-git.github.io/dampack/articles/voi.html) +4. [Probabilistic Sensitivity Analysis: Generation](https://darth-git.github.io/dampack/articles/psa_generation.html) +5. [Deterministic Sensitivity Analysis: Generation](https://darth-git.github.io/dampack/articles/dsa_generation.html) + +The same articles are installed with the package: see `vignette(package = "dampack")`. What changed in each release is in the [news](https://darth-git.github.io/dampack/news/). + +## Citation + +Run `citation("dampack")` for how to cite the package. + +## Getting help and contributing + +Please report bugs and suggest features in the [issue tracker](https://github.com/DARTH-git/dampack/issues). To contribute code, see the [contributing guide](https://github.com/DARTH-git/dampack/blob/main/.github/CONTRIBUTING.md). Everyone taking part in the project agrees to follow its [code of conduct](https://github.com/DARTH-git/dampack/blob/main/CODE_OF_CONDUCT.md). + +## Funding + +This package was developed with funding from the National Institute of Allergy and Infectious Diseases of the National Institutes of Health under award no. R01AI138783. The content of this package is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. + +## The DARTH team + +dampack is developed by members of the [DARTH workgroup](https://www.darthworkgroup.com/), a group of researchers developing open-source tools for decision analysis in health. + +**Package authors:** Fernando Alarid-Escudero (maintainer), Greg Knowlton, Caleb W. Easterly, Matthew B. Kaufmann, and Eva Enns, with contributions from David Garibay and Mark Clements. + +**DARTH workgroup members:** + +- Fernando Alarid-Escudero ([@feralaes](https://github.com/feralaes)) +- Eva Enns ([@evaenns](https://github.com/evaenns)) +- David U. Garibay-Treviño ([@du-gartre](https://github.com/du-gartre)) +- Hawre Jalal ([@hjalal](https://github.com/hjalal)) +- Matthew B. Kaufmann ([@mbkauf](https://github.com/mbkauf)) +- Gregory Knowlton ([@gknowlt](https://github.com/gknowlt)) +- Eline Krijkamp ([@krijkamp](https://github.com/krijkamp)) +- Alexandra Moskalewicz ([@amoskalewicz](https://github.com/amoskalewicz)) +- Petros Pechlivanoglou ([@ppehli](https://github.com/ppehli)) +- Jorge Roa ([@jurjoroa](https://github.com/jurjoroa)) +- Mathyn Vervaart ([@matverv](https://github.com/matverv)) +- Ervin Wu ([@ervinro](https://github.com/ervinro)) +- Yansi Wu ([@yansi5](https://github.com/yansi5)) + +**Other code contributors:** Szu-Yu Zoe Kao ([@syzoekao](https://github.com/syzoekao)) and Michael Fairley ([@mfairley](https://github.com/mfairley)). diff --git a/README.md b/README.md index f466e97..d389daf 100644 --- a/README.md +++ b/README.md @@ -1,54 +1,173 @@ + + + # dampack - [![R-CMD-check](https://github.com/DARTH-git/dampack/actions/workflows/check-standard.yaml/badge.svg)](https://github.com/DARTH-git/dampack/actions/workflows/check-standard.yaml) - [![Codecov test coverage](https://codecov.io/gh/DARTH-git/dampack/branch/main/graph/badge.svg)](https://app.codecov.io/gh/DARTH-git/dampack?branch=main) - - -The Decision Analytic Modeling Package (dampack) is a suite of functions for analyzing and visualizing the health economic outputs of mathematical models. -dampack is developed and maintained by the [Decision Analysis in R for Technologies in Health (DARTH)](https://www.darthworkgroup.com/) workgroup; see [The DARTH team](#the-darth-team) below. +[![CRAN +status](https://www.r-pkg.org/badges/version/dampack)](https://CRAN.R-project.org/package=dampack) +[![R-CMD-check](https://github.com/DARTH-git/dampack/actions/workflows/check-standard.yaml/badge.svg)](https://github.com/DARTH-git/dampack/actions/workflows/check-standard.yaml) +[![Codecov test +coverage](https://codecov.io/gh/DARTH-git/dampack/branch/main/graph/badge.svg)](https://app.codecov.io/gh/DARTH-git/dampack?branch=main) + + +The Decision Analytic Modeling Package (dampack) is a suite of functions +for analyzing and visualizing the health economic outputs of +mathematical models. + +## Why dampack + +Decision-analytic models, such as decision trees, Markov cohort models, +microsimulations and discrete-event simulations, estimate the costs and +health outcomes of a set of strategies. Turning those outputs into the +results that cost-effectiveness studies and health technology +assessments report takes the same analyses every time: finding the +efficient frontier, computing incremental cost-effectiveness ratios +(ICERs), summarizing the probabilistic sensitivity analysis (PSA), and +estimating the value of collecting more information. + +dampack does these analyses in R for any model. It only needs the +model’s outputs, so it works the same whether the model was written in R +or elsewhere. It is developed by the [DARTH +workgroup](https://www.darthworkgroup.com/), alongside the DARTH coding +framework for decision models ([Alarid-Escudero et al., +2019](https://doi.org/10.1007/s40273-019-00837-x)), for health decision +scientists, health economists, and students. + +## What it does -This package was developed with funding from the National Institute of Allergy and Infectious Diseases of the National Institutes of Health under award no. R01AI138783. The content of this package is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. +- **Cost-effectiveness analysis.** `calculate_icers()` identifies + dominated and extendedly dominated strategies and computes ICERs along + the efficient frontier; `plot()` draws the frontier. +- **Probabilistic sensitivity analysis.** `make_psa_obj()` stores the + PSA output of any model. Its `plot()` and `summary()` methods show the + cost-effectiveness plane, `ceac()` computes cost-effectiveness + acceptability curves and the acceptability frontier, and + `calc_exp_loss()` computes expected loss curves. +- **Value of information.** `calc_evpi()` computes the expected value of + perfect information. `calc_evppi()` and `calc_evsi()` estimate the + expected value of partial perfect information and of sample + information from regression metamodels of the PSA, so no further model + runs are needed. +- **Deterministic sensitivity analysis.** One-way and two-way analyses + come either from the PSA, through metamodels (`owsa()`, `twsa()`), or + from running your model directly (`run_owsa_det()`, `run_twsa_det()`), + with tornado diagrams and optimal-strategy plots. +- **PSA inputs.** `gen_psa_samp()` draws parameter samples from common + distributions, helpers such as `beta_params()` and `gamma_params()` + convert means and standard deviations into distribution parameters, + and `run_psa()` runs your model over the samples. ## Installation + +Install the released version from CRAN: + ``` r -# Install release version from CRAN install.packages("dampack") ``` +Or the development version from GitHub: + ``` r -# Install development version from GitHub -devtools::install_github("DARTH-git/dampack") +# install.packages("remotes") +remotes::install_github("DARTH-git/dampack") ``` -Trying to install dampack from Github in a non-US locale on macOS may produce the following error: +Installing from GitHub on macOS with a non-US locale may fail with +`Error: (converted from warning) Setting LC_CTYPE failed, using "C"`. +Running `defaults write org.R-project.R force.LANG en_US.UTF-8` in the +terminal fixes it. -``` -Error: (converted from warning) Setting LC_CTYPE failed, using "C" -``` +## Example + +`cancer_psa_data` holds the PSA output of an illustrative model that +compares three cancer treatments over 10,000 samples. + +``` r +library(dampack) +data("cancer_psa_data") +l_psa <- make_psa_obj(cost = cancer_psa_data$cost, + effectiveness = cancer_psa_data$effectiveness, + parameters = cancer_psa_data$parameters, + strategies = cancer_psa_data$strategies) -To solve this problem, run the following code in the terminal: +# ICERs from the mean costs and QALYs of each strategy +calculate_icers_psa(l_psa) +#> Strategy Costs ($) QALYs Incremental Costs ($) Incremental QALYs +#> 1 Radio 16,129 10.68 - - +#> 2 Chemo 30,901 11.07 14,772 0.38 +#> 3 Surgery 25,110 10.88 - - +#> ICER ($/QALY) Status +#> 1 - ND +#> 2 38,492 ND +#> 3 - ED ``` -defaults write org.R-project.R force.LANG en_US.UTF-8 + +How likely is each strategy to be cost-effective across +willingness-to-pay thresholds? The acceptability frontier (squares) +marks the strategy with the highest expected net benefit at each +threshold. + +``` r +v_wtp <- seq(0, 200000, by = 10000) +df_ceac <- ceac(l_psa, wtp = v_wtp) +plot(df_ceac) ``` -## Vignettes -dampack has a series of vignettes designed to showcase the functionality of the package and explain its underlying methodology. -The vignettes serve as a guide for proper usage, and it is highly recommended that new users read any relevant vignettes before using the package. -After installing the package, vignettes can be accessed by typing `vignette(topic, package = "dampack")`, where topic is a character string corresponding to the name of the vignette. -There is some overlap between the topics and functions covered in the six vignettes, and they should ideally be read in the following order: -1. basic_cea -2. psa_analysis -3. voi -4. psa_generation -5. dsa_generation +Cost-effectiveness acceptability curves for chemotherapy, radiation and surgery, with the acceptability frontier. + +## Learn more + +The [package website](https://darth-git.github.io/dampack/) has the full +[function reference](https://darth-git.github.io/dampack/reference/) and +these articles, best read in this order: + +1. [Basic Cost Effectiveness + Analysis](https://darth-git.github.io/dampack/articles/basic_cea.html) +2. [Probabilistic Sensitivity + Analysis](https://darth-git.github.io/dampack/articles/psa_analysis.html) +3. [Value of Information + Analysis](https://darth-git.github.io/dampack/articles/voi.html) +4. [Probabilistic Sensitivity Analysis: + Generation](https://darth-git.github.io/dampack/articles/psa_generation.html) +5. [Deterministic Sensitivity Analysis: + Generation](https://darth-git.github.io/dampack/articles/dsa_generation.html) + +The same articles are installed with the package: see +`vignette(package = "dampack")`. What changed in each release is in the +[news](https://darth-git.github.io/dampack/news/). + +## Citation + +Run `citation("dampack")` for how to cite the package. + +## Getting help and contributing + +Please report bugs and suggest features in the [issue +tracker](https://github.com/DARTH-git/dampack/issues). To contribute +code, see the [contributing +guide](https://github.com/DARTH-git/dampack/blob/main/.github/CONTRIBUTING.md). +Everyone taking part in the project agrees to follow its [code of +conduct](https://github.com/DARTH-git/dampack/blob/main/CODE_OF_CONDUCT.md). + +## Funding + +This package was developed with funding from the National Institute of +Allergy and Infectious Diseases of the National Institutes of Health +under award no. R01AI138783. The content of this package is solely the +responsibility of the authors and does not necessarily represent the +official views of the National Institutes of Health. ## The DARTH team -dampack is developed by members of the [DARTH workgroup](https://www.darthworkgroup.com/), a group of researchers developing open-source tools for decision analysis in health. +dampack is developed by members of the [DARTH +workgroup](https://www.darthworkgroup.com/), a group of researchers +developing open-source tools for decision analysis in health. -**Package authors:** Fernando Alarid-Escudero, Greg Knowlton, Caleb W. Easterly, Matthew B. Kaufmann, and Eva Enns, with contributions from David Garibay (CRAN maintainer) and Mark Clements. +**Package authors:** Fernando Alarid-Escudero (maintainer), Greg +Knowlton, Caleb W. Easterly, Matthew B. Kaufmann, and Eva Enns, with +contributions from David Garibay and Mark Clements. **DARTH workgroup members:** @@ -59,11 +178,14 @@ dampack is developed by members of the [DARTH workgroup](https://www.darthworkgr - Matthew B. Kaufmann ([@mbkauf](https://github.com/mbkauf)) - Gregory Knowlton ([@gknowlt](https://github.com/gknowlt)) - Eline Krijkamp ([@krijkamp](https://github.com/krijkamp)) -- Alexandra Moskalewicz ([@amoskalewicz](https://github.com/amoskalewicz)) +- Alexandra Moskalewicz + ([@amoskalewicz](https://github.com/amoskalewicz)) - Petros Pechlivanoglou ([@ppehli](https://github.com/ppehli)) - Jorge Roa ([@jurjoroa](https://github.com/jurjoroa)) - Mathyn Vervaart ([@matverv](https://github.com/matverv)) - Ervin Wu ([@ervinro](https://github.com/ervinro)) - Yansi Wu ([@yansi5](https://github.com/yansi5)) -**Other code contributors:** Szu-Yu Zoe Kao ([@syzoekao](https://github.com/syzoekao)) and Michael Fairley ([@mfairley](https://github.com/mfairley)). +**Other code contributors:** Szu-Yu Zoe Kao +([@syzoekao](https://github.com/syzoekao)) and Michael Fairley +([@mfairley](https://github.com/mfairley)). diff --git a/man/figures/README-ceac-1.png b/man/figures/README-ceac-1.png new file mode 100644 index 0000000000000000000000000000000000000000..6a9b816b4cfeb3a44165962d8305b21ef76fa6fc GIT binary patch literal 48131 zcmagG2RxVW`#yZzva_;6NJz-uBC=NrNo31jA$yCAGLoIW_ueuRl7x_~knFwJ|5V@a z@AbT%=l491&+Aj+J+Aw{&T*W_aa`{ZH5FNW94Z_HLGb0}r0*jL>U#u1DaA&GpBz5D z69a!?8Y;?4Bj*?Yy>H5ihM!!rlY8WdAO!6f|D$x-<{BdiJt8kHso|QmGU4JuGEa?c zER+v2Dc74n!_8V5;O=2bd>bxEWX$%`#|Qf*J-uJLpI1B##-T6BpaLR{2go8TAw6E3Hxr$WsBjb*KgO4(#LHSI#C?F8{K z{Zmuvx=x3!_o75az(5lVO>5o`%CEslm!l!x>uXg5i7( zwmglV>k$(+Sjd`W*gc~=caW}i&y0aw4YoS0G!})70lQ16qN`jwMVf4}Y@|3Gl5$a% z}poGdKb4bH1|jtf+L7QFVqI$plSrlq?IFC@j`LHmJIr;>%4nfcl^gSk&{ zc+9>SbE-KTP?&x%eAF*-?ja@BR%<_N`uW`ly=iy?_O;1P&FSs!Z5|$;+1c4ke&$2D z>T+^&Iywo*m^ww@Evopff5gzg#Kflx2nd*)o7>p^a}hrdU5|bHdwN16BAlO~(2$YE zrlhPj``*jA93rh%qKD!0`Sa(twzeNtmfQb*{J7A;&!0cPepMUkYi(`y^F!Uo50IMq z30LtszI`%D_<6oGgZ=3){D6{@a)n!YkUz2Lex%po*1W#gVdwMvKH1Mg)1?{SEMo;r zGf)cH=xS?gXSJ0x+S$`Q_hpg~GGIFnQ^;@Yy-zGMd813v0|N!g8SLiJpp=g)HmC8o zlQ8ROW~afbPh$y<+fOLD8pI?YVQ-%B`#LrvNJHRk{GE+aKI+tKnz&=p3DqaAQI=QP zy_B3yn`s7PI_bNPbJFmThgl<-FPR z*x6Z#nfaqJ_rs-usTT{$66YrsE7m)1r&x3zYI&s_>o?*Xdc-qiuLNX2(@P=clyp6m zn!)Ad{70!lhXd>rT`R3DluQqvtoe>KxG$JLb=@7EeEke(zl`HJpyw7c( z&XPxAs_hZT^zw4A{T>yB*`MyV@siB<@9FDwQ<9QCbA3MCUQlL!Fj4<}Lcn}5yKf*v z)^?&E0})QPS@@Ftq~2*}dEf~}Xyd7T|4+X5TOw0^XNBHtgd{J#vEkJ|9Yf4s^3yIi zmNzu~HT}VVA1=~q@qTrh!_m&-6%JhlsnR)}kJGpLS!=&B{oxh)Ew|&<8oExWmY^2z zQnMrjE6kwELH>-*4$kkEvaa5w7cbhoDHN3~mG(};+p8@0rdF}jEm>*l#{EfmQE@~0 zw>Lv7Csyja1=+tRb|cEPi~bDooXPC^=f`43N68y?-CI7MS6}FRQ&zN81mZ2;Rm_CCq@Gm{#m#f0keBDY20^6)wI}RBWQ%aRmfkgJ5rgCb>nU@ z!Bt_W+~w@{87>MuFE6ja$EH+gV?)!uE!dZG(nu~E!41VKVJ!V9W-kWo0czE=rA28d z8EIYl30GX@wQ*m}79VA*x(SQ0Q7xhE0svYs^Gj$YBrs1&<7m29d_=w6$gbhusPWq|kyZd>h zIDy5nTg@eAG`b&je-y0=)%n?fHmy)%b|k!yVOlXZV`t}a@1pP3V*PJTuV>!jqUjMl z{!xc@n<|_wHgtBsdB9wQE%qn0wfXBJVNgUAvTy?Z6p5mkR-Hw<=t*T3orLM==*r#= zqJPyac*sO-PiV-W*fb}`O@W8k?kdk67#PSMuf(yy%UdhQM2u~ru)M|i&|gp5z6euX z-7}QH8w+oCQTHMUWD79KT#>IGs^wG7PZcF*{)L8Yfhn1*%yN#*5|Nm5)m^|46T^k5 zs@R>)Vjm%Zq~f2ae8IR#g8&OnUo7N#4{l%4#`CT)22~rO?RSP4tqc7EJDVup^;MJw5TwKntNBA5YNb|%; zR4y6s^Zvn*iE*otvBcY%u%q z@UZTjkCpXVXDHd4rLfZ|cMsMooeY$l0|wb_P|q5B@xy9b{J_{we2uK+V1ZcQ2o0 zGU3nv*71L|^NT-UiqT*TkBIoaJWvp^YT;rdkk`NG#0F0zvUL^l``7ZBvHhAT|MvXJ zT=k5o$5Bj&omcb=A60zx%o?_ME}Xo=4f#>CX*SxR#TAj?`_9iAcgMSk(A8NoFHMkn zHYG{`eTKGccWaBdqabRAxI-)^u46OGxX~gQFw`2C3D(TA{Iv+^I(2n*X=!Q2iN3zR z7#LZRJ1H?SLsga|hGu_S4t~8_9j~eAy?y&O1ntiU-8qxSAt|T>rf;LO}FBw-WO~{za6i{932uT_|xCKN@q=S#aU#Zep*>1Y}jIc{z0oE zk^hHQF>L+D1+iZ@2{EW{c|;!n~c>CzYng9BqH8jVo-zW`VZDn7iAUyG;A@b%ODY_S@JE z?JThhS|l37qgMQo`?p^XTbNLM?CJ_uB$|E}3>OB-m@O8Nu{x=I6pjB^UTgkDPY;iP zwqGr6H#=wSumh$3UT{|l>-3`7R`u0;3~wOKIV7bSE_V@LHiWk7@8|bZwf+hzX+Usr z+)wtKHx;w)2L=R?KBgnP`0C_XpJ<4ibwQCNg8ISmMq|T>-Fn~w^^eaUPEFYpj%&;E zQH8FrgRjX)rL#ozVk2%$f1iPoQPJjYOboBx^qpqf7ycievn57YB3nYwdg|K3Y^QIn_t4ZcpCH-FVPRaP!Dg?CzU?wdYr8zpo7`PUp$t z_NyY@I)^OFM;vkMKdN4J$x~%CYt1Pg;Ux^R)I9rp&HIYjD>iMdU%DUI!W8YRhHMMW6iV4Jbt1B@fxs^Xl*!*qiPHBF39 zy=J@p@wSejL$SN?#oU@pT-qAvH#BuqxHdG66?w{X!x0ou`Xy;O zKy9VkuWYJmrOfK0WA(J7>f^|pKL~O-b4etUTdv;H6^6p5mSy0}8TeIT)V9^iR|x~s z5jC&;`zw7nVb;Iz)L3>!OnK0>B3==F&yc@hcuD-8avg* z0A>F;`pF6?n(bZ;wdgQhyK}D)7*&kX5U6ZRzh;t;V%n!irmwGGmQja?AwKOh)2>yQ zkHc!s{9t@0LBAH=T`O%ToJ>EU(R zVl|hp7|4^#Y!U~gv%sKoQ8NTdWrS!5)+DX?6XDwsK~KvuEB-ff9bzx{J#xqN1N5Ts zj<+1b7>Te2yN=~c*zV9!-UH}HmiY2fQogMoHaT)6Y?eg0i0AR%uCCw3*2@_9 zo=4uxI9w_y`8o2g&oo{sTlt-};GWTGtc0lRDh4&qIx}c9x3o4nhO}T02=F zMzcrxQQ0pa?Aef1vA+mdLtT6NrVqWw)!@W0Cm#i^{~?2XBK8fk^d+vaYRW!VZ`Z%5 zI9IB)RqJJ?jitPtkyVx~1bb#Hs{U>Cw9u;(wwATV&??XBW<&Vz1?Zsk!koKP^`e8IXXj~2ACVeJc(f;p=CNmYS z<5t!E;>T8_r3Q3f2WiKd1suG}wqxb<(DrF0yiROePCs(tMMj>vzC2m}ME(3Wv(Zg4 zxPp?8X1@BB=ITmHO29jOdc$b;B}snJsSN*uR7g#2v7h}oT%bhFWXGi@77Ft4Vv?2 z$VNtlhd)2v%}Mn-I$m!XGYL%(4Hd*^Zcn+&MHOC5NAvKmdd%P=JEio{l2Cg;k5jxB zdO~R^IuwUl9#kZXsSa!AI%SXnkPx)!2mp<~kI#tp{!SYoo=(HG!4mq57HBnOBkF2U z_OphGFWDLVq5vPFt3bR5d`}iF<;2`zwPs6@^RO@1Ff>% za?O4UZ_i%mH36CaJTkx{cwI_9kvV8KC3szeqL`3@l6=$-8Yj=b2D7gckqyXhPUGc|G$QzILAv2#37+q>C*FOR%Hr--%?Nr{Y!)x z8)qJ40zh3W;=pUSJ85-)%jO&aqbbjR_P=~m`0&`55Cx#rA zBv^#Bjb{BRS&DJx__2dIsx5XiewO|ai3F6~#%Cvow#|p8MLJanZaZD7^jbG{_5@N| zFPPm=)SJ!(w8;M*kMzI04c(ejR6G|e&rtT~v0LWsnh6Z{stbSTYD9!vBRPHdj?^#z zzq5NI5&1zri5yuEzbedfsH)TNk1pm30rrigTtiuC&f^td-2ct}DB9CP>eBlppFHhH zb)bz-V&aP42++v28(@&B#E64tTn)^sn!U_p@#Cy*)}*gmnh z9VslRY5DEA*@NR0p*#O(raN^QUem3d!?ng>tRuSjuZ~Nx3Va4_ z0jd(GPLP<|6TR<4uc|!Wd8>%^Zx0Y3;JzA9wt6A16|%}LM~bJs&zlApolbWro7^Gz zZn3cBN*QzPds_7f`K^C1N(9b~$Bf{ZN-QObBIugDO4Be)sNs399kt}!?8Q*5%Yik- z7{i>x%`#NOE!sqD{n!Vtk4EWvl^w(G4)StJB7jEpCM}!&h|oV=Cpf0HC;TNQf`{07 zmjJzDA3SrgRZeHICh4=ShjrqaTQEk@F3z|X&vgBIXIB@R&%KNRbKb6qt0K-TqoHIs zRZ_)ok-tKDy85FE_X@|!-ncFBVcKKUx@*Uyuxmgr9`^}MHEw?qS#R7{$x%t31d3MN z^GGf$vieej0mWaP1`KeK5H>qE&w>;I_R16F5H^m@%U7oos~OFp-j)+V2%r#}%&>i(}Z}x%s0PwY|Hj(+_6bheIIW}^qJ0Hm-F%xKbGkZgm!cNrAp2$MZnCnU{-c^ z-gqS%e}K1ZAd`IV?{^WTp_$cYC)e|XlyJmr+t{InQE+kBjZ;1wTBBi^9p2VVFx_N> zOQaFe$mkNL|A(rrjhS{_a`w>*Gs1Q*09?<9KLA@DWbn7(+S}Ny?JN%7m6D+de3qJd`^oSP)Kw0>MhYS#yOq&0 zfF{mB?5tqYd7o@%%13i^b6X7O?*e2{eaH1-(WyMv36vU@97ou6t2-G3u3K~RIfG*v zvV?ki0Z^78xKlj$9>ps`lc?dd{2oTh^BeA7qp01kG*SMj!EfAS^TdI{JbK`dWr#XJ zbU~Ru*qrSb-ByBLfQp8eqnetMkkDge_vdR`Anp~^RtO%?^+~ajtNlqr6^;vEX1gM( z@q(J%_d`1koAgpsQoNu}Aft^h9J`NAjg1Q**AO7(<>hNX>!RPi!$x>aKf{<#{>tQo z9|p3~MduBojcm@~;n7h&ZGn2my_uOAUgcB?Z(2kpMf4H^<<8hU!4B#%ow&!r+C)P} zMuzvS50zPf5thF%jT+^gVYU zZvXCsr+P8o9B;|j<}Bz-6p%PxsTdd-Xl`z1lBdGMEYj@V`+|lf_GxZ4d!L^QAb7N* zwJOQN$X6KOeS8q4*5lAYK>-Ir$HXizI)bOc&z;4{C5cx0fKNhet+pOVo_@=e_h>?$ zCp`e2!pzuMx6#%9U}HvHT-?Mq47o%6O`2)|^P-D=nxzu3IT?_pn}9n|GD0V4_g*=X zAK|k9CH;Fw%#oVUVopb|$*roS#8z5b8fIXS+31|(R#vQ#(PqzvIeoY+1z0~(b2Fi7yZ1{a|=i~V4_@BMp2M^8D(JwPe$vskm+BYuWTDPgHluV-Q4Q? zOvgb7Y+oa$p!mC?b|N?&y2zp zN-y#?Q=TM5`qz(gpgo7M<2dy(eLyg~ckdoC`h`vtHD0=J5?Kehyq~sqVSax1qA>QS z$jHdR1|zSdqoZSD&=J#$y1AHT+ooM|L@7m-5`oor?Z!8-HnrwtXY18DbXKY1MH1@Y z3PD@$rgGx_H;TLn(&ae+ooI_Z-T5~ZU-j9#Pl=uN8;5%lP^g^S^OFm ztA~mF)(Ie-!mu&Qy&}Bh(qao?2PMUVmkJFHEh9bM&)=VrPMqfI)gsLTG*nbQ_f1r$otMi&}zBStHD}zt3gwAZv6x@W7PI1Yg=k;F# zC%1LLGXK9`fFQTsVXgjD2?^&F?LaB}#c$WEbi_=0Vqku`xZl)av|{xzFe6WE#uX=< z0RF&IL&L%@`WzI~QB&vO_BYYd54M^?b==NYPSmNkicd)R0Yf%qiud7soK@1p0!`1e z!vzBa10Yx7Ds`JZgiel+eda-|bN-Mh{G6@p!&oS*CzHsVKW9gkm6goO3AN{oz%YkX z3;e$AZy`T?7I<>2A}puI(oJR0i36a;-71UWrC>TQ*HQhuckgoQ*0yrti8$GJqa5gY zYyyPx7c$7zJBh zjzwbzx+7Luze3Avm{?m9e9E7AQC{zz-|Zki4D3#H5=fET_)F;bw3!(CIjk4l&jYz{ z^K^e|1j;1b8jC77zy*e72Ov%w2^^ek_d_Gm)YOEg?<(QS(bW2Y^g8Gx_gQ7sq4*zh zX5C-B&?{4dsHBDs-t8VuWv&(f1!i-UR_l%Py>YU#vQv#O2B)UHdtzAQtwP+LyrgZY7j~HY1gbCco~1>+;siI9hdeZW&Y{lIAH?;G8$LXM0(wdXi101v zXXB?H1)YZWbbY{lUnJE^-QDwp*+}S*ab+lo)9R1dtAFstyyk}}s?~rq;^gA`YFd$} zk>3?H9KD*Clfs?YcRq18n;bPB5E!_-HgUW;4_VH!w>s{!{ablN*_mHv|L#i&kqRxn z*clLB0SefTjk7ijv1_;_QK9!sYETr}Iem9tz>vKYBD;#hd$E?-{bNV-D zy9U2~qvAE^a{T=n0(^5kvgHWqa463ps*A6dlCNpASWe0gl@=_|)>w~M8+Aq030MGw zgJEK3RuP>Rdy9vs0%mg)D>XGW1c6^aWQ$#)3T~gVP;7xV;){kgN;%rY=XG-6GFj(M z5uKb`pq}qBxBnPbH>2%ZBuh>!_$K~9+XPJ%22}_ly=IRWPoGXVzSvfn7&Wbcim^1; z9S!^epmUHnnJ{Qc?oNMIDy5WRxTdmv$L}U z%XuLz8Sr)186fwFo_n7~RsHv`yiZnK7XiMfa+@enN-8mvAVVekQ+(GiiJXQkSMFbs z4L@tuKiqu@a!+7zFq+R7o4OSD8WVG|IypQfQNUIl$nMTmD&!}7PpXHjqnmoO9KA=A z$%WQpLCzkolEFBFE z(~mzb%p3YVay8t4yPVU3iqhG7BRQ2$qvd($`j_$tb6=-gdTXdqS|uM8a57``K4he- zDY2sID1n$6Tr`$la(OYmQl5BnaspP9g9BGTdN9L~5)((gefwMd79okxY8ySWsYHS9KTqa25p>!k)?4n_H!x8SbDX9~{nP9W!`Os3>#rp0IVy z_C~TL^4j%yX*aXat)Y`_P)e9zPZ0=Adq5w6iak$vM?id_1yK-M<<`~~08ibew~rn@ zf}U86yOf@u{tY_yzE+SyrOxUF5>86`30Kn{eWSv0N7Ly0Pzy!E$vS_#VEF^FI07lY zFQA|GB-4ScGe19Hr$6?PSUw5?D2=k0Sw3M?GiGH`m2heabRDo%GsW1KR31GU7n7JsO+~d-Pf0=X$@!Pv zU+v89)9Y){)9+bWEWsRgw%>AINAL$wSK^yDA2=S6cBVYc=iVDMuNfEG!(x&zw~&PK zW6UbRmgpVsM4J%H3q(d8nRXJsc}(*x;rBs#X^6e|i$fe#L?tJBw)N$Td!OQKHjj4l zb}$mjf*0?9GZO8QXh_25AVilh4^&xdY0>!6>#h%^h`EP`hB^s98OtFUyIS}lM+L-{ zKb5wsB8k@JA4e`NQqWvNU{XbV98wt@5`{on80P*Hx&G}H&I)gQdf#N~3zL4IZsG=# zpObtE2A?yZ*w&B>S<5(EzL`C1*-3nc*gZ_-aqS<`sv=Z+9A&`r^3>#+qzyBobbeFP zgXimM32~YnGf}sUVVxWUN$4u+B?Of|Mk<8%OyqzZHxO<4-k77brGd+ODRYT49Vo(4A0Sv96vQ_9`3;0yc{C?-L_Z!@?LN0N=+U` zLbiRQh92yer>7?%A$w4_!@|P8e*FsQTWqCV&RH-ecmf^Cx_`Lczkfs9a|!_$`*%% zF5wtbItE7Bh;%s)4$kP8ctR3N%5Ih9-V4VRVD?^)md%gU zpuZE==?8*h>;GYTVWDf20sk*Yn@=Q7PXv>c)c%{dt*`x_&!BDk1w8Iqp6^m@%{N$BISWs~at$_#fM*zwQ5Uw*QN5 zbM$R#4<-nZ-e5x;8@Ga(IXT!gv1+YOSIqhb(1$(4`SXYK2_7=_J2ix%2gM7DY$0J{ zlypQy)9dT{_h|O)u5)p1%o>e{Ch>=6qqHgsSDR$>Vq@1?a3DS;^&%0Kv!*!{bCron z&J}|;PCuRqB=efcN#G-v#kAX(cx$oDeCo{jjj3r6xG?_Xp_`UZ z9rr?xkB@8c2+!4TuQkn0xP;Zeo32+B#>de*DicEky?RXl!AhJ@WTfiTi?8V!H(Rm> zB?4XjpxA3~i9)L079U+E~b1S2*>JH*%UKdQX-V^`XB*v9Gtu&dhXL{$3RJ z38?GUZ&|OFP5InKKzO*pOw6~Ep0HEUZo}g-$>XyBt@Pdm+i>g4uhHfJ3gqSGixYX3 z2YFW~OK#EAKZR+Z>@fTQ@I&cca2RBbEW`GW-Cb3T`C`tG%Y2+K-?i-F1}P z5$Ag^My|*uVEQNqjs+-NI*sC~Bn}uj-lCx4_iwjw8g;mZiI~@JKDiz4I>i!SoeQJr9^c!A*I--7f)bajM6b{OQpy44_XOdjD%( zq@kvEpJ~I0?$#(YvQSXyff-BuU>Yr6$r&CBG>X=o#p@dyVBdSz6{BT9n}entkb)h} ztn4;A&(8sa!i5F5E)6_dsi~<029KYM-#*mS>22M3WPOa;6GO%-W17P_UWny0&OA|G z;cZHRGrB~8s3aWSIz6#nT-XlXLY{WeEu)7XC$FHRrxEvhT<4CJ&W`%z(OLeKZpu5r zP^9(oZO{FwW{>9Shys*H98o9V&WF)3&K$jkL~mrThh6DO$e&G^qD9Q#6S&pu?I?Sm zjl`X$Kht3$Rk~RRXHY3J8mpymT=B0x0A0$L zwdSo7|F~{r*$A39z4LUQI|D9Lo)iLxlN12r2>G!#cM@gQK;?Ai=U!9V) zbM1eYsUyD|(P`IzKgQ_xoDQCl>D`DdmM%jsq18Rh8kA`zz2C`--fswLm3sny5gMk( zmyD)dI!ru{z%3yAb@NSL2RB_=?Hu+kku1khgjx1E_0Lk}spOlZ%r1f2TUN#e{F2X! zLlgw=d@-Nc9y_`fY4*gHl=O$VV0yVx8_?VKR=Z)5krF^90bAP6|GL{4@DGQGT#MJK z%~bP~HDey#Ef49!%j?<2nHvljPAjwJ9|5)%`90cM203xXv8NTK8}SM*#_T$hqM z;s87Am5iA4uE?9GspSrshwg^EA){^_4wA2sBu>fuA8J}!wTB9ET(2(jQGED7Joiu%`2gBqg=k>GVb z+w)#xXGPSPcj_2AWUsEwj^EDodKqc^QYv}b{W<2zjz&|i{zrCNcf0nfF@aG*afa@K zTlGz&ne8`-DUnJXy#0}w=LC~k)l#0upB__lo=4|#Qhd#QxGt=G^GsJen2^r-Nr^$l zSQz)%BLo50tI5ya>)hPskmEo)xfB^BewbMHgsAQjv(YOz2J4UF&k>EHkY;WDg0?m( za87d4rSIfe=7qZO&Qebe7Lj*Y$$k-g_9#z^N2`E0exn}bXb9rlAa3+ zH!vbx+}sW$#ZRCfWBLG_1Il-@=nDrRqk7{wN0xVCLcGdr{uSIpz_`_eEkaB0#l?<5 zgj0yQ@gkX-nM&`tW`G-kVxj|*dh9U;6fGu*DuES`gCoLU5ZK2@^Yzkk`mO-s8PXo%EeqK>{%^%&THL%OVQ2 zeHmH*R+p8Sbe-^6tC|vFLcNUAkwBXw3iM*4C0|MRuKUy8j+usr2@6OZ{#t?97Re zh)cLU&EmQ_Wl>ohUfxCpG0d$A-A1Be+DwB1#IvT5@w}$+$KH1Ft)X~yYH=BLd4+Dp*)h~<3gxD zZcS%j)9zN7_T>Z7hlGJl2V=q7Hu6wG2t%Bmo&Cm*1rQ^e05QQ?0LV{TQc{h;U`cl6 z^^xbtZx_!F7|iDuz)|<`CaFwTc6Pms4u4|bTChKOC?Unj>H<`YG{L6N)q3*Q*`TY5 z2gPAHM$4=Y+g} z`$;57Aa6b+|GL(2RDs`I{!_ZxRE=v#HF+=EIL~!a|7g{`IDaR8$UhrVxy{>-jnSrZ zZ8!CmRjk9NVW)l)&}A1M+$W~q9kqJF+-|{jk3n>X+&_0OCzVl;Id&;l$Z{KHN#y|6 z8R%9g{r29huw1-bs8X+r|w91;k&JFhsKpN>?t8`OPC11skUz2eG?x3CLNf4je~;O8{}8dzMRLqjCRix{pzH)Z7g%!T`I461PVvz-YcPDu^BN9-9HHDV9=Z6+HJLEZVP%xX*`hpvX*a6KJ`csUTmsdbw#IqF+y6F^#=%uMmSI+av z`Y4sn=8gQO9d~}1`aqSewsLzrHh&FyufZm9^d|#u<>|WjIoOn7rY&jQx=2vk>MNa{ zoyy6=&wq7@J`4v+j8{#eelXXfNBXWQ*KKOm^+tmD7{mgzsan?xdDf9n>i&9*S`fQFlE5 zft8Y2#>ywuZ|G5Muv}YO{!#O0keDbKP)mA7vl%jG<)M)MRxZ=l4RqYd6@CUS$Y|P9 zm*F$ep&%7QHnzC!T7cOC%C6eS*WDljz-R(e=w~=+hKWZ}>b}2L&Y1<>0;DZ#kOImp zD!RfbXD!A{{J!CyfM_LgI2S!MG<15rf1#n!1W!tL9*FTZ{8^_91^3zrq z7p~Bp_sPEzHyjIHyF1%PTgx78_(z2KiH^FYj`~?&x~Q{na?Iz11RR7sCquT#P3SE{ zS4##54tBbkYc!vIh8J0rByqJKSuKAUQ}?U$6y|kK9Q#iv^MTc`0`VEeH+e-9#~p;@6q@0^-C-mF(KHOBm;U{(z3tf;*SsQ<6GqBJ(rBxS*mGBDVSQSSZfy!mgW$x z8dlE+{m^_nJx~i|olqru%f@pDZ=5$0kG~G!!S*Pfuzij|pZjy>K zB%rX|`qHSx+pgX*X8Zf63axkvTX#xiU7>M+3coWh_^be0eEH+&w zAviT}h#^4udnidX76OBA;}1#V4@+KVJ2$)%kyFQ47is>p`L5htssINjiv}8X^;X_z z&VfjYvr$i8>L7#B+mD`DqN1WgoqV|IgPq0(#^jW11tZ^cWNcR_RXrQRf`~I5-zT)s z*m2>J(9)iP99(7zhYgemiwWxCkcN=`FLaUBXY!HWbsPmWv(NKyjnJ7FYaG-j~3cOi;C`R z0fKER&a|`VJB=c#JKV|RSyqqFVyJW7w)gkxcuesSNIOpf&QUqpmY^a(PY;x*=Vzyo z?snFT`F~&&Gf5|#3j3;~R1mI5;Rsiu`najOx?=x|OE@hfnF4e5A2)fI^cdM&zG9rE zE|5R&`{GvAjW*P2H!z&bw<%Jy#f!)s9rxuan#yj2YM4S59LVD%V ze|h4bCvR2oyGV*`sqTD*OSsV2VHlFDJ=QCZ?d=_aZsKPLvu$nX*?e#+76{bSN*9wG z&t$M);J~VA{5I$W1>EPK_Qu{7zqR2p7rg)$6f{gcJ7t>t`QVLeNNDKY{XNz4XXQM? zNa4dlM>zCg!YKt5pArsc8Jr9q%m7tu$|-Joc44l<(dmS>BiYJ7bOY&XCAeP?#HSwmM^-zk05nGGvQ~+ZFR2o?HRIqJ zJvm(`w2~jJENI7R=Yk=F)7koSySDs7q?Rh|!gKw__bjo}-#a)D@O018MtW8E*B;v> z!F0eHyN@d&qYhwYjidvP&Ao)`1l9!e1gOz0NNL7m`4lODTSj>Y_%o8Hb197IK!Wdv z*~HE2um|OzwvOc(IeA^yem<&LQ&qL$rHWO!KF+KfHP_DZhs{Bolgw0~?Q?TxpSbt3 zF!Ga|{If{&T_z%*$5>bL_F{je3z8dy`8kb>@3z);An^PBKGZ ztYYHjpHE$%#^b6ZpUrA(U=19L!>!2nhLPmH=nvIpN-(Nb zZ!ViiQUBG_waae@KmOi2Tu4<-6;D2LF{eVn_*9(n0(mXRv~r>WzLE5FeY!{zXRy|4 z7kw}9H;8-XS@Pl%*hqUPzFl?i1HbLfq2-k0?L?YaZ%oZ`^HLq1EYs9Eue?N|pR%yp z9b_n)Kt&wVYV(+D-R(}Z!~GMF19>$t6apBEG#fDQC}2M;XsXapUayZBb)C%&cOodB zs#i59fwM-=zjzY+=9{OBc`(=|)Kzm{|JMs3Au*!8Q^?k6g4Yh`fA^>GAKn?{ZReu# zFBJ)cLj(UgYIrABuiw!1P%WCNn=34i2@Qd9XC+PhW>m;=ECHf zro$vSyE7LVo*UT?qV4FiH0?c(xEoANJIf;_Vb)2!7)VG>$(gfN!@yxHHrJbG6p3A0 zEtj_PJ9S&h&L(|ZZD*Ns%F6iox5VwAclW^|lV3e-8$eRCkd`Z7U+aD8vZVK*@>qO$ zRiBnOe~APuE%d@q4aVh`776$FV0mcgD)z?&=aGhn6AboI7%7=~ju<)Gt~deUoLP94~aV4yRfI^;H&<(>>Pz56DXB~>jU@54x$7WWM<*usa@>sV0pE6^c3B%=qZ1{# z%7L^1jMRUb@g5C0yDsM(zDAt_6aL<3IybA6qrR_n2MXSH4_mj_-G$>4<~8(wA9pvb zzpK0|wd6gtS*IhupQH12>=tziGhLI*cB=cLTZ0S}Q8j+Ph)}+J+*i!olFvI)XRgG( zsz$+vPuCE}9FY()JR43Dd5&sP7FTItbVW+(n$fClvW2qJ^yus(K-`gm0LKYI-S7KRon8nexwP&5xbK#UoZ^)RphWg1x-Poj;#GuGSCPj9#nN%j3yEh0BiPn4rV?4uqnc z?aU#2Kx6dGf$9tDY$8;YyKhkLk{|CT>jlym^SCj*`2&`S6hV6icSq>M;Bcq|CnB8F z2lb`h`JOm9Ic}CTN~sS40fL8sE2o(8ASNc}5(%?zliPE!qQUv~U!(f9?{J=%H@K|R zBBd>wUo&RUhgQ-Vr>v)tuOC1+c9eJh4v-Bi< zC#}!s+O=zl9dJ}Izr;1+OMOxlEH9ZmTj%>|V`ximrPhu= z(Hh>Im6cT?YYUbJMxjGc`E!OXU{3df7t*P+sKe|uc2e$F1u<%{x+v8<9zI&LWI`PfSv|&` z`6ws{tTbCjJz6WYZbnANiIS8Wog(h|&rMw-&JlM%`EQZ@*5C!V(r&+nM4tZ=Ls1qMwcgZMJ))|BFCC#vLm{Vjr|TJ<-AsJ`!kby~Ch8JrIQ zB}p*1nSINbZY-if%j%=!34Fa$;?C$H8o+6f}9r;L|rX}|1+a|W5dn@if!w@Dduzd2tlPl(`(h>E3K z?-Q!7jm6qaN@JV*WF@B(YH8YF@&~*g`L>g!YlWZJ;$o?X`%$N3lJrwd+5-}%Yq2TW zAD$EFv6Dwq3#@_h7Rv_+%J=WzPk9~VBcS*0f`KP56STMx&0bA-6P|9A0 zO!0AXIrsC}66Gm-ST}$ZwfvUZ4W3rrb|t+QFSxq$dm1t{hlC!-d*4z%^0egrAl!7~ z&;wJh1OtQLttCRnxZaA2iV1I4Gs>#RwkfB(zvIW5pI8ST4UG5oqw5*c8 zke^JYJ10K%+t3YYKb%bBfDbfCkbBVZWjv&A`Nrt_6B4XpT(1|(Wclv5;rRSjm-|e_ zlz5oVPkTU;h`ZXf6mOQYI9fLQ@-ibYRUe%FoLD~?dZ7Az?dJ<{(gL3YjU|=K0JY`r zfmmg0#Z=lYok}pqf`8j|1l-emIKox|9!_)0^}SS1Lu-%nEd(ewo_nd;*c7+VVB_Ik zuw)HEtA%Sv8Z;u#N(IBe7ZxmesVWA`_BDGr1^Ss5G8F^UQ7)^eI#K@2az?v$?cfm$ zst?@m)F!9cbe7=uvh}G=z9&t;E=vtZ@}?_K5cY3Ss1<2~?{)t|y@E3VUPp^J0;LAv zgaCY`#mCN0Am9pNiUODE%a9QGuojR7fnaG7M}f3~y15D`=A51Pz((49;J4uW1#ljz zOr7v@1)5C*e5yp@S|qvTfyAlvUZ^A-8@o7^E=+AZ1h^D}v#eLR;te8|QR`iQ6P&WF ztc$v5t9(KW`;!V~uOpnR7{ph(58~Z!O+}Z+$|B&_F;aIot|F@A5GD`N$ z-Wef6Hj(UPrR*8mWQWKqdxoU4_ugc$?7d0$miasH=W~4j`ZpuV;!nV3Fr zhic?-YwG0wBXHikqr+mL`pyHKi5LJWPCMWHEq4SRG#$)qHE}i;%+L}&N(qZgJR6H$ zPIhEgMr$;5v2&a_PUth`;$KIZKgj3)_~&=d@~{mTytjbHyUcJr(BMQGeOHqE#hv~R zFLk03g9DE1R`9^~W$VxfS5^@jrZr=s>$=hRugvRa_EobVZ&}J9B!cGMO*C9Uf``qv z$N2o&GYL)WE8JEpx;y81@g@&zhN9lJ4OO{jM*G!lTIX@8`TogbqLLO0G@9EniZhqQ>1^^f zC+aH9i!!7y5%Zo+DUZ+VW3goNTO5Bp<1aM0LtS<+Ao7W@VK870z1`h44l7;IHHbL= z0V0ma`|2Xdrq+BgBR)PJa)iKQ10=+O(Ptavr2;Hca&bNx)`tkgIl+&ey24^T7O;bbSahuwwPIGS%XX1E* zgz;avbt z@l7egVMXzs;T(AIgY(f4J_I4S*7qp648m&y#x<)wMhgtRhh`MRsG}e7EEj=-a&Jk0 z&v`9>YGYWH>9Yj@N&vWXwZ1CqFy{U6HJ{)W=>ym{!NEzYG^G!Q@Z;{ zE*u(P8_U@6x{BYsmkIruWU70(Ex>EO2P(Z~?3U6RI8j5A(e~Y2HmckLY&>B&kE1$Y zU@4QH!kz%>JC&&G8*oXay*r*b+8BFJ_iyUQ_wNML``GB{=;mdU?rx3`X8`O)!gKIs zKQ(gz;bQq(1F=8-Mg>-Y*aedS`rU?uQc*=~OB9lUK)~6y5j9+XoO!kH)z1>xQ+DkR zJ**tH8$`B{*xqTkcV=hDFnf4P(o9dS zlJ4g1-#`qrfT>1b7irWR6v5ux+lx}n;+iG{)BGk~#XV7>y~-8hH=zXJ5Ra{B9YP*EPs_X@bTjd+-@5n?kuo_@>7qi85 z{m2Y(&TDA!cJ&O&?U#P?OVD$HPR2U1Ki=hy(>!uPMf+x1N+y&D0`&BP;A+ocF?)|S zcLKCL0~6CtN%O>5c?R^y2!y4T6{tjP_{nj|mJ*-Hm6PZ{iz-f_Iv1U&m-CtK=913S zGOTg#9VEh`5@%4DJUKl-7xAcgk9tpxAi38RyleFDpXX^|rO!)gf2C~0TdUy$B^5X> z&&wR~!k3=M60=J0?&N8(y+>W*f(}L@Fn)ROW>vhKN3C^N!F6Y+Ns(Do=eZ;=Fb( z8J$>B6Xu%x&P8CNkrg~dnYVS)+c@?^Hb2RDX_+TFhZ2z7?AdyDhtVMT*}XfgK_)s_ z>8CIJCajqhei!ksp^ElDVIbO*oObY-VU)hua*E>xu^W9T;~XFmu`jaBKLfK(8yiMv(cgq_JjN| z6Uw@tR|u;eip|IG;N)(=!}4vC*sBfg{zvU~x&WH=*tj^bw_EWi!b4VSKXMri@I_ib zTwPs3yTyaM0Y$T(o}Q_x>4AA*;q8mzu_&BdKWOyH?z`aR4@7%E!YiE=VDMfs;}ou< zYUioZ)nQcVQ?zcocMC2qC;KbC2@kuSvp+3|E_0JH3HEYp8!mns-+@d7dWaC6E=CdB z~2_Rk?D)hZoxWo@T@BbeEH2tP> z!TfmGy;Lh6V-ZweiOQ6pibHlAeJ}EGWs2^!*-m(-7k(+~V6C|Kxl>JD9r;vHg!AJ1 zFI5(@QAqyz1Aht~x;iLuTU%S9RfV1%u3bvl!yvc-x`2dt+`1ezy7>lmFAB!8zkUTH zRm)x^@aCY3H;60&TOj=VVV33X&zq^PiyuH~p`6HXFcMKmXu8P>dukdu0frX#W zpBi8ygkJRz?Xx~upX)sJFk85RXuQa8S#zesK%C~z8K*qW*2DfcfcR$dbz|iKPgs(GEmq)Ky_mQ8%0wNT+C!n@OT+oo3lc-O zwUO#=Osw>GW$X@m? zSFs^pf(X3tK~Uncj5nM_pomFMP6oLz_~yE|g(>Y-J_HB5ULNV5E#P7J%c`5jB10eKYF&6$00+E&o~ zfeef}JC_w|O3GT}PNH_6yrbG$qSySI9 z$)|o2W3q;r=v*!DeM+s%lnYm=e3Ue5`XagNvcGmn;mSx?wr)U}hX}#rdAbXh1d5x5 z*PC6vb$Vo!8PsCMuj{UgY66TZ^St#%-~N4wcs^Dd9QB%rfvEJD$X9|tup;6jVa+|) zFfJ?SL5{%w-sev##oGuU2p5Gea%--k!LY%LG*G|t%>vSbQ0@3bgY6MqE;cscGtsMd zT;BrKlh7OBut9qQtLz*q_Oj6jv^+>BCKs-I;RCs|NcS8>1B4B*`kFsglC?Z9@0PhSLKx+RANF^CixU&! z>O`@|_eXeZ=YV(*zhla=!pxNPX+gzbC#|}^j4_=f_Z$PWLCap}4X!U^I&t2GD)O&M zX|sqDqG3%P5~SJgLFud^E1yM_jzJqk#9f*KXUa&vReqexkVTihEs z8toq%q!_MP-kqI#?v~8>|70G?s%#8AaY$-G3-?dU7tPHnFXS*Zu&wvh5jhF-5)*qW z(scD!>bNRbEp`C~kw7NH(H)T%y_rjZ@X4Pnwvu~l@;?3btB%H|E&1Ov%b6sNGMcZiyVD3A5cl&!S}v||hEDLxtgNo0B5LdEz?$^{#6iGRBFhEca_b+024sYU zs#;pH@$r8oH_>vND!q^xCTOfcgF(P@7f}d@E4=QBy{m99p`c+1d5~wvgN6u#8n|Rp zkwi|@*qHWL3O5j($91>iq+3j`j9@;XI3N7&Sn^@jEqEbb@^Ss%s6^JMudG_ynx9MD z%cEj_#ok)tCL5TIY~(Tx^_rU+SSAz{H216g*wdsGaSSur=)w#1{qpjt<35!|NT$W8 z>1&I`OqSSh&qEf?6SG+ROy|Cir+MkSuKeATjb^}d*gs`d9aP|d$=cXqVq6#}%M}yJ z@c5IE0uh6Pye)&JWiCl^H&x7s4@WABw|ogCsU@*ovFPyKu;|d8u;G7b14QxZ;6GRD zR3U@q%8$g%g@l<8+3tK*Vu}Iq37x7+Vi#X(dgRAtMK zPh(A3Lx@-dmW1jIcDA-Se}~nMSQ77q&Vs+E*VL#ynsj>qBQ6G7z`)d04I_)XzCKjt zG8lQF1lr^47ky0=RL{-I#^w$ekD&GV8^MG%u#moNL>Nt`sPj+zA^n3Wh>3}b<8hwP z%zz%2m?cQnc^Y1WX%R=n5~%`_dfF!Wo#{8zU#4{GOvw=dr3OsCV3)C< zm(3z*;o&aJE!WB()7I|B`H80C^;P>ivhrm@poweX0%_oDvRO+LTnvhqU5=LfxOPQ6 z&KtL&>(%Qn;NloCB1O_P_bDN1Yg+!LkU<3v{0r?gOuDX`b$PzLfqRCPKtO@W*<;82 z3g_lwbPQp*5ovyLbE#GP$^g%fu2UtWuTeY!&XO!DAE|ZwE&WWgTL_6E%J$|hZ^PtW z^N%*^_ZG7`s~QKSNYvEIq2;EP>Q zAZk07rJZhU3kBEaUK^Z%3qn_*f)u2ll|e{$99$qxnjAWd&14i0c7bmR4GH;4@{jZ3 z!)mBwKtpZ6z>hUn7{4C6)~ypX>y!#PdX5{^>l-<{N9t$)SPMfb%I3t#L>Hgz`1U`b z|EaEM{=D)eVd%!q_xxp6rE2Tbay0FGP%I7q$;;8A%wP#z`l7_NKKkfYPi3pcwcE-X zaas!c*evzKtM^YYO{8z%7U)vMGxT&1`1Y4ek&Tg>e56Qajx_ra1>t$rR%G?>p;0ZO zy~ueFg)eQZi8ndymCb|LVyMy79>6=s8j^0E-1}>sCANC+^Ps%ygYp+<_WXvpE!jJr zHA7@OMB!^H#s=b1Zu<01?@?~CD#)7zXedS+&qQJ>5UH`!#VCAr!uE?Mn!2htPCJ(x z+)Se=URp#mQInK>NT}ES>f3eAh0jRWt|$k)U`jum$Ga;8$uBC$k=V2baZLkT1SuHY zdpe4!^#6<_5Xt>q=VoU4y=Ql(>9sXV%rQ;?&fX z<>scTnm)2nsJ47!YpdxhL_@f~FcXcFvKu3Va*S zRqG`b10+^Cxz7 zDaM*my?3D0e({2>Ui;=C1Ani6?KAsh4TZiH#c+Sgx0$|7jTCxTxYyZah2uK*lRN`<@M_eNR09Q9@jHi+v&X zvXWQs+gHrm?w^KER<|yhN#OTBcl~-YRN_t)Fw~_dWO%jFZ$=stL@N3&!}IqNYsx37 z6yoNAY3A!k8oE}QtjB-SH&WCjnU#O+=v~-Weo{X=h{$$JU2_9Sd46fAh5$YB@I_jj zf%|&a`9ev=|7igr3*-C(3OXeHJtSZ~4_cl|$Pj_59b!hF7w(Dndmj$UK`x1Kb(PB- z%Gyu5PREX&|K6_f_r4Fb2n_o(AS|#c;Xd2hvjMXRUOKD|gnu}CRDr2{-W;2zbKI7x zx?{3f?D6N))Z_7SMu{%xEdMSBDVnpLIL%j$2%DjQ%x?w^w_Dw4DDS5I7Uo}bPjg= z<(j3v5oe%(_24I63|8Clq;X8GfWywEHsn4kP&LCm-nq>>flGU9?z5n4{E&cTfQNb z8P7$pR#nraK31FXHicuh@7te$tOqk@LHHnPI+|9!9&Av+tnjfNF7b^-oArOT5ZnAocCM5Z#r1MO0plxlg2Z(HHLy_x)RhG zy{}Sj1as6_@7pUXK93JH`KNnD8ouHhzM>OrMoQOo-+G4H-?!Q^-EYa(nKP1skZMWS z%;YrbHZ>jkv2@B~hj=QB+$`62Zd?oobbG1bq@d#sj=DNCzwb|Qu3sVa>R#YcJ!PZY z4S3olitk1(NxdKLU3f*^Ar&4J8HC{A6k`;o}TFvvxveHaQ}j6yLN2! z(ojaGYOm+VC%)__M-M|48I=|Ao@|j*6K{EV+z{q@#`yV$Ill{kPx1FYUv({wtHV#c zMup$TUZ2PtQFII0c41=&mYs3m4fKtFPIHZDo||j4MLd>=mwUV%GUN^V`tnMK%2Tep z3yvN+3q@p1P(PZg9Ed|8{!BA+h~yk9+Lq1BeYdIzj_xz+ySd<2`D*>@uVv`VsIO|T zItnXM{3nIk*CLMAo~G5>()d4P>$tHaZoZ>mF-9CPDUp&+?ELWg#B}n{EnTjOsTv~e z!Axyq&vS`ze~xbKBVv!loT*x!7{cEVLfXD$qM%X+V4%qeIZ{M27>rrh?LiRgnpWQVmYi?dxcsc{twpE2e>_Z11Hed#`PIA`)*_<^GORK!qXI-yZwq>-+O z5{SmiS2HLG(m>^;C@nn=b&cmE<@&)^R$FfgltEq)9Jm0~5b%tlY;g_uX z`gT`0F7XNzE~ZeR%DbOQOXI?UMvz@=o-uxo$x_lEqtE>_odq8G0yR4G7gHrq{rTJq zCR}-viL|@Red+3{@=jXvY7QNO*WGs>LI|`dt-P@9D`vCyeK+dWH9*~ybOPIXXg?Ev zrrFr^tI3v2lfPU)S8ry}`@}%~PITlds>-9eS-?#@$3`pB1T2jog7>L;`Dfp|5y`w4 zm{h#oE`{(>lEw8RIO_LEWnx;rfdM}cT)uLruJxLhkYY=oiU3vOpsVO(6z{&Rks_&U zqI2G$E=jR?TJ+lsEyKhCE-(mMEkpM$3q9Ex61-nu*H8#byb^w{u%${SyU?>i6`LnC zEdM+m<~KZe6wvBgMf*!QHa542|6b6fO_!}qRBzHgIwRVhh-nbFT6!N;$Rh5%e% zF_-pp220IbxiNy>ngZ9Ll18Oh!7RoRYKw^oA(n-oKwv(Z(!tbw?>SeeQPzO&#=k zS&$giw&A|NKHPOPbEG*y&h@iev%ov*w9)n7?|)`UzB7+^QFAElHMKd5pNcbw6=%HSj1Y z*+19px89?V>8~5-+)yGz=c#%d-_X!pHg&$(J;N7aSimA!9*QHcfCnOYLf&_>UEwxW zZrV<4S=PG)4+x_a`?-7N6$1~h)KIAg8?mtGExJ4RjL%~^ zrouE;_-qx05iHYP@dR)bg^Od(ZBniU?B=KtHTR8YTu>1&ECv;}VxZ@o(6DdXhaS`HDWhZC7c4FNoFZxO~Mh8tifJU?Il>cT)6B``?zKx(0RxR{$JBEA9to?dd_3ljmc zqVpOnj4hg}^6zZ6$FdSw5F7tGBB!FF;!)pyfZJ(bpt_U_HV>GBPHMus+$Ej|cc|#K zT&#&IiQ80akX8K6dEKdBcFAh;Y zS!u-1)oAOp2Ju5{MG18I+JCqE4}h6o9;ZWo&NeF^g{x71WMsaeesg;}rq0PPR(65h zfBTh6PelQxG!!>6|1PP-FC3@Jgu7318Ee}e)D?~v^Fb>4;c@cu*|2xLdQ@?PtQbws z! z&lMSnJwI;vF&Nq}U%N9J*5vz%y?^tZRd(sRdw|UMrYhkRXx$J&87;MoRCY64 zrM#;1c(XbASbapTC^Q$yFZi8_r;89J*qjPumzwwar2X|o4U3L9dV=M5%nbjG5arhR zv~trgC+u*L9RDrnvg17%Be_-Q_~vr=dutkv=S>|KB-T6v`T%bIYJP`hc|;>sP`Mqqvs~FPGTM)m-Xce$A_VRVV5dkwfCGZ!US-;SycAI)9HvMFXBZ zwKj~d+WHY**Fef6GpV73{M{0zZxVTSks-4tCM$qg01r1_=i!p4bq)}!V1#L2@W;&e z4VRuwtir};J+=(%7%GIe%rAed={#>fT9NumQqFy?qA`(Z5m_|hZQ=&v4vPAR-n+f| zEq9M7m8WG?#QxiDLdyKZ5I5-X_AM|)VN0hV7n!WEy9VJC@-yCFp6{32RSnmfMdRFO zPQ?2dDH-l-tv}DIzM7`#koM(A8Y-d(_tN^TKS;5_bJ*-6`jkK7=LeWghE)Ol8VthW zO|S+U3LHLh@(h06+pB}QD`<4%b@$A6#I?_kBd7!(6y4yW=HTYu0j;E)o3IZ7?}Imd zFTUOIycRy5eu3pvtRHYP{lN-T#e=6fli5iIeszGZMF!skCMM*ZEr{y@H_9?H_YqP7 zfW3jQ%}I&1FqhP~bGg>#-?FIt4H^-Mpa${}Qkr8vG2F8Kch4t)Ay3QdsA_uW>``DY zlINN>;~(@x256sbLv;`#c4lsEoeM#BFxVqWrsytSe~b8Nnn;3czu!Y&|#c!>jTSP2fhL8SRWLo9Yc~Iy5%*kPAXK!g~ z0g&&Rrsl{ucxnaw^nv9TN zIs9c5jQhE7HFh+!z5J>-Qb%#%LlA#@1D zG2)QE2iCED(Vf$rUQ>^QUO|{wh7?ILr@FoZ5fg4p1464l_2*m>s)TD$Ch{u`p24{d>1^i+A4qWC7xjkR z&-2u&A3uMxac}^`K^tfUEk9_*+}zy4$@UKpkkmIdHJ35g&6~e1%WSJdJf7O*ZFTpD z|8s+sg+TbZxC8K~nde=yc)mtEmiSW-KG@ehfa%wK z1eM5LYU){Vrz0eQMp{~2VGn#%1R-RK+tRj1fbLDKYIq_i2VFs zIEqw${d(M7rDjKCqER}BKbQK&V*B4jofH&tjskpAo!CU}Z(;%h2&AnpVo#{et_9xg zJ3M%pz8|MeX7hM_6R)y)%Ks-llJZB&nJzCb0!aB_be^c8s;Yf<BjtiEIBRCtqZ9x1kt0Hk$KSQ&wM{7JWine^1%{ zTIkoovlW@A?|6069cAu^FF{#XfQ4u*&MlGbG37(}WUQ|pU+UQiIB)6kJMlR0ISbphvRZz*kO6NkZSEjc%*TPQQGKw)s%GCvmft`CSK4P?NRi3YVK@kX=<9paj1|pC6uV2j2O>&(;U7tr6L>J zvJD88zQ`WE%6-Jusv0eioyx|8cN4|7`uknAmzxeR{0Ln1`!2I_|N8^N%r#^V0VG61 zWE~-50EB1A>meS(^9Sq)^m9gJrKS1fZ?_bJ*bpQ~$zty&ZRdfggV_PDbU^l4hsiE& z#t6Yg$EKISDAHqTOE#G2{KUpqHJeBAApjJXE64ajl11WOHyH#1GKg{s76>&~N&GIF zm1#4ZL_q*GHwR(iFDE0&Q?Yz9k|q)_cTYl1m4nzz;GZW7H{AEH+r>S zK+S#riimGds)5iBp+P1GVizs>p>W$blrXsmY38Hoze`H`{TU0y5Cc^g6%zwao{S7z zvi7anx^qapI6L0T(aL$Es0dL9`ydbq`F)wXu*IpObW`|kH9Eq_NP_8JyJ+$si>%~t zlcuRvC)vl3YOWsry}@0a@0&9Q1Awe+_rkh0jJPj#%C7LXcI@sGYx{pI_=Tm~ONA^V z_{H%kPT_@qzWvM!!PYS|*yKb}`} zdtQ1zI-+yFvMbHt)zAF*b%&Vw*6A~x{~d*qfp`~Sc&6Bs9Qp#o!oF+#1USBC_G$Y( zKjhAUTO#5>@qr-Fdj|(Uv&AOLltseyxLj;#=SpflJU2ds_!$QFeblO{x12K@cx(L9 zdV=J0XsjrKm>CWk>4q%ce|L>a)si%ZA8p_*sxwehK?DX=H^p7PtHTBVjC8!Z4N0a- zk;LIqDHqrNlc%fraS?T&9~Ac`g<)x=1j=v;wN5F?gt|}UQ@;&$|L;pJl%#A}9Gp%)*3tY(TSNK5&jT^;-o)nZ65WJ}j z#t73A&>jB%@c7BM@?Yh4{Ug+M8ImonTqJu~S*w#~Czr{8NUC#B?vcLH#gkdQO@80b z@ZZ=_SADWtja?b%HA5>;o}DY)0mU80E-5#VH^1%I4`nbiOIAjPk(pV1w-F8O4hWx7 zV!B^C<#u%pbh}3PCOuvdY8Mvw^wf#JA}x2`<(KO}utkac(QJC1uS9CyDcs7ZC9fMF z!agQWj$x7Gp%U<4!8cGyHn8pp`Y^aVXCRdYP9+osAt7Oe=za&V+~|DXhH9?L-m4m& zZ$91_ z6cE6J%LnoUGCosKidR6z;w)Z#!cNfK@E`^Mmg>nS)s}_6-a^i%RMJ7e*F|GM{^PK! zC=%?zx#BNAW*m{_(=|hM$m;PF)SZT|3-4v!zhL{F&imt=_NUGyznHPw9+!z&a)khR zkSfu*&Z9KN{78X~Ga;1VcIj@v6)#7GopGd`KV|soe4&$ykLXM(+^X5PR=zfH2~syHn6SZLT5T#XQO7x9s%BTep*hzksBsTG{s^QQz2#uajQd4_ z_!7JWij03aL;+j~H>lWkkU5Vx4VT9MU^zLUv9WEM{ z__XJBd-|zoE}8^{Am9s|TAHyan*=v8GO9*WS5!o#2DE~+v$M!3X&F_hTL3pi22Ddm zw7R-_XlN)3!p{m+9WS)BOzoOTh>7vYc%?H3YhiqyWNDB4r<7-E7f$8Q<+k--)D$rB z2saM%{^{7Srafwl(>h8SZW3&5`8gkO;$!wIZtWD;{b_}bPE&38uaAZ(0o0>6VIL+& z0cXs}#I!wCQSj}XDTpVLmx)2a4xE#);SmUAOv|lXTKv*b_GN#En(awh$MEECl*aMM?KYaj5`B?%i!5YWcL||_6PWW5>&U(NF(ias z%u^`P2!=GcZk3He7O*{~_kJfcB&YQ7D8Y#@8$k_y9+cmp7e_%z%0K97viS?E3;_+~ zG&C^&1T<{OkQgK0Wc^}w#j6}V(e^#v2l~Mw$I4se+ZOuCOMdx|+Y56+vWz`jqOI`l z4CKEvXa&s5d+~qPK^2>=S`e*rfLH%L+UOF?c)kXr{ZrCv= z_1nsz9YCX2Zw6B(;I7Tf&krPki3rD#Q*QfJ66)edkKhz5s6SH-o^-!BwU0s>O(J!L zH1(C&5syrHNcj_2aq3a&9BW^a5MXc$h1aMHVj*6hk3JDamkKPiTGkWo7vx&Dop&?1 zAbyOnvdHn%tWByCC|R65@&6{rQl}pY46&QRS6TAA9Lqm2Q0j($@ zhWSBaV|kCI58}Nf$GZz3#4gPid1Jb}u=*!CZv>j?M7LvCj@LahWqm)RI%z5R)U-?UKKg5g2{ZE zQe8@)9hSt?0ysp&JN0uPhZOkwQrwXBq-BI>rkB28N7zOiL;+Au+5RE=-nhnI)-7aX zvnyGJ+g&V)a+ei7*yG0y5cdQyN#m$uAm(FH?-_)-Lu@W80yLik@~JiWfwsH}wsT}K zp922Y=c}NSg9!v%NwD>tht}8DTp;3*4wehz#DD0jpYvdpwA8oR?{|>+2w^M8fN*9H#qF$2W|Z<>iOI4Q5_qA*vL3gmT)RsLi{UAYK*XA~;SkbGheo#) zbI7RDzsD*GGvHDm8)z{F2R1-0~N!Gw|o0aI%-oO+do zAY2bS&+y)D!kqC?ix>*b5zhWU zEr0~AXd4yY1#$=_2!;R~69W%+-@`?_&aXRlmi%h3sh?W^*y#St$MiraWNy=3Rnd59 z@|*iJnmv}qbGF-Wlh|g>ab6A>xNw?B)4%U8|dwYlSRl1eqKy`j0Vo@0?_UO?& zKt_&x;Y6^6x*O^m8?L-Zcdz~Y^bK{!d!&E9TdGCl{-bD*d+^!g-LRQn ztx9Be3=wAF``)V=SJO~2j|evU!Gbx*kI|S)&ok!sio8yk&t}qGPdvI#(#KQ>$l<~! zkZR#XDv(hTNJA9JbwQwsKm>T}is)X~GkjJE#0@s~12$)X;QyXAVi5Oyjxl?OLLv+W|$ysY+H;U@cP1W1xNw;Iy?xu~8G6La{ zv^*+y=p{&yIPZf^Q~J;DF)#(I2Oc+PpK@I;8|fOz4#^?mj72S|%$ti}8o9vbLaN3L9mci>PXF~CP*y%}ScWe}g6J%nwGxtY|Pl_VnUm0MWbBV?*aA6`NJMCXaF zxLApqwYB2d7Z&&s6rLZ=(2d;hehUbEYXVGIBQ-cRaDn3$(eL&`0EqzI;4i-=2BYno zdOenDw+TJiM@Nd6XB-+=TT3Xy?S#h{Td|Z99Ff zHhl_HCGXBKx~>WyO)nJvQk2!5rwfBWbw-mWSIX>_$r(a@lx)LjliSny-6TpsuSSBXgBbvvl;w^c_?z- zt)YJqX@JE4)GE{PM7#HOudBzu<#}gEKSS;hZ-Yl~>#Gk_w(ntYdU-v3)l?XneHGMM ziwFd<%ZfpI{{?~QygiF z8K#JTDM=vt7PebTpNFPn9I|tIYkt&qH=s-9ppm~9ZOAog5axESFQ*CAh!5KQi^Tm% z%u!csga5{3N8U>J0*UH}E%iN}hrZ5b{kI9g?dsro6miL)r*4k|!k_`>W&l>ap=(!H zWhsnC5-+QB9hq_VB#?g(Q(u&uccT4!o5Akw-sxZ=zWf~FbAPE~@w)Zzn#|q0F+8^h zz+6u7Y^|D*`hmN5L99;DQV3>dz`h{O1c|3&WDds1#hrlF9~eJ!^0NUcd@yEPgRgK~ zO=2GOXCc*9erYSG)5%?t&&oXd9U`8)TV{WZ>{%Bb9}1bad#Uzb`>%Py@L=TY+_+;3 z6&wE4(@fJHOGzr)h;`elWG;?x**E*|a#N9>4-)GwY*`>7Z%GdLO3;9SEd;yf{;vdx zm`F?8iBGStu7e78;oy|m zVW6WAZ}>)h_@D!CZRDzuEx^ry;8$Oc0lzZEj!xMQQ)9=5|0Lc`Oi*RuUj)uNQo6M zi$tHmPbTX}=I`7r;qrqTTT7V)b}!f2HeEE)3m9lef4I3d+cdQ*uPh1marRH@ z|5sx=xTcaAz6_ufPyF`&*>kRz;@mG0q;$9~V20}8TXd`M1PnaUm4UKWo?9d0ua@URMgKj|Oia$Mu z!dT~g7oo(+%=3>L^0$K~ZuZAVTw@>_-!*rX*D>0hN7{HSc@4U}-ke!knt%u5gu!a| z-6hW5ImTr#lKO`AN|ViijkYC(+-@0xeKITZbX5j}+W}L!o{%YSV37Jw{XdL$xEqab z8oEX>PYA%QRv<=1uJ|D+19Gph*CCf#n3lHDL>a=+78j)eat2cu!QQAzk77YTR3 zxAm#9JLBm>$3vGv9zo?Pk;9o$g@u5SXA;)XOB9TKW>|+{R#~k7DBoIw|BO&n%#AfZ zyR7#59gXV`)h&+ZXb_a`CI&|R1?rBJ2*f;YOl-2$X!knIC+x|-LLlTX=wqnP*dN|{ z=KM$H!FX{{;kVOIOhV*wKcw{tVC*VVO9PE0>a@m7h^JCwQvV^XPuLAnoFDE0uxn6o za(sN3tZve>3=&cIfD0&z?~nG;O^s@L=|kY|5t<^jc-I6tF)&z5(sglHf@EUCE`;Ds z(tq2x#5C+3AGG{-bM;62o^ZUOKhD*!(W3w1J?nk2r020GU!>Y#I&k~?=a(H$%!a{( z7=ObD8h0>cT^enpr6g~EeM(XAlmZ`~iuk{$I_HeLMH+sEOU`>d$}XoUo`%I58v9z{ z(iHts(NEtPSL7w}v9jbw!T*QWes~4FVQNy+ISju2K$%|KKQO73X0nw;IVLd3ox46A z?13gL|Hf$`#eE)Ix$vQl-$NVJLW=Vz>#^N}VRiKRaqF8}@;U(*fh}`SS{|O_dZkYs zBBA0b`y?b^4hA#GP<&DE=?1SV(q~D!4}X0(&@Dq849xA~0tcX=ZSrL#D;&3p{GIwY zOVz~04)8G-kay+_!d89Nxp6}=^Sau(=&~7GATLEtMt7dm;ixc3O+Ddn>UamG%`VpV zp)n+ZAl;SjA`ow?PIn)gQ0U4l_Z&}9ssxA|mW}IU&t66rJLL|`34x$sWQO@?jmlwl zX~W@AklTiGoZ`pwLe~je1cFQZiPU0eq^pH|Eq`epKSMZfc*}~5=d{K`!sJZaLA1=n zmw~+NO{rA{BxkF>pO&aqzi$<&w- z5}p6HQ1BZzL_G7Exp9FrH;wc?AzdGPdnc4*~r;R!1(g+`p>h4}8g5toN-#CKuzn7eiKEcGrY#^Rqc@+?VYu)lEU#dDkhh|b~ zgktW^byV;6`?k`s%_+&OB1C(dfJ|2;26Op1*tTrpF7Z)SJJ5k+WlwfeTk^7tJi zdd!qYzYD{QchN1L`|*pWHz|02ESIP;N-f(>X3W=~|C$xR^qs5#8`kI!tPuiZ z@@y~RQReHmKFEs6OQA$ z%tqTgFeM^7&=6QL1uxgLjIjUvJULwEn{!*gbS+Vq!zMis{^^6wd&xo7-~$wWR`5Xh zyB1p2AJS+49!F*qKQ84kj$$q9=C+Es>{E~RNnPq}d~PJpq}i3BZZi1eI61Ota7Ea$ zlkw*vy}rYs`@ff4O?D{#D2O*7KCYRkx8Vy1R{XKTzn&L4y)j%7A#pJ!fl*(@%CF`9 zcCRD%Ykib~i=#p4>6qkqw3xbY4wP~a{{FmsZ>6~74wO9a>Zpuat1oWqyDL;Uv1&$} z^kZA-j`VYDUsw6y2GgNsFrwTyo-r(CpbdO>Hr=^<5Uo>VRKGWDs<(byX!}ut?GkxU zAZT3OiA3AqCZ? zKdxtqL+!=nvEM69cM0m#s>=NBIx_cWMDuVKuWqzRre1|p3lyyWC7U~bw2Agl(3_co z=$hq&cXY-j=Q&+_u>hmBCQ2q}se!Pi&_3F-9fq>m&-|7CFbpBOJ?xlMnNRf#MAz~k z48OJt`W=Qwp2{imHz5jWT%sxx6t6@+v2~%6rr!%L)PBIacs8BXElW#q^cbW5-*~*s zDs-ULMtsiPH1u?(08N{O^ez;)EWQa$!l|e)(1OB1`;< z7zL&q^^%Z|qr)!W++6+WEe8&>?*yB5FHe!5&RM!jZEDpO@ZK|gcXaZ1KeM8g2*F>8 z4@|+&Fo_Z+B&p=TquJ}c9$d^3<$9L2u5_3iAtZy3s#l$<8M1Fl_9bf~TO@19*s?~EY)MF#tXZ-}NVaUr zK3PIp#=geyn$PFHACK?-d)$xPAFVU*dCz&DbDis)bG@FgZ>Futh|hu|u2X$aR?>eu zaUl>~PuoW~pZK}cHVPro+R2g%oOcSWPToo_c_OMm{-fZ{ld=#^uPm2Ix(U02gVpP> z^2A96>QJD}B@Q>&E)E>SB^{-OA~bTvp9lQqn6VF}tC5?GNo~+(h+w{carsVMP(v6q zU)@^gL`EmyNkk*U?zU6@#*z9_FPDhjj&P(a6^%{M2f z0BO{k_{BQ~?_>y?T}8LHg8ZNwR-;{Sc63ejYF1+9_wUj)N%<~}kB(A)SeLX2R+jM=nDDCYCfgy`G$V_Qh?!b9Em>P+rkUxINQhLcfWw+IUlB6mN zUn}7BCWG0Z26e(4`RNQ(!;N)ETL0MOcjz8iWuBRMy1u3A z*Mu{T%f4C`BkFMV=9N1GYdJM!clz!t@tb)jcMRU=D*P6SN8a)R|7|B%K{@QKy4UVB zk6cqn1`3+2x$XcbRv8#4dY1K~V&i9fLt9qfx%K6Oef@FjTWC{6h&smE$TJ-Ag4PW;)JJu)u zZpLr-du;o>zzcZFS6*OYbMT?{isMk7y>uCQAD3ue!N`ttV+w_Ea^xzBv0l(I#E*O8 zp_Ymc?6vX8r%(_YJ6+Y9JS$Uq@rRJ0py1=T4MIbg|H-i*8mSn&cGgdIU;>-CK3`Ka zEa)_nLrZ{>1+hW?E5qjG+4zqPn^@{_!|Cj&^2oD5GaYoO_`+mq{nILq4zl5Ynk5U| z^hA>TqgPCz{`omW?M0^OUoGHnJ~29)!l63E(MYNOPNKf;CJZ8hZ7ScN2C(+qETfzz zvAb~R-LEd6ZoRH#kJRdy5(~Jrf~7<%<~N0*a&zPC@{ImE_O8{(4eVV*UREU+?N3J; zk1lA&eSqCur(ce0$-x%KqbwGiY>fmK3y1%nBlE+@4S&^AFpIn0#!sG8Q)4WU#Cw`b z7|TIV8Ix7<+4&VG)6RaVZ5`{vdfSmOCYMbm(2|$+#eC@lNL<-hFNiBMJS~trkP%p4 zJAY8iCsG}xA<;?>^7rE*_D6=Vj*hmLF|tqA9x?CS*^*UJQ6Z^TwPFuqKHh}Zsdz>5 z_$N=e&MMy%s=`2B>g(i}zd4A}#g6dk?*_eAL1bR4+@%acaiA1q$W#a7tjkozeOEd$ zlX5IZl_|j-qQge!@G+ZQR($tG;YK5Z!`*HeLQKibXGrbP1sF2 zaZNbGFWn<5tOW5>qlV1}&&^Zr-IK*$GYoUvo=l1HU-X}76ud(h`o%Zo3n|ui5B0tS zDgMjl?Ra_kBip=M!h`vO28!t9$%w=fvx6MZeo-w2@0(NMk7(-)kr7LZCh!6#NobWkF|3|!M1%G^Kk?+ zxpvGGJ$YzNi4bbsPTSXSE9z24x(ydjE`8kVsV)~HMSPhV!To%Rv)mXsc4u~bBWOP< z$XUef+xt1%-wG*Il~uG-1W)&+ww}B!U0a6>SpyKQBL+0J|w|5`ZRCMwq; z9>MakIMMG`y@SnxoPg(4=ck;eo_7+R7f)Doy+KQH+AYW`W&g29aR~4c%;pDg?k7Cn z6VN)}T>ayA!7_4uhw-_%-SAM3dK=&1tiQ`SUfM-Gt;M;}8+SL)8I7a_Sby9&=^_z~ zQ90sd2 z?wg@(o6Ng&kJ&_gFuCt{Y zPvMg^v!xtvd=RoQ|Esm%*Q4a_C+oQm5B{+WFK#%@KfoDNuQ&G{D|)HB{RVT1{de+( z&$irJ(r&|ExzSaNpYKka4Y5c%%>=Nne^8pxT{)NFD&XELz8+DC`*RR+|Fdmn>Y*Tm zSG#4<3O(7E{M@Z^=7E)oDzTJ7o(ED$v|RHS5@k8W-Gq?TA%@kNqy2?J%FnYT2&JNM zV?r{QKv|ajODb`LfSqUihfN|C^+z4K&nzjgA%CXgvX7Y4X#*du6`aESs(L5*eQ7Uj ze}D|3Lw4Q6r!EcvIS(UTcsi)^-`7h2+7`j|lc5K$h^;nH4Vek0{*d>Qz4v*b+ssb_ z|9$vFpLO)n5vfuUNzVa>-EA8;vMRVWX%Twi~fg1fCf4Evy=ne5L z%7L{$4l9lLaNoULwKQt%Tj>M}oI_YQ2kr~0&qEQi=8d#{r7r_#nRvQ4RhdTGMW-3L z_Nwb1c=KvbE?H8KA9wk*< zZJfa?R)~j_bXcV7aq8L^^X=;k=o=5)Xr4)(#i6h=OblMaOGm2*=bE+Ek55K_TtT97 zZ$ik;#mH2)^8C+P3R_zrSNZp%G}N{B_3tDfekfWq8uM31qGu?I8zY0cIrc~YDl}QG zgsa8>b4?_5tx7%L(enOYeh924E;Dgr0!%!wzsk1Sd2A)D4|NqAMyYj=)8e z63s4PaR1}`XQpr8Kl1z{`R`qED}K~-f*)`1T>z0VhN^^1e3Jd z-45{9?q7HlONs8SRhyv$$ys{w2n^tYxjIO%if?hFClNC3YKUN}ynNzUWDdSKjQ#-0*3w8>%^z?;|1_lOj zieAGFrqw5pD@67@Yd4k@0H5(Pz*y{g0Y^qD>xP$FbaZr6@QKuait09ZN%=%VU>aO&W!wy6kS| z>5<5_@^yTm!_DqWbhFg+i>ylYN>p9|U`3E483I1R3s8DINgW|B4%dgxNdpN}@)2hg z@pdf>9#7~$2S}axt8__-rlS;x`Z-C@#}Elj{vF$xIqwiG8WVo0i&r1@6Mhk~D>KT} zK}G=?3PusOylfr{oz>LT#uA{e5QtV*Hp(@)CfDYFxPW zqr@5)H#KEdHVnppy!x(kpV!9zv`{RE*Ld4c9_r3LMq!cOrwSWIPm}>-2wbEzfP)5G zW`KptvgQjom}Uhq8vl3QgSPLbH_vFc-f`Q$fX2mH&@VA!5MX};uHwF3M9g39FO3v& z&CelfO)c)#EflRvw2aIQ7axGRI!=e0V7!sGYP8VK+O=7=AA`bvae=rlbhc-ffwflJ z?zWFT#UrjjxRYIah8>#y8LQr&#N!xm7Wg2!W|B!^tr-95&QFvjm-En1hXTa0NHhl> zAK^mT?o>p6VrxtAO1u;ueCxOThIr}5PbGhPm|rp`Nxjv^4S-QO3hVTVWb#;w0Q372 zasKwkEMnw3xWKTpRq9@}d5325e1;A$qmF?%j?d8Ja?olh z807Yad0*bbkTgjW&$g}u|yzM%B~`1IPL zA^u!wuf5nmks!A}Pd5+zKDo`!9Yv4+9MGNrLxx~$B&po$tUqbZx@aomaq6;bG~yoh zp$b%W-BDGP*9){Y`Y8|b2e#iC!{)rP;RH6%k7nT9uE&FRn@=?MLcvw_^QHer$4yCU zJ!6C2aqX}hmrj#RyE4ytNQF?8j4r|G%|w+CmX0n_g z6-s>TIi$1sLsU@8nGrq+jGN}vwCrlk*l@%4@IK3fH$y>QtPV%2$3J_;7S}|QYnM98 zSBZaVlET;td3hTXdkzC1VWNA;K*j@48X9O^gg{)qNUfJ=5c4HoN|4xJNe|q@&3VYa zB&u^C85dvFo%HQ_7IH;@eYv#%=`nft<5BbZ)NxYM|Gc!g@u=ymRMMGylg4{X12~iK z(Jv>|^H>;9C201`I-_Hz1_59d|B1cMK`a(-S|9qHpIso91)T(ddsDpk5^z{ zj#lrGt$1IlmLjp*{>f5UH#d}cU*X!+wd2kKppyZQC79Sj8NkTQ%(urbaAg2fwIaum zM2gV*{w$;5@{!X(m&=4(K%6o6UMaj4-7wof+EV|&q$2Z?KfDfKw9I;F^VqCKxaAp+ zL(^@F)$-(Y;H+6Pa>2<222)#5ps|tt!qgKnm2MMgm`c-|M2yk9S112F;RxNG zWv&EVEchTr)<4Umteij7V6JPt7Cb^p zAry3+=diFKZwe-EQB;gBVW+;4%0Rl1tbu(@6pl0n6NMv>`YgTDA4)QdWou9K6*m$3mkOFj1NFo=efk6m%&s0WNWg`~iy1=viRums=2moC}WWH+|6ZuzMrPOcLzMY+zyWzJKAF zn*8JlNK+xvU;!ZUPP;FJCnWk&yoQ-s8k}-Ckeg{}Wa>a_AF>kjsja2u`LkzL+#qC_ znx6Lm;6aQ7%ByBUwA9r74i$h>x`;b{_H0VmuOB}QGx%XzM*Y6wX8-}&2ntex%?V7? zdXQ{DVf|0G6VN>euKW3LO2bL7-|l@20XpLAlZpRA9OApj8N7(sKi@B9sl~RUNQs`za4j9Jg=3YK#;# zD%3C3=fyy8BN5!BoqE6sxm>S3jx6(`?P*5b+-^<|SDpEL9ps3^Rc0EP;V^Kmhm)-N z{fgeP1^T>bMSS*Qi(0FVNZz1B(01tC!oo8zj)P%%BN6itreIk1fx*Ev8DGyI#g-)d zylrlv>&9=@l$Q_O78>*UQ%LO49i=3dY6v2p5A3*!6}FVvmBY&f7K~PI*{fH#L5T~x zr8D$l6Z*RNq+%dw1H&eY>53ZuEcp$oPT<$_GpO6y<;1bE>4sgtbZH(ulZZ%Pw$~nv zkuADjXWXGOPWB~y8KJ&v4{>mK%k=@;0NbiJ& zg|UxMT#(Y+VWESKr}18>mjjs_lt?S*or0mhxVTtQz?WCH(*jb$O0bU0D`UgOK}kes zXD8gJY8pJR#fds;QPG5n=j!(bYY9!Hv>3Jr3=J=YESo?#lHCpQ1xS@}Q3!b1TF?tX zrG)|PIq{+ZQd1F9H2B*zfkqD978#0$v+5^H7i>J__n5D0$$%1y#y&puvDTSDBvm z?CfkHwm{&P?S>)r^5skDbv{tpwwzqZ`E4+w;)O=!cvXJFUExy_@6-8I%i{EXN>>u@ z7M|Z{OWtys;+SOGb{%&LumAAjLv5`e6mbKyh?6rUEG+5y^Ief3XnPO?f#X-ZO=);| zxbJdx?yFbVnhGRiI=Z?x$ByDS&gJ$K4P!tch!sczRw$6NLVaBne`P@(HdbJA=<4ib7P@^6+JVEU4$91+w0^TjtcF9=rfisoRkqfu zTAk1AM!L!Il+&JKhQsCESrpUMP0tZO5mF&|5C?P4qK zy0(2*+FKdU%xBKbme~}2j7T^T4XH744+{wy92~6XhVq;J0k;e^?%wE!B`_+1s`YNr zi-HRnhY?U~+CG0)R95!wO5y-O{n=BeM1k%cY6%10t~=Fx@rQ6;{8cCf`||YSDSCRw z&;I-kN=-*XxVN{ zD7=FRLWF0FiimJ@{)S`$#D?imYY){#ur#g!69TDXE>)4IPoHiGn4hlpe5jr`Avopc z&kdzE`uUtI&^_30S#^CZ(Q7aNk&x|>2p@jrVNGcF9q86foyoWit?y@O_(DHl#&eMo z{ido4PEFccTlYkQU=jg>p)7cqrc0PVb9c0t+18_?75^(B0TsxZ=Pf1H_ zkUt2ZTwENCyF!!7J`*Rk_)bt0H|+jO0pvd7n!o>Fu!RLd3cq*HUH)UL^o}Lx)t_rb z${4-d&TmOyl!H-oDC2O%rM1>``gds3JD|qHUKd<5IXO9CuRp`j&jOSyv<+}DmOj*& z0qRzmTDYC`!E*($LX09d$h2o1=b(HNd``geMOj%{4mA6qc%$dKX>NW9v5Ns7UrP(% zHe(O%dbX3nRN`18G~>LDBjpa4aF31gpVoTS)_fz3_m<711ofcRzkSusfV)3aGqrqF zX5i#Mc?Yg$6eR=s2M+OX8p8kor6D*7^&BVD=TzcGRtxUM8dnoZ-4fgPK1Vr9xq`Sd0;9q1mhP3 zH7Q36fUJO73~D%?@84g8H=Fp;0JCirx)Zqa3^~?=3Li8?ef|9v#!hN#U%@&Hwr5^8 z)YjCz{4fewc$qoq`yk_NZB^- zMGhG^7r|KP(;*=t2qC8tF{!JTO=O6;j#=h5I;`j7k7EdUydIGNuvI~3&KB=l_sor5 zYmEi(8Q9)ac+Vteud_~j!_qR7;3pZQuAyOC>m`h~fpnSwk#Q`G%x~CqyA3GyrlcQ( z^IZ)KOx~r2eD|v{?Ce(HuC-FTt_2l%N=izRbX&{g)G*cey4xuzkk-Aa*5FcCQF&!x zx&V8(kl_w>z+|aC+Ld1oPTR*`WoCYI=W27?2$cdH@+)`HmQo>rVtf+)dBNkiM-AAO z#7K~0<6-kg0g^q0PM#ifTkrVl%#01jtX%Nv{U@pZDWAs!u6hjA%K(jNjXn=0kAFO* zQj{5CntgWpX02WK3t>_OgMg`h*)S5VmFo`l280mEy?b(3F|6m$pQoqC3N|WFio9zf z(MD#HauAnqJ^hhyp~*)65_`kf+52F^vzmBKg~<@_Tcn!6b?n;p>*RH+>&mSDWC##+ za^i4dVc?iV^pyZdjE(vtr^QbnGZc_+)uu2-fo@?GZpP z%k96esOSfJIJNBipmK3)z@c=FjIKftg`{8NToT$^bU?h&LzF}LGKF-vlAeBK)4Q>J?w3V z!~qz&6vvN)wyZNlt|2EU6jcgKiEQ0!5E$*H0`J&7Pp>aB*;h?loA$(sH*;y#FnJ9* z;u_8H>+9>MPMrcu$54ThLYvu$A$8esfgVr5{^(?1I}k9kb$7d^SIWmt`HQqhUGz0I zyTDLu!{;wBGKhQo4Ag$Mww!h-K%Bm(5N$lb}G+nzMN{Zt%D9op%b_Voi0EhF%t z?6Wlz=i#A3Kv^lccg)|)Gbl10ZdFGT7ur&3F)^R|C_s`eZKeZjN@?w|W?`$(;gBy? z2v2qptqK|)D*6UhwY4b_(Mk#l9sxtEQ%3l!F;HXQ27AG_N2o?)*j?$$t zO>*D}7*-}Grqicq1lms*nk--BaxOgMRc9`k&^}hV{H3GA(U%_%+~*D<{3sv8gQ>;6 ztZHMEH1P?(+iYN;EG0_zoZ$QEuy(lX zEH(&R^j8AT@DuJ=o95{?UGY$^2D5Z=@x&a>*ueQspS^c*RFkan%R`EnhX)ZDW8$?3 zT_159>O{@gO?U~2YumYien29??)a`R|Hvgm;S~eN{QlOIP0%%yHo~dW-CH@6kBs>U zxHw38z)mn#zz4Ld<_c$-+sG*>8rHt@n!jIT5^>Agws2$Bvog|uKLqJpZ!yf@#{(g+ z>TURN15b&D)#4kY-sW4*8v-Q&d*`pGLd)Uq^lqDis2i+EAS%fC?Fd~cYFO`XcvV{3BvS{oJ?0L<$-}UsV%O=F-1nE#4+YFNiY=QF__1pD^oNIApWT=X-C_1$ zHkkw`pGTk!wkz{K*jc5OJ1Cx8Hz+XtwC;D~2Azy!6pPWeBDE{nUyOA3^Mk=5AKwWC zcpM-Cx!lJq*oL}kw=w5yD}FgxKX}z8@dpeW>NL1W1HloY{6+2Xy+crIf;To~aJ$f| zv`k`e-@Xl>*cvD-E-tRGPk+>~s_3{5J$5|57*Z}siC83E5s0yA21rgABC|6y<>5?C z?;T{sO%=g9Llw6_ksuhaIGi4jeQ3>i90A4BzK_{vXp0-S?P0*Gz`-6=w714wpdL~q zE_HzxL|k|-m(KrMgXuqaGxSk(m4KAmLB`QR+kyvxdLX>WfcZSMmKfI5loM62ZQlS1E{STGCsh9{>2MMsR^OmjflOaPx*OZqyrOhHxTH(>!80u zR2p`u0HW@X+514kedKUUyuPeH^7wPBM z{px>CMM+uhOolrMQM^d;Mey4VQ^c#eTh-fI^Kp`NLW(Q0UHdQu5QL$I!$G^UUWIC) z`tSe=s=StE>9ObFnUzKk07t-F`N)@*rb?k-tdt?X3O*!WR%HOeTa0GZMcEEURY`lg* zo;(Q=INNu(D{LOREENJ?2a=kjM1B+4Q3HEPnb?Yci(R1K!og z?UC`5j%CBJKW8eieGs9jq;wnsbLnH#^RIK;jK=J>*@kG=W6z1UI#!GE`jnDjCCJ8t zS9<65X^WN)$jQjQY3A-fkznt$7lTteK;k$!IH;+r?!u~XWc?isPFOL+PSN>jr3&LZ z=;{G~|5$8jOQkJCBf~5WIn-?-@DQ0Bb!oVD>2zV&_LEw$eZb@KV0Q=NjNY5A4E)A# zhW@Cls{=)&8fJAku%s!(Lr`FWF4o543rpbVmM>t+fE^PbI|c+FR5mPlkRW7O7OxlT z$1(~(P8XiY4DoUB<)^tICFVuAozU*Fzf{?uF>?b#-)Me6d$f`}_>T~+4K#8F2K0O| zud4|J!t2+s!Nm&JY)+u6fO`c9&h}U^k?6JQBO^oTHH8C20gL*B%7(xEva*NpEqI;- z>G>8UGrpjhUw9f5OhLPbIReR$_i_jn&G#el%NW-s$;i7fuO_MKYS2Ph@k_`D zBQW!X(y|^xY5*CBMUxm3ot`yjHV2C{5c9%^d$?3pLo|RXJ-eVF>e29T%B_{jjKsw6 ze$4f7{I8T%qTbd+q*8$G)C&zRDSS}eQL~KK%tc7z;hvU*vo@k~!EnZ;?@gn`pg;Ho zdlb*KoVB>ObQK;N9PodIEev_tM^=#kLQYS|dAeE1%y>La>kgPqfQ4IeP1boN16&XO z6P9KHFwo(|I3Ye>OdHUR@&tI_L1aDsx-xx@{rXmF=0A%2E;c+?} z5Qf%=LJ#jXhn|E1-y^e~N?f`elT?h2V+GhGSHiBH>9Gpdp7j&W`2EMueE+>O8#e!c j?a&gJMz9MOdPvLt%tudvmkrov2!!f&EyW`2?Z^KQ)H@>@ literal 0 HcmV?d00001 From 7ba5d628aa0f54bd322641fac459e394a2adc3d6 Mon Sep 17 00:00:00 2001 From: Fernando Alarid-Escudero Date: Fri, 13 Nov 2026 19:20:23 -0800 Subject: [PATCH 07/12] plot.exp_loss(): label every order of magnitude on the log-scale y axis With log_y = TRUE (the default), the y breaks came from number_ticks(), which spaces them evenly. On a log axis they bunched up at the top and left the lower curves without labels. Use scales::breaks_log(n_y_ticks) unless ybreaks is given. --- NEWS.md | 1 + R/exp_loss.R | 5 +++++ tests/testthat/test_elc.R | 19 +++++++++++++++++++ 3 files changed, 25 insertions(+) diff --git a/NEWS.md b/NEWS.md index dc7d086..69ce9ba 100644 --- a/NEWS.md +++ b/NEWS.md @@ -22,6 +22,7 @@ ## Bug fixes +- `plot.exp_loss()` with its default log-scale y axis (`log_y = TRUE`) now labels every order of magnitude. Its evenly spaced default breaks used to bunch up at the top of the log axis and leave the lower curves without any labels. - The `hund_strat` help page now describes the data (simulated costs and QALYs for 100 hypothetical strategies) and its columns, with an example. Typos in several help pages are fixed (#219). - `plot.psa(center = FALSE)` no longer warns "No shared levels found between `names(values)` of the manual scale and the data's fill values" when it is drawn. The fill scale is now added only when the center points that use it are. - `plot.owsa(col = "bw")` now draws each strategy with its own line type; all strategies used to share one line type and could not be told apart (#176). diff --git a/R/exp_loss.R b/R/exp_loss.R index 1bfc911..ceb63e8 100644 --- a/R/exp_loss.R +++ b/R/exp_loss.R @@ -145,6 +145,11 @@ plot.exp_loss <- function(x, # formatting if logging the y axis if (log_y) { tr <- "log10" + # evenly spaced breaks (number_ticks()) bunch up at the top of a log axis and leave its + # lower part without labels, so use breaks spread over the orders of magnitude instead + if (is.null(ybreaks)) { + ybreaks <- scales::breaks_log(n = n_y_ticks) + } } else { tr <- "identity" } diff --git a/tests/testthat/test_elc.R b/tests/testthat/test_elc.R index 0b8289e..013d06a 100644 --- a/tests/testthat/test_elc.R +++ b/tests/testthat/test_elc.R @@ -68,3 +68,22 @@ test_that("plot.exp_loss(col = 'bw') uses lsize and builds without warnings", { expect_equal(unique(b$data[[i_line]]$linewidth), 3) expect_equal(length(unique(b$data[[i_line]]$linetype)), length(unique(el$Strategy))) }) + +# with log_y = TRUE, the default y breaks were evenly spaced: they bunched up at the top of +# the log axis and left the lower orders of magnitude without labels +test_that("plot.exp_loss(log_y = TRUE) labels every order of magnitude of the y axis", { + el <- calc_exp_loss(psa_obj, wtp = seq(0, 2e5, by = 1e4)) + b <- ggplot2::ggplot_build(plot(el)) + v_breaks <- b$layout$panel_params[[1]]$y$breaks # on the log10 scale + v_breaks <- v_breaks[!is.na(v_breaks)] + v_range <- log10(range(el$Expected_Loss)) + # a labelled break in every order of magnitude the data cover + v_decades <- seq(ceiling(v_range[1]), floor(v_range[2])) + expect_true(all(v_decades %in% round(v_breaks, 10))) + # user-supplied breaks still win, and a linear axis keeps its evenly spaced breaks + b2 <- ggplot2::ggplot_build(plot(el, ybreaks = c(1e3, 1e4))) + expect_equal(10^b2$layout$panel_params[[1]]$y$breaks[!is.na(b2$layout$panel_params[[1]]$y$breaks)], + c(1e3, 1e4)) + b3 <- ggplot2::ggplot_build(plot(el, log_y = FALSE)) + expect_equal(diff(range(diff(stats::na.omit(b3$layout$panel_params[[1]]$y$breaks)))), 0) +}) From 48a22347facaa0d4efbb0d5db6fdb2812af4b777 Mon Sep 17 00:00:00 2001 From: Fernando Alarid-Escudero Date: Fri, 13 Nov 2026 19:24:05 -0800 Subject: [PATCH 08/12] README: the CEAF and expected loss curves, with the Value in Health 2019 paper - 'What it does' names CEACs with the CEAF and expected loss curves (ELCs), and cites Alarid-Escudero et al. (2019), doi:10.1016/j.jval.2019.02.008. - The example shows, with numbers computed from cancer_psa_data, that the strategy most likely to be cost-effective (Surgery at $50,000/QALY) is not the optimal one (Chemo), which is why the CEAC needs its frontier. - New 'Expected loss curves' subsection and figure: foregone benefit, optimal strategy and EVPI in one figure. A hidden check confirms the optimal strategy's expected loss equals calc_evpi() at that threshold. - The full reference is listed under Citation. --- README.Rmd | 34 +++++++++++++++++++++++- README.md | 49 +++++++++++++++++++++++++++++++++-- man/figures/README-elc-1.png | Bin 0 -> 45946 bytes 3 files changed, 80 insertions(+), 3 deletions(-) create mode 100644 man/figures/README-elc-1.png diff --git a/README.Rmd b/README.Rmd index ade58d9..02b9fa3 100644 --- a/README.Rmd +++ b/README.Rmd @@ -36,7 +36,7 @@ dampack does these analyses in R for any model. It only needs the model's output ## What it does - **Cost-effectiveness analysis.** `calculate_icers()` identifies dominated and extendedly dominated strategies and computes ICERs along the efficient frontier; `plot()` draws the frontier. -- **Probabilistic sensitivity analysis.** `make_psa_obj()` stores the PSA output of any model. Its `plot()` and `summary()` methods show the cost-effectiveness plane, `ceac()` computes cost-effectiveness acceptability curves and the acceptability frontier, and `calc_exp_loss()` computes expected loss curves. +- **Probabilistic sensitivity analysis.** `make_psa_obj()` stores the PSA output of any model. Its `plot()` and `summary()` methods show the cost-effectiveness plane, `ceac()` computes cost-effectiveness acceptability curves (CEACs) together with the acceptability frontier (CEAF), and `calc_exp_loss()` computes expected loss curves (ELCs), which show in one figure the expected foregone benefit of each strategy, the optimal strategy, and the expected value of perfect information ([Alarid-Escudero et al., 2019](https://doi.org/10.1016/j.jval.2019.02.008)). - **Value of information.** `calc_evpi()` computes the expected value of perfect information. `calc_evppi()` and `calc_evsi()` estimate the expected value of partial perfect information and of sample information from regression metamodels of the PSA, so no further model runs are needed. - **Deterministic sensitivity analysis.** One-way and two-way analyses come either from the PSA, through metamodels (`owsa()`, `twsa()`), or from running your model directly (`run_owsa_det()`, `run_twsa_det()`), with tornado diagrams and optimal-strategy plots. - **PSA inputs.** `gen_psa_samp()` draws parameter samples from common distributions, helpers such as `beta_params()` and `gamma_params()` convert means and standard deviations into distribution parameters, and `run_psa()` runs your model over the samples. @@ -82,6 +82,34 @@ df_ceac <- ceac(l_psa, wtp = v_wtp) plot(df_ceac) ``` +```{r ceac-50k, include = FALSE} +df_ceac_50k <- df_ceac[df_ceac$WTP == 50000, ] +most_likely <- df_ceac_50k$Strategy[which.max(df_ceac_50k$Proportion)] +optimal <- df_ceac_50k$Strategy[df_ceac_50k$On_Frontier] +pr_most_likely <- max(df_ceac_50k$Proportion) +pr_optimal <- df_ceac_50k$Proportion[df_ceac_50k$On_Frontier] +``` + +The strategy most likely to be cost-effective is not always the optimal one. At a willingness to pay of $50,000 per QALY, `r most_likely` has the highest probability of being cost-effective (`r round(pr_most_likely, 3)`), but `r optimal` has the highest expected net benefit (probability `r round(pr_optimal, 3)`), so `r optimal` is the strategy on the frontier. This is why a CEAC should always be shown with its frontier. + +### Expected loss curves + +Expected loss curves go one step further. For each strategy, they show the benefit expected to be lost by choosing it instead of the strategy that would turn out best, so they combine how likely a strategy is to be the wrong choice with how much is lost when it is. In a single figure, the strategy with the lowest expected loss is the optimal one, the curves show where the optimal strategy switches across thresholds, and the expected loss of the optimal strategy equals the expected value of perfect information (EVPI). Alarid-Escudero et al. ([2019](https://doi.org/10.1016/j.jval.2019.02.008)) describe these advantages of ELCs over CEACs and the CEAF, and recommend reporting them alongside CEACs, the CEAF and the EVPI. + +```{r elc, fig.alt = "Expected loss curves for chemotherapy, radiation and surgery on a log scale, with the frontier, which equals the expected value of perfect information."} +df_elc <- calc_exp_loss(l_psa, wtp = v_wtp) +plot(df_elc, n_x_ticks = 10) +``` + +```{r elc-50k, include = FALSE} +df_elc_50k <- df_elc[df_elc$WTP == 50000, ] +loss_optimal <- df_elc_50k$Expected_Loss[df_elc_50k$On_Frontier] +evpi_50k <- calc_evpi(l_psa, wtp = 50000)$EVPI +stopifnot(isTRUE(all.equal(loss_optimal, evpi_50k))) +``` + +At $50,000 per QALY, `r optimal` has the lowest expected loss, $`r format(round(loss_optimal), big.mark = ",")` per patient, which is also the EVPI at that threshold. + ## Learn more The [package website](https://darth-git.github.io/dampack/) has the full [function reference](https://darth-git.github.io/dampack/reference/) and these articles, best read in this order: @@ -98,6 +126,10 @@ The same articles are installed with the package: see `vignette(package = "dampa Run `citation("dampack")` for how to cite the package. +The expected loss curves in dampack, and why they complement CEACs and the CEAF, are described in: + +> Alarid-Escudero F, Enns EA, Kuntz KM, Michaud TL, Jalal H. "Time Traveling Is Just Too Dangerous" but Some Methods Are Worth Revisiting: The Advantages of Expected Loss Curves Over Cost-Effectiveness Acceptability Curves and Frontier. *Value in Health*. 2019;22(5):611-618. [doi:10.1016/j.jval.2019.02.008](https://doi.org/10.1016/j.jval.2019.02.008) + ## Getting help and contributing Please report bugs and suggest features in the [issue tracker](https://github.com/DARTH-git/dampack/issues). To contribute code, see the [contributing guide](https://github.com/DARTH-git/dampack/blob/main/.github/CONTRIBUTING.md). Everyone taking part in the project agrees to follow its [code of conduct](https://github.com/DARTH-git/dampack/blob/main/CODE_OF_CONDUCT.md). diff --git a/README.md b/README.md index d389daf..af7f2d3 100644 --- a/README.md +++ b/README.md @@ -43,8 +43,12 @@ scientists, health economists, and students. - **Probabilistic sensitivity analysis.** `make_psa_obj()` stores the PSA output of any model. Its `plot()` and `summary()` methods show the cost-effectiveness plane, `ceac()` computes cost-effectiveness - acceptability curves and the acceptability frontier, and - `calc_exp_loss()` computes expected loss curves. + acceptability curves (CEACs) together with the acceptability frontier + (CEAF), and `calc_exp_loss()` computes expected loss curves (ELCs), + which show in one figure the expected foregone benefit of each + strategy, the optimal strategy, and the expected value of perfect + information ([Alarid-Escudero et al., + 2019](https://doi.org/10.1016/j.jval.2019.02.008)). - **Value of information.** `calc_evpi()` computes the expected value of perfect information. `calc_evppi()` and `calc_evsi()` estimate the expected value of partial perfect information and of sample @@ -117,6 +121,37 @@ plot(df_ceac) Cost-effectiveness acceptability curves for chemotherapy, radiation and surgery, with the acceptability frontier. +The strategy most likely to be cost-effective is not always the optimal +one. At a willingness to pay of \$50,000 per QALY, Surgery has the +highest probability of being cost-effective (0.369), but Chemo has the +highest expected net benefit (probability 0.356), so Chemo is the +strategy on the frontier. This is why a CEAC should always be shown with +its frontier. + +### Expected loss curves + +Expected loss curves go one step further. For each strategy, they show +the benefit expected to be lost by choosing it instead of the strategy +that would turn out best, so they combine how likely a strategy is to be +the wrong choice with how much is lost when it is. In a single figure, +the strategy with the lowest expected loss is the optimal one, the +curves show where the optimal strategy switches across thresholds, and +the expected loss of the optimal strategy equals the expected value of +perfect information (EVPI). Alarid-Escudero et +al. ([2019](https://doi.org/10.1016/j.jval.2019.02.008)) describe these +advantages of ELCs over CEACs and the CEAF, and recommend reporting them +alongside CEACs, the CEAF and the EVPI. + +``` r +df_elc <- calc_exp_loss(l_psa, wtp = v_wtp) +plot(df_elc, n_x_ticks = 10) +``` + +Expected loss curves for chemotherapy, radiation and surgery on a log scale, with the frontier, which equals the expected value of perfect information. + +At \$50,000 per QALY, Chemo has the lowest expected loss, \$14,954 per +patient, which is also the EVPI at that threshold. + ## Learn more The [package website](https://darth-git.github.io/dampack/) has the full @@ -142,6 +177,16 @@ The same articles are installed with the package: see Run `citation("dampack")` for how to cite the package. +The expected loss curves in dampack, and why they complement CEACs and +the CEAF, are described in: + +> Alarid-Escudero F, Enns EA, Kuntz KM, Michaud TL, Jalal H. “Time +> Traveling Is Just Too Dangerous” but Some Methods Are Worth +> Revisiting: The Advantages of Expected Loss Curves Over +> Cost-Effectiveness Acceptability Curves and Frontier. *Value in +> Health*. 2019;22(5):611-618. +> [doi:10.1016/j.jval.2019.02.008](https://doi.org/10.1016/j.jval.2019.02.008) + ## Getting help and contributing Please report bugs and suggest features in the [issue diff --git a/man/figures/README-elc-1.png b/man/figures/README-elc-1.png new file mode 100644 index 0000000000000000000000000000000000000000..6a8a99eff1fdd5e26e6c7d7fc7b83a9fffef0bfb GIT binary patch literal 45946 zcma&O1yCGc^ffp*1b26L2<{%7;O_439)bl3?i$>JTW}2o55e8t-I;BE|L@z~+S;vJ zG8v|zXS(~n`|dmUoOAi6tR#*6f#3rO1VWaTkyHhNAb)^B5Vi1-z)v2HMU#PVu%`0T zk|6N=zaO2YNx)AKoMm*}Kp>R<_x}(>PG#mG5E)2TQcT@D`y|K9A8YmroR@LhzL>bi zh(I2XMs`DT7zNXbBgtZjSP)tEqa_libME}}Z^jQ%+P&3ekWplvp`!5lROFra_RZ%f zCu7;`q?l>ci%e|fVLKg0kCbkAf>ya1iy4y>^pg@~p`fq5#%y>B>mesLaBO%;uwQ%i z{c3y6-d|3`^+Lw|_ghkEL_y#;#%uv-Oc(zhbR+hE9)y112(>Eh>3OGR6hJ+eYJ!(Y zC9C@9oIG|LBwEWnwp_3p(HP)O|nn?y9LI%B1&x8o-TloY}9$Hd|65J`@r~mLL`1z-*W%h&Q(uHvnT@Q>%)vNP5vMglh-TkTLz<*=bq0uL zEGcwo@}yzt1zjeba1(d_9FSh@wmo-NTwEMTFSA6ECMhp3uU7i&u=zi0PDmtl@$vC- zVPRp{R{h}CS=*G+-M-xLPlX9ovQ4_oJ#&ZdACU@`>7W;uPCc{6_s-8*dvN$&D)7VE z@sM<@G$|6?_X8nBYvxCHU5i!edr}MN5~+$+%R$Q%D~e2`z?)3GlKMBAQir!2AE6hW}r4_5bUj*1Ng`H&<6V zd3mSyQ|`c|g0d#Nc}q)68d_Rbz2>p``FX!q#5io|mT({X+zQw-kas@M8>o(TA`G;hzX)R+F=k?t245O9Yg zoNo{Mvvz9s%bHvP^|dd>?2*m)jne5P8@eA}GP&$X{4aW=qCRNJnwgpT`1ttx@{6G% zARv&FlmGtx+i;|l@Pi{V5S&klkqwq$+lK&TX`T4#-o^2WHviXir~5{||M?`f&q7TO z%gs{JIIVIF3=FEOs_-EjJV{AO@^W$=$$mUoD+EA{-NV-3Z)8M@gR{4U1P^a*YkQ@7wi<7t15qUeez`{K z2L;Pe5qva{T(rk=R)#0fED5`PX<(8^gI1h|w8sw@gGac_u+TASen|j=cYd;@GQ zm^mnb7$pLv7j6P=v<1ZCOTvftEVONB&gArTZE=e3qU?{;*ozZaOjbUr4KG-BWtVdw z=tF`bw2lQy4o;x-{l0j?k?*`cI6HW%pRFou+bJq4cIa9+>4wJFIx&lvB{f4L3A7rm z>DNyF<(m|^c0Y(&CK^J(^nR1^+c_rp?iS4H7zpuNRZG*psP=ooFgoR#4!l9UPSp^N z3O4iK6K4HJe99!U`}2<}Lw&{*$1SffFVT(V zryGaORV!!v49^v4l0Frp8%LtqIy!37VY#-xTs1A5H0kcTw#rbM_tj2W139Zo6S3BV zKL@Fw4(kI%vuFFNWdlO{z~66b<7_d##>(aD7~hHn?h%7l73GRmgPEmNsAPc`5Y0KQ zS=y;h%cp&3G!gwK+*w=%;~J}qvGqR(nJ3AY2Awj4-n=3-WD}Suq=bzd1sxj&J(HQ3 z5MLq>dwGet#xWl39M!r5I3MU;OM= zhy&!Kx*siw3Zy?q*Eg^~NPYS=n9gqI>gvjbBd@NWoOU%z@WJkV5&6=_bd4Bhi(wNt z;1atf=ZM=!G<>G&z07n$+;J(~t)rAW3O;kkaLf7E1o;_l;JBu{%W^!EXWgHJnfdfP z&)SA`K0kv|hxg^bVtF++HGv9ejsLEAvM-ZB*Ykx`!zWcCXItQvU7Sh`>9ma3ftCN$ z`?)OGp5!J|iy}dy`p4>$s@1XHmT6RECCh?CHxSjGJ(F2HlL2bQszUcGZl{$#OhJ1Y zR{CysLYluS=76oL1kOb9wiqnCdGg6PV=v?@bN8!=jWLr(YFKvi&=ZBn4~oE_f`J|H z&z?nvG&P&kzhjpW|HXgly|3v%ifz?xyF-4nU2Jw(PWmjoIBod8BkqevR?m`}W=;*c-^nZzB(N{db}zTFF1M+^xNOWl)5Xa~`|?$CT;sF9X0K6SFlzoh zu)zhc>l1vW#X$u}Gh)OkOvDL{w6^lK}IZ`%8= zJLUmhaH=j_(z#JvwSlIQ+^ez{G8~WLbCFmxL%9c$OBH&E^WV-J!}D{ooBQoX(9W?Q zsh#~G7c4;}jZM;oOVVt_tB{g=!F!&A_RWHw61Z08(&oB7*d2E?j8@A&agU|b)pL=O z^$-{mV8ZzaGOzW=F~TWoAZ1$%-eeSVB7Pfy z77Jy5g{TzR^M&~D{oCnr0S4g5!t8!X(08NVgP_0!b{@!ma`H|pvQG(|8X3veM4VD5 zxWl-%G)zTDQ77ZGZJ|@4>M|3eNF~xe@0jU{g-AZdYCfngS<4F~iZqUQyiytYN|F4v zgdyo4wyc`K2RKHA7z=;#+BUOl*7pfb?}DAyHxcR{@;pP;J2Ra7x6&1&Oe*eA>ZdZg!i0gL!<8E<11N(*#aNlgCEbP8sG%kti80 zjDY$gu%=hHn7!Pxi(_DYbFH=JzZ`!7@9^0#^bq2I-VbAnU%}g-*)R<}R6OR)h{c%@ z{kE*6wy0D~jx6teMVO-3g6`w2YQKBtv5ZSJ9V2fZ40e?H~%1I}0j=e!uYukwTo4!Hx+JsBh<|7e` z=X&1ndnUZOhBK44yg;f80OO*K`obG|gzfd&^Tkrai>EtS`S0;ob2Ni`>Y1*v`mXEY z*I2s+fE_>UIv26wbnFo0q~TZ{u{LZXfH~i-R5zV`TJ`kLBJvWM&4{}=6HMsAQN!G{ zfP+UxhbZn^Q0JRwStD*}Gse0)ml5y(-BDhOTth(!60=R?Rs5m3R( zXeS(_^?mj}PQ)m1)bNP%vz!+ek-1*QvO1+ChRut&&y~H;6_bmypwYpGsIW)5 z_ZB6zSfnJy$mXx5POm-3EdUZ@25V_Tc&SNrdJM7R5)k}^_%hPkLIWG|u;4JUq$}_6 zJp(+=v&_2wAy}P3j%dfPX~OKwIn)j17g||WhLj3*@lPj2C=rqr=$7&-SYl`&GZccW zew&)$eOd_Qi)~o6@3Xhb^yDRyD^A2})K3R++4F8*H*ge78X{YZZWT~5P+=g{THu9K z$(lFqaFP(_(?xYBz&xyEw12bEX>(wb4#XPNJxoG7ZM>H8=fH_0s&>45A z0o?MCF`H0GL;gZy)2#arqn7AHG+HQtI2dj}X&im`CJ+QI*iRhENmvwLx@&St@fqwA zjW%g{fk7IQp{cmHweVGP9+zI9|HW^h7gAigF-r8m_V-`5fdfDMw#l zT8j%aE)siwmF>qxa*C~<=+^XY8dp@tXrv5|e2~y6&Tk!_b1|Uc9iZ*dm})ZFBxqe( zhSD!i92_{DvEAjm$9L-te&2%h1$gHG7~^VNNo`lzP}evfQKbGwo0e2LHTgMI@Wm!i z>%QdEajZk@cjUd$fejrVnvq@8(xj>?b`4FuN(STVXhJ&#dHHdiBlPZl125YFU}}2b;n+ zEsrt36mf(HQsQR420tb@i?XDRrUWz}XN;NBe}1aoT4!1z67H~i`s>ssk~pqU$;t-LKPRi zc-O>S8y(+1D%#}LP5ho6z+7(2Cy(WQgRm7<9`Qg6u7Ge1vKZ^@il(p=-PWoKxByTphrl7W7+t)0RVE@}1 zE>TR0<~X{u;)3jk*J2Nv-8z0RZS?r9Yar)LPmSCj6KAEot+z)wXu=Kko_FKK5)o2s zMKX@)GB6@lpWSoEQ=#Kje~Y>L9yMoasXjN89xM?VY{LO->M`U_`#$?IaA>0o*LN8o zyXwpxc27RVdB|jZ-sn897+P5!@A4odiU@qTi^<;WbsahPIxD;UV4C|4Xe`4e*hbSNuLaYRxZ;!C{F^lublL zRM54wQG=(z^Y#7Hy%t>WXC3^5x#lwcLm(Xka@p0hIJ%0D5n0P6(MH77X3X}Vi^ee@ zA6gz|ijZ5oW?o-fp8|q#cx@cVejT`A7NCbL5vWy`r8VOrqJ zQodor!6;>A{O#;AV}pEW9d1~&zf*I8vxZbGLB`0lw?<4hYGT61KYhnmkEXn$Jo1G$ zH=1aVC3jQvCZ0WmUNij3e znNU&;#>Sl)uCwE98UwlS3H^^O*Xu(^L=h{f{!ji<-F=z#IyFMfY2Z*BeAc**?!S)P zp=-|7OynF3zQWDKpUeRK3okb_eKj==0G{2u92y|QLo3ePyvkM`?=O2BcG%`IfoMt( z1c|Hbq#j=Emzb?Pb4z5QVnYqWiu@8I;Ql48-@d0k7Mo}qAS>%8tKw!*wP;3Mu1-XJ zJj&>+-l#;=|5;pt6gsGiJ-rLb&X$dM)t>*QweTo==tI#M?ZLTSjh6=xL0RI~-unZ% zj;|4K2wv(8hZVCNk`#hcDAPm#jiDIYabxb|s`~+l-9)UsNpXRC`6}U$ zLQGdotRxC8ixEvi6gEpdP5O>sZ|Dy)^|$jU6{WeGhL32FWoH!LA}n75US{}KJJN2O zEV{_2Fv?rB4>}{NL~;Z?>Z3^Iz`Ofy-KT#FPjan>VPGEzkT0SMH%J9~K9|}NE9bG% zbBojrZaSmL#t!7KvcE#VNDUcRe?}{ii&evJz3NRW>P5(2wVrsG@IK(EpqlQ)q1UP$ zTB*u$@AmJ}){^4^fkMbqy3&$wG-F_o!~?f4V?Nwy@$$t=ILru9sT1R#lj81^^7Kqf zGuy34YAXy59N~rc9>pGtxIcQm4v_|}boeybT?6$hQ2v8NRc4l&>^L|$0QM)1S#R(7 zczN)zSy=2(mSqCi=`>L$wbuznby=D%Udkc45IElB==kfz;mwhv@=RIqg>07eh`P{j zsjx!c zweR1*@9yp%9v(uAQBhHahK6F{;?h!4dDr}yJ`~h7Xy!sHIS)qD$j)DXYm;BqLzm?& zJfBf`iuh>5fYy0G?;sXvHuvr0cO^a()oi6wW1VUskE{Px*IrOpncEsU+){FmUV0>$I>Cm`H+Efh`J z@zD<>wt?a*t=JA`&9*5@vOwVMm)YahEWYers;MQl z=MQ~g8ue|wT3f0%Xy=WE3I@1GBO{}Te)0`bG;yZU4-FQ0lWqYu(5V*a;zGL|iqoM# zWVHcC>@H)yoy{%P!X^`ko4-Fu7W%v;sdTBGd0O9*ueR6q!ELUhTU!n-7z9E$ZT%yn z`_cVOgAV}zAl@9Mx1Jgth_qi_bN<#r2%{9B4YV!1=AVtfJ400I@zH?bKF0-ud;at0 z(^2a=9i6rqTIjSa;zcWm7bbwiS?i_`LrFTbpcmT4*yp(b_6VT4*8`uK!QhSFuU2E} zE;k2Liw#zkp(}ZP4&2LW7g@wQK28p|N3)!YP@<};|Hd*n#w{lks#p*er~L)`e$X9V zrSOZ^PF~FSOX1;+T0n+-=W__FVNpN0d+ZU^F)e3IQ=ia}Z9lq`Pv|(u^Vd&hv=FZ3 zHWvlO`1e6=J$#$_;$bh5@7@Fb7=w*N2;)3U3$St>{`$Wr)B5VmD*mbKnldqSb8>qx zq)9SFMX5>;*%q2LLVStpri!oA4h;(Gm|j1lXQ6r>(_$D+RO|VI{K#Jv>8mX>fzLgpItZjO#nRE`xVY6^X|TzlXWg->fb+#q~T;a za{Qdf{Vs()lfwy%I^8Oyet_YGXeJt3JoQYG`-5bgSpsm2H8nN0EhY_%KQ$|1@UG1G z8B*j}?z0V+1-P>SVE#>#0-_oC1{49kgE>mHKsn^!ybKfptCN$Ho$(EXiV@wMX9m*q z*!oP=@ERlbr;WuG$r)9JC-g}$dA8vuF`o(ny+I~)Oys_6eUD~2>?J$|QSo<}Vq+8rF=-S)!RhJg$b{U#MkaHG1Z@6M-Di?80z_t<=9-?H;$_mXd5GVzH}oBl+M*bfa> zla5P`Hb9iQws!pZv9VBR7AJIv2ZDx$Wx$GxiHTu18{FC1(Pm05E2H!3k7e*?>ae#1 z=*xU>9gbh;H#8a};vgV9ZF*Ci`@<1w) zCvS77ff2W!uQph2cBrhZl$VwDxtWrd5Ep-l`>A;hNDoT6f*Gl)&Hj(BU8uH~TfgPg zSO9kXv@B#uoOSo?WUX`Ao*P(ar;sR;fG}kK(S+FjKT+yu+L&n_8&cu)A@qVPkrlw$ z-||g!I(*D{d3g!-V8~P`xwE6Jfeux6b87(9s@gNh%2|fA(J&8Yg72Raslsl<_}`1n z2Y4l|Sg!q+OpklA5y+m&^mcDr42rtRKg%A|Bx+O&Hzf8-7v+|CgT`#uO-q$W1*ODe zgf(XknKsa6IrF#($kRwfSn0fUY&3HNfMw4m*QyT5T zOJv#5T6#tU=wK(C*ALrqH3sdP!Q>_Pf18{xpvA@j4*Kq7@l-^$TzyjLUO%hPD%U4G zZ)I5pCN9iyggxC*etNP}cb2%PILXu@k~^Tz^Qhm{KszE;NYX9I<2;ult4U)OfM5GeCd^^|M<#qSuJWp0CnY#0p zpr`fvD75>66IP;#kx^aO$@%@9N~eqDa1E9;g-n7h;07sFBj!uY*gB*BFZwl|C6qz9 zdi^qdcc%G{6y24yn~%}&Tbmk|TKsytUze~*O|Rvt^KQO6P40aQdH*y8e?>4rfX4%$ z0O+C6|AIR`0?fo|ITH@;`8V1WTMwn{kq8S^$ld}BJkbB zdWFA6K-+O%3AH2Ygm3dY_6?+)Py+NGTPGG!EhuP(@XF6=zQ*#D|4XX=sG2{k7&PWJ zh)JOsU)jz$El7UY<#HVEJ36_ZS6;;h15@eLPMKlyYvuaZXaD%vb{STkT$aKy zI0z`sdnH0%p>#`sE2a~vX)NMl9I+;B6#oNO4h%rKO;eMFbCW*nc5gGcorP2?7a`o` z+!mdMOUp>v{lZQ#D$+h%&pC2r7ZCP7hnM0*54oyO1$PE4aM?_RiPPhA($UfL@wzDg z5TEkQ6cR(z%^|83fSn$FN1*y>h>45-uToFZ%ORi6 z^0`@mr1uw1PkW`G@ec+YnTw5`jdbno@!cm%0fFbsAyO|?kukOy87fy)_XhLfL}?PA zBHE7r8_+w(R{D~iEaW93p;J732nxNNMAh>I$8i}ccu_^m+;tx20i^Xui`>ZudG8l7@i`weUj#rg67sz%yXQZiuNi_|4|<2S@I#CG zhL7vRL>92I4yHL&vaXIcOS@2mPvfs?leJmC3}WrGa(Nvv z{cr<_!&AYBVI9wB?T9!`AW*2XynJ+A94aYpS!*jFJw3vH(u8&P;OM zCLqoGHHmaqCdU+9g>~p~4skq!4V3Mj_iKem1sv%Z%H7lvBn@ z;tj%^nMnLi2I5$mb=3E%f)8yV1XKruqlrQS{$=o8+Ah1z>V{}QPHTNVpUIEpI3g=f zW#C$J$IF=+XIp$etI<(?P5P>)Dgj3WdsSz)zml!3sY{AYTs8%dfFFHGVd#5Ue0x0a zSL3!d@a?n6>x?`MO1^++V6s(NA)Zxgm#y^gN1+xWdMbV9J$^VK4&7$SXe}Pxa^B|W z(MErzC8_Eh{5x+aXGao0sGO|-fS3sa<-a_Hf}+E!>6~oT6pIs4WD#U#k-vYw!2QO8 zPZK}-n>@ebj$?;iD0zu9^ z_iSNZKt1H<9{h7VC$IZG!>cZ!^fhRJx!SYc(4MZV$qEeoVpDB>jX#-GX)gLuRm}YBROOiy~P%0h&y|$G{w=W^KkNeP+-nE?K>K~nnS0qn`r6zE!i{VQ+K;|71-NLQr7b0;c1jbGLaT`lnRR=PVom;?tt7ok2U58oqDS z8NPORe+m8?pM1b2+gJ|-xEcCr_Slf7GhP9ML*^OVTPv=1ss-Ms^hu`CdZy)s<%xRC zJ%`G}0?KL(*%|oTs>)&nObAf{SBN2(A{q0KE#INNba-PW;0^BF*^#u=e~9>%&BTHZ3vC|97n}XdX#gP$V#`z{bVKp?An(7N$NM`ElVOBw{_i&z z=&yN;^XZI%Lez`797hjwKGc3(CuL#XOuzBA{#^YPxa`h#($BnSZm0|x2K$)-7yqAHs3lpF6yXu{8dQ=ptt~YmZNs;-A z`2xPH@ZCAB#=q*j>b+h#^-Q%8;5Q(Okc-OvY@v}SpeC=9#;|2HydXuPTA8AlmoX!J z-G=+%f5DJ)RB_eYj^_XEKX-|FeeMVx>)2`K2;5{46e9yeUw?nTf$)Q37B{nQ&bYeWCpRLt14V5U`9J8mt2>Oa`- zv=!Y>;>F2Nh2xPlw9Ib5gdCb{cJSoq(exr0%{H}DrD_JRh4r6{M(6LJ-kv93!uj3& z#d}U#wMx&@M@<_RqoShpvxtUNuy)@HB`aq?NlW|Le4&y}q>7IDUSn~_#Lle*9}V9B zqeD<0f;(Ybx!KTB==0R}lG=tC&R^NkG+Oc776ih{=Z>?4jFep8DEeYzw^<3q@iJ}~ z1-TG7Uda6bZ9U;DTD@MC)K%ZdOLh>OQNTUMp6`49nZ}v0{Cob=ohq^>5hPVbzH8f^ zl{-7GA2`1MmyalBO7!uu$rJ7%%VAMx7I-)$)+AA})6rXd>plP*35Lya((4)-B-lrVM2a~H^b8^%0$~yUhc8iDV;K)M!cF;yzVs-#NNj>Z} z^dF7>{w!F}9HOwR1xLWO9 zw-TMl)`NezP23s|MeC%vJ1*08;*H;}Agh=@p^;Y8Ke#y_vqZQnJ!Wq9RY( zjm5)O(JWwPwQHYYrkQY()P49nqg-5}RTOnjo1B`80nP4rXCIi9H?>9ZO3%c^1lV5~ z8SR`RL415TpsqiY3pRAk|4;&}iThnX!Fkk{{Q1<9Bl;tCB*|X__Awmk3i;93?~Ub; zL0;E;F1+QQuA~hTSXEBIzHBsdUCrciCm z!t3+~+HGXxg$ZS!|LX;#)+AZlsMp!bF!(da1Wq)c_WqDl*UdlqO0AkbuoZ9fpM~_w z8ljq+;o14UBfIM?JJ7~!b|=Mpvl>50R5BhE{{zw2d56ewMnZ!CQg`3&B+=-E;)AB` z8_(kz<9X5L{XKZT`)zk=$EvXmux?=nvjqSgZ~7aucvGHWj|xt(ylY8C_yoeQYc8Fg zzbs1*233zeYZ$2VtBqGCHk*|H`Xt1;cpI&0fzj=7-ZZT?j4IGh^~hr&#KGY<9e_Xo z?hN0VaQj@^Dh;8Wyae7RJiAV-WMF0nT^MW*p9j_9;2q_xU{}>7d>c5P#+`}$yaK*# z#!rxqD|Y{-;CM7suJw-$cAFTura)u^qsTIOEo5}bfrj=J*+YXmfrzie!_~uMKo!N) z_$Wg~;nW)NXTe2<*YbhuOF@}Qe|YP?mLC-l4-Y;4uzc9(NzzBAbGqOB5I=g9gZfjw zctaJWbU@Da`TFJa&E>4v->^;Tk!wvKVqCte3JbQ2ROUmXZGkkA^4KXq+F))vScB*E zoP_)zYM%+jL*t*d#F=p34I0m4CTxNLihIQlRgmW43T&@e{H0Up^uVR%75byTfSRN5 z>ZbhY&gm2+icwUPXF8_+^K2d|3n;r6{R0Bd+m2bf1`gdRLiOcOfNDccUw`HA-@l(W z@MD=dx&Jb4?<@dd!@VFs#x43}vT1Y>r%W034v}~Zmw$oPz1GM&o5r~huAL*_|E))5 zGk+6P9ByYO4VQB+Hd-l0`Pp#pwEe7Ybe9b^SNa<|5{|b*P-u1g-;3X7%^k04$r#Y| zYB1~IDXx_S00Q#sZkqn0Xms$>5ditp|_eIh!V3kTD*>k|r~tUUjc2?23eenn9!Rf*|hxpOY wzbI!MUtg zy0!=)k&MYF!m}p+LkcrWkf6sw#NIyxrXx1nECnh{-q;#ZP$+#PoMX^a_rBq9C@I`g zh){(9Da+xkmeK7NMj-u=JJfp^!0WAbg-?~KHP)aHeew$>`>rMq0xR#j$VYEQP z>t~GWbCpm7xSso@%S}%yR%4v*z(Jn*O6*LRYgr`?*9D%cME{0}6m_QIR97>c3t0)b zp;lEns;y-=D=aYezr1*wRn^lQbpyr-hLKNvovvE zt)+*&9pqox_w&9!%{hf-QN%tnNINRXyWU7UQceQ2$r?C!W^T#_0jd?e4KE%}V%Drw zn)Z~-k>1+g&hy?vn77PQZo*S&`2ytGd9m!v4J*ZUa8gtAV0=P2GD;U%;&+d1j??py zliQ7==UENPTK&IhaSZ%oLLqJj5ASsS*zEeyY~|dlx|be*@g-2Vm1?R-~AIa{(14Hdoo9ObUTnf z%9YyR4)b8pERqnZq~!~#5TJP&1a!Sc51qh7JSvJ)+;>}~oM&T){%UyLffv&$@7fSI z%d!DDB3=je0btVeO}yi8b^jT2h;*D+UXM3{to8o{x6rPsSkCb*je{;))`crD_|O}3 z)+d1Bq1D1C^6ZK>l#6nTRCqZEf5x$YIfV|^-QufKF*V()GaEu75nM^E+UV!?f=udfGOt*L2gs3d|9q~Hf>1qCD{rT?ccuW>o zzubZKD>ssdIYPj1U#P4_Y4t4e&Va;52{&3%xm)bnv$gBrNO! zH>70wtJ3;?qAJtt!%FYx2~tnzld8J%$?t8K^^T;x&OJiAjHC50yls_|Rhn=xFb2)U zfD`Ue_e&|=Q6BhZU*Nn069-3AUHxW8?Q;P5_5KBELCes`0-)pohZ-FNgYVN}DUajI zf@Sc?{}Mk+j7)^uD>p}rCkE%ZlqeLP|Au(Y7M_PpAb-t@nW-z>#g`tf!fA284Jvj> zk&`O#@2_oZ*#v?+EG&#l)-yS6ZqwNxIMXQ%2}DI50fRaP{)a7*DB4mv-ljrU?=US=0tXilJ@V8!{t_%LIdi^xkw_XSEFljirI z-7U%pIh#ykK7Lm@U#oVwCk9~PypO9|9>BN0o%$?ZoygD6&+S{|hxEsXB^4)wQ73~V z(~26%kH6mwM!l7poCs)BC50=Nlx8B^`w2e=`Z+fzA45Y!dwV9ZhzQ~O?R0m3{-!Po zIqdZFf$kSlLGKHIv3~@Qj*KYKV6K;IsHh|)CT<-S$}1_c(9^r5na5#6LPGv0;V6*8 zxo9|z7ua0|oAJ`_mRWy?WQ-;v@CWlh2G@lUTI!rBD4lLX6{ldQNv0~wAZ7FNc`B=t zfve<+N&47g#Pbiv#BSh-n68!Xa!-r-x<1a1G~%oEV-hp<*L)0m>ES6QVC+=zN{bp$ zirsC$x1hKbf|BmyziVb60jEG&7m-YKXjT&QyB}n5Sam2wA|fFvC@6H`NQd{{d}=dG z%o#S<{pCz9e((j3WTP~HZhk%@EUc9Q8XB6Di>o!u`)5!6Pie;Us`skq&>d3L*m&b$ zst`H6nZ2h5I(C}~1#UeSq!&VV;@Qsj?w8N=aRcnqW;5m=4Vp`&wNtTAH~=dPca&|G z>h$v25n{HYMWkCauRNMCPj^{*eJF-(8RK=RQP8gIvTbog!_nuQvN~-RL-)E+1$+Wd zTLo>=9~hvR!8vyCP&DyBo6h>JlPunPy`}Pijfu7Hw`ptQ#xKwIfT398GnZAqSUZjn z+!iCr2L}gt%^fNIo7U`&g04HyKTG7`hLZJk&(>cp(pA;WcyzO|wtiJn^En_D7twvf zT!4>1A$&RRt?b(a|G=+)6!2-bKJfhJ_lf2N{3N5Ul>X-W}BKr|k9S=sKCh zfbmTxu_F<4A!Bslo(S*zCli-Qfgm1)zz6s>dSE3GL)J`LH#@HC8X0w){*D3!wSfUi z0EbQIk6S)t1U66tgNjVV8z1!HqEdgrlssZgOwSH zPmv(AKq-3>h%S!QA1PT-$#39p11qhwDlelLX}PDSwI;6E#3xYsiHbrx2FZ+%_+<{lK07@ejKRfO~g7pbpI%@>LT7ZuL}b&$b_!rAr>)YRsEQN))ql1MYOrJ%G-Q zZoB#FyIuch56^3X{d-a;CNqOgBQ(9s`93KU$Vi}0Rqn1;?~cUk{wwLAtaNu+e&HFZ zidl#Z3T_!1H`5L$gP$tSx9nJ%KZ620m`)jCzwoF5rm$_^I_;HlFH8^{l0;%rtrJZN z>R-jt9rzG<1ccVzR6PJerKG3h;o<#}{<3FdYa10A>0EPO7ou8zW4+tiw4du23%0sH z1B~7}3!M0wB|tmJy|oUp*ta+G8GcW^3Jsxz3lVFOA~?d#Qf@4G+j zzQeoRAz*H?W0$%=mS@$i69Xg6@k*q6*K63}k;ftCZ>ZGoSZRz#n*lQm3(zvn#Y)q5AW>?=kaz79vCiXNx&k za-MzJLN35Dd1~&F&|k3a3o7S3Qb-XE2M9dptApX2ZW50t^8S5=LoF$7DFJUj_v_1L zq@}@u0G2^cN!;qJWI~~STLA34<5)l^2iV*E9yWYN0)hdx2H5{q%GKAr&5jGO=CS7f zmz-3Q^Pm4*z9827=MQDN|4BS~0|1fQawhCIr4pDF{|6xidZ7H%>o|CBzor&F^e?O4 ze-;kV7e@FTZG~7oe0CF6B*Tfbt-UNK!wzqD&;kANw~DRMAGN%SAW&mtW5?~B9(+C~ z@-FT1XafQ87$H2xSsSr*yj;jkC)*!H2?!clg^*vb<6HcePhvUzdtd-4(+a1&uXcuG zJ3b&I|4w*e$BT2X>WL!}dg=zho)^g`&DeT)c-+=>zqL88@`K=!kksk0@bU4@24jYY zhwHSNR8&;X!ihC5w7nPS*aA@H=Th6gqq;QD)yb~&F+!d_)U?*gU)V0;KoW>G->R=6AxB3?`fkLao=xXQ9hd$6ecBUb?5~H{ zz)eU?NdZamzzF=L<>24|sIzu8or4iS$l36pPj&ohS}~=H%jJ}C5xpkKA{2O^2gKRGiy*Rp&`I@-N6RgX+0rPfB*NOZ zEiE~`<=1L!Y=x*Sy?@5XUpKy@>L*R`#%{yq>I>CTBv=EISARJ2^*@=`R-f5A2hZl) zBI2?x=O0gyq*!wS$&w0*!*83EUkj!o3mM3+FXMu8&@*oS#(x7qT4*C>vCkLb%+35N z8@ne+FV+I+VTssLh18r>QnrG?bv0tC%V~OcYiE;O}lBm9*p;3`1#9!lj ztNl7o^IG$<+8+GAc6@OAv6X~(G)k}OhHP&Ph4LK>fLqqYO}VfBVEt~>abom+|nUb5)A+T80cbE*DJ|#pd*Nj z|I_|5Pg=t?Co-s}^G?^+mH|}DWY?sdB1bJE(GG~D<)yh>$DrWF!#}okB8FO-I!39h z>Wli$kf1BQBXx!4v14`u@bB_M++II~wjZBMnC-ibfX)?=G!Xi}H`dZ!2Hg1quwT(d z&oLgb88vSj5!2V;r85Y-^lkauGS;!4r9xR2-FFW+2neg5@33G3H;yc`fF;g;&Sq-1~WC85MV7q2lHnghHZCS&ZPl_ z%QDr}W-oWao+)W;($~|$A9U2S3T9s?iWV{89en?NDRH8K81cMMogu{ZwnUFj7!S2(khO!gsQLg=Kt6{DC#|G4 z@yqQ^&%W-U_|aYYRHn;++uuO}2|-K^U1NYmVQhBzQOzx*Jfqn#5ua7Pm>83YgP+$^ z!P!29<)Hg_jd_16T~l5tmQ8UKnCH@oCP}ui7ZD!NR*?a&f;Fs(b3WiI`1Zy#P>moP z?P-ArFnWq1q$bRbs5rNq7tHOYFIEa{KBi&X26WgLKE__&DISFs43NLb=~a> zch2a6*h|^gc+Rf#mm_v(9zbg#o@UF-8P$#6P;Yo`eq9$?fiLH{Aw4M7qAK7h#@t%cs@wzCgSh09{>1a1a%2$8mG1 zL{*G@ntJ+H98H{>AkpalwLPBLbh2|LB0e+Yo*1nzg*(J{v!2MiXT{H-!8=Kxf6hf; z2BV9gMXP8q%y#%(S0C1yKB2U>wmNk^?g5=*iZs);CjBJd8(+c@aj$xj0LlaSDl8}f z;4^gNitM#8Dlo84y`GMr$J-twEn<`eYLU9DGn30h-rl}ghD01LvSeodyyddI|9c7A z*rv=9ikS|%A%0Bhf0oI4%Zlhmf6o(Cb+}T8a&#|{*C9s(m zzxVHAkzx>3pX?_eut9ek*w)dk2uPwBlJ!B}RV)vj*n6L)}60nhU!r62yQq_(D37Hb6LF>e0_=#pinaMK3VG zVz$+7|HIXlblp)m15KH%f>n~BR_{9%=G+PPq zO9o!6yA^_#V|eB2N9+Pyw=i$Jv4_}zlD_kcsvtwqmzVsK$ZO8h?d-YX;7_lr)*pfl z1tbvk7hV)tD(JD+;br-t$;8~;+=PV9&IusKOF!qWmHZw2OOpv13=`z9EAIQ}V*GQo zJTiQc%Tob+FFU~@bu`RFd?Tvx!;iGrEojqpyJ&sDYrcymHJ-_JECPO<1ZW;eC@3I< z@p~MpgS?+lTMW1i#FBHAN~aI?e>ragu&|f*qT}^$+3RG19OGL-K|w63NZ`xEWk*NH z{hD9gWXY*xm-pq?XW>U^5Fl2?5%IZ>GPeN38lH#$v}9LPrP~w*K4@7F6mZ>%oIc!v zDxCn@fQ!=%fmq1{+=Y@9>{;U(85tlppuTeCS2TbH7Urbc@LSX2K5o1oo15y{w!o-_ zxL0YCkZYO!dM-zTk(&QJI z?qP~E1-0A!%svS&DO$o2yBK7UvUM5mqrt~FX7q2Q@)=u<1UDB$%YzXp>F60Vf!zR_ znmfL&wxSS^49afZ8~G+#0AxeMXQ;s&Gx(4NAf@z$A;u94cwUe5Km~VqgFElno`Gda zLrs0U+I|n*m2Ri+Euc)MGU;qNrl+Lbj zZHbAdY>g!qaxyZJk&$H;75o8DH)Se?-X3`(L2uYJRcGUX=zPOA2$eSz^hQU90+MMZ zaEXy2CUQR)h$<>So8?8Z$SHf;ZbAG+nYRQDA0k7D&F(fKnp1tN!9^xYpJoJyxLKrn z2*lYVxQaV#$`%C?X^+P!@NpN{$iMDey?;W`iL&BU!jdV6Vv`g3iuAlW4+nf55iPDf zMx)qg)T}!L!)3Osvlt*;0PIw)7;XZzduvv9Ci-9C#TcKppLb0aN(Ws6i49nP09bc( zyM8_g9DL8`LBQw&t)8gsHZxtv*4KOE0O;KUTj32Jx&sRn@QnFCY(@Y!-YTB3_Q!mQ zh>aThGLUmWGS9>FGaz#SBamw}jE{!K1`HMGuQlhS;Pn`=;2%E0hj;_6qp7S$Z`YF| zaRgjPzzT&7e)phMPAc#K{V3CJWHT?LqvLu;do2LRC!Q9#Zdx4R@4pfxKnV@_aDtW- zjEt;2u~A4c1fNU9$n$e?Ck$}OM4j@MuBTeR68(T!`Y}JWu4&`AsF8pO0(%{yrnKcN z2IMFlCesswnpYkU7K)NTwfmfAy!t3|sZqcQ37UsQ_H-Vn8;Q* zUYD&!tGw}v37}?rZ=PRGkQMLy9TQ6m^uv#&Faj%7oq?FkR$ZEgi;D|%0o+h~0onDd zhGj#B(Pl=ndz}@du<+XI-GsooewY6vZJz%Xkgxzpo-9QIFeio^Q+G5S2ps@v0F)Ii zPV2z9ck^&@g@uQ|r>>GLN5BB5NqW%UbHfyAK>*-K!3%`$a{M5p3n)TSDhRsSB)j~2uQ_sxg!6$$^NS^a?Vs$~&qklQsi$4L>z<-w2i z;Y|e0{;i0IFeQI*3oIRm4qYI{+EygGr9+5rmE+G!I1sQ%Y((%B;#9P>?oS#fA$#`o zf|L@AG%It21D*hHjH#*Vm}SE=nJ=Ma!JKTp)kL;Vm%k6d34Z$Y2_T_tYz~7Vpae57 zE8V+#RcHg6$2TSD-vOGRFIAkT1))1K{gVA?>Jt=8>$y_IOoA;hp98d*9 zU=Slwh@?PyyTkz;c4@P+em~KwuR3`wMv7`fReGnYJ1RjvJO@Ht^3npojoV18G~0cy zEn~MWC$fjv(c$xfwFm>kz{E7DS^88Mk&Z05KbZ#?3~a-LgM*MgfNO7QeslcM>;GWu zt;4G7+HUbhsDLO)NC~1+3IdW6f+8)cbV^Ehw}|inB1#Gf2uOEHY#NjXDUt5(ZaDXT z-uL^(Ifu(X++OVM+H1{u$GFEBb3KHLH&3IqZDInz#<#luWk&5$5fLOvMt1f{t@lNL zCeRwPPfmprBAHcL)e3wP@iyzQ2(;e4+qmSSve@6>KRrDSZ+WOzKnn}1xjBGN>f_7@ zaRZ>WcMD%snGIyYe=t8cHHBNJ78{=8pMepICo9zDaUr3F}_u{l^^ZAx||YE z8(6?DD^^iYilv+q+5Pp24Gh=9R*yE~#;g(robv#VPk4)UrINWY-j*i!;kd5$YlpuS zA3cQ(G~6mzR`e zVrI6NH*Vo{Ue`hjhAbXL!e}=)SWc36Mtu?l22KEpqZlEAhtUDqyssDiqvJC59+p-K zu4-Vt*3CDs)b}n>pYu?_EmNc1Zc$xU^u|I^V-M(+FwX_ z7kl~MTd$^WvT7Ydq<}>YBT5%8bj63p_`J`LLWPbO>#y7fW&u7fZjxzbULGqQ-S?0q z$ocMome&Bh1BDnBK*i{a<7j7LXDvq?#Umti9`8J!nV;_smB7sOG>z-*4d;+P6E4m- zU8ggl!ZhBeTSCP;{l#^ZoCkxzJ*f43jDJDk%QUg`BHzL(l2Db(S~M5Qk;A-9A-y8?K0 zZGqq-{YW(O=ko{3n3IJO_g~$+Oq0htZgWl;T#l{1S-I<;ioP{$VX?8t5R6?>g@yH}0_- z>)1TL1@-Okfa8Vw6FLDP_bNL1DHs{iKO?+xnySSBgx~Oq7pECola2$o8G-_4UAgVd#= z69n4R?RK_!xdlMmpeDSIJom@vb6WkP;Ypn@+pvo>KQl4lS%p(5hkWv<7t8s3e?Jrx zi8VerRkX3u#ckFOvr~FNEE`KS!datpO`hhs=8v&&t;nuL8+_FI4R>g%`tA{>`xD%{ zK~B#gg{X-Wa>=^mI`rjU@u}a%f@%!dI9DJW!h@NcpZE9oANRgE%T~$7!?%!)|3G+G z%C0{{@wCH7=-r#QwXgupuT4TKvlOJ8?)F;Qhw0mO$-eXd?&pxHD&V~C z(EZ7{Yy{WWM)xA-%a@in_1pjS=v|zwcXKRDk9gD3|K#yWeV)!*ZSd?{LwaZW*47r# z$YJAL454<)nc6NeCKF?-hqTo`QHc^_)St*G?x^&#YzX%DOET}pNf>#SS_eeMq^cO7 zZ8Pmicb?R=YKkfv1>=x^pSjHlIzxl8@$YHws*2ir*@6?pAk@@yfMxo8%sh$VrXb66T=~vtJ2)mWHC{rojvfYsmU)Oz+gmM zi93qJI4?%auCU}nNqOR6wE05vk!b(trts$Qj+v%I9w$nkB4q<%Qb6RG3I8U?TsfGu zfZDTj&34xfh|u#Yr56X@7YCE|SRvx~?3EH+T@DV<{ALKIi%aQ|y2S*pgGfA42LE}= zUjjHk4c-Yj?O7Z%EA3&@5lNvxrWiQ=*#kUg{TV-W?}{vnsq)l^)A89(H>npJYu9>d z(-8sIeiOO*P-Av|oraj${`~al{Om+RRP^sXrGEATF0qC5zVCx6$_6E6@^aynC6~J2N}6I>Cwn7kA44+wZAYNB|^>+W+; z7DthSI?;jtiM|D0jtJpH|A#ssAk2l@> z<-eyolPAodk-CZisd?KG z99sVJlRfKO!(wxNw#Px@M6Qfvy2B>Ry%|ygRtEQj!deY4S7RXMkC<+VwA4JP38ndW zz8&%D(o&XH0BC%FK|Bh8ypK49@RJB zl%NPnHSO+u!DnHZecTu00m;DixcAEond8#3ZuN6jR-%Jl-1UjaC85?k;xG9GYpN1I zeWDF-ZOgxQ$8R|>h`1^`deHlPClLVWfcACdo6-YD`GgMH^wiX|-9C9Jmr|J8isEE; zeP8L0HO7`i(ZBDU7;#i?q4~G!PY{B^s@5k@1D(QWNEShO6e;#TU*SO0At&Sqb`fsXZUowN%I35jN)G}yfpw=YlNW#OHjo7)@RbqXXj z`}RO?6P;xF>}DaOFRZSP=Iho@PQI?t%Ih()8ZEb-?}&lwmyIpIJ(|tHNg@gdlRS#< zErt)~+X>R4!~QGUh;*}yW#6BZfh0XSxzay^=kuL$+9)UG>;Yw#J}3yn8Ra#iW|z2| z*PE?_gb-xV{vTmR4>h>Ghg;1!r6NR@nX(umS=o5-$ z{80WnIZ|fr4dXUwgguVtqTyB9sVd58ZN5A4;&>&0#=x$H=mmvz9u2K}5C(@Fzo0mO zf~TNJMRuIW-+!W>-LWb~{e*HoYru?=k27>jSi=({OoX8Sl0>idZD|x~ zykn@I`m6cyXUmelw1`R6CFIKI*~TAdB43}F@hbTd7FA1!mpr8<2e=#t&L9Akmc~1f z1y;nXTevjJK1LjaGLJ@4N_3lnc3FY;x-j_9j3tfV&%wDs0@BC55>IhD74Q; zeC*kYz4Q(FVxr+UsYlPSv0dldND)Z1p%ui3986ihZzgVtWMI@|hnjt#ZA-8paniLp zZIyg%Lr*6aB!-Rkdnxr-=kpbhPh9>+yb~22M$UOn`s`A)$b88fSWC&J2v`kDIb)yQMGb~c*;hueNt`u3JsaLzF0?1S50S}v?dkBdj=@S2V3m1u!@2+WKufLa& z7hh9pzu;M2aF2JSs$!(&k>#!U?+78`;_9MW?wenK^0ZH&QdX*#0o7S|qHwqDis`HV zcTH!Rw#^&O3r*Ts$UWKUp)gUVul<4^8tpV!3&XgcA^nepe6;%RZ4? zTwTAAG5>8=3I(+{9v)u#?KvweD=JqbE&c&*!3bg}{o=$H%AI{lLxDH8nk5 zK8AxH@@In|E`+CWNji0P^>7ARN~p^r$HNY<$NrTE)6uAe9o%{vG=(MHSTQ{nNX|y^bsby6L3{|mw`H^mqI!K5uGx z_VcAj-JXdmoMEwA$@x+ey!`PW(@e(=I90q=qn{}%>FFhDvF7K*=uumeA=^7U^hAO0 z-eHASx>YC8NfoMDc=p(t*>Ux7<@PNdQp;X`mjqOphLlWxmkqNhBPjL_IVft`ZhQ5p zEQTKLC`AgB;d7M=P3gyoSXX2PVJAjMlaVMb5%Ysq8otW|WOrrI8V26L@Q;U;@0`xY z!Lbbr_6sl#K$A|G0FiEf}l}`}l1G>hq5Gt%X=!3>R542Ny zS~lLjkx(Hzg}D?wa_fnjYWsyQcq$SS60qsL0aW^ddwqae{GR9QjkT)_(o|`4nZqi_T@8lWQwIKq?+(nd4xZV|_QbrM;?kq}sVNME7xsZ`v4wN7@hE|f-zVza4cxYifNwa-r?6LfnDurDI1qCzBp)fGtNu@2-7=ylabX3db*p?jG+}uV)Sm&(BBgCaW-s;0yqKw+Uv|YYVb#h9|&UOMmY0BWrPyNEVRI$|*+&ie1 zrC0m+zvu|nzm;2`(0!s}*uTGNw(mHY#M=7?g^{JeRC9j3VmtHe>}VkojH@y-OaP^T zIFVZrsRkbkGlNGkrh(@Q|JE79**qHPXmt4+B?%3UkXb(y8Cf)tS^)$7GyUXhTw&Pc z+1XjTY@E~a?h+bT2N2Z@x;!8lS`w>2*|O@N_vR#Wg!_C~-U19*Ep#X5_pfJl6|yLh zU?BvYK^NmBaYq@u8fBdJwdro}nd4`9tYV7#aPl?*6oNs?4-0*9#fUkDOPFt817qLw zWWRJ&^s}Nf-P-=g*=%G}ew);Frii-9X@Bz~x1A}}bmG2f*!B=Lg5$f`pBZi>6_aOn zLuBtyL7HU%x20w{w+G-rnm;UDp*SQaB7%yn_;o$zI^W^;*RNl%-+Ih$*zyTV12l&f zQk077`pitR+ukzlMUy+fS|qM`B+9f^K_o=hp`cVnRaNNNIqV7~`;2*aZ7Y z;=R6(VLUb9VEM?mJIZ*-9KEqI=Gbnc-z4(3{^_ICCv{I>oQ;31yoUK_QP3u8$A_`| zN3Krw@G{0dqQG+O?zuhBOx`X*nPl?H?@WVL_|9)Xj(pXk(;0RA;@8yt_=X56pifY1 z0|ABk)%w(w9?kRKpFijRlwGrt;%2piCWeg@uKqO#o5@db`5pGE%U8X*ly$ zdR3KRY}(l9C^%96$=4P3T9c&8xTPH^NsAr)qWQj91x2jL?TWUKiUXOm->bAQMmZ?& zBfk|>uj$6e`;8UXZvNJ}j2NZguri3!P?@4hKZ5WD^>te$u%&huX5lh`{q#s7E!Wv# zKmC`jYjubC2#Wj2x>m|^jFPR^X?5ZxFeH+7M}Pp&b{`1BS?z;C&}Vb5ypd#OWgQiT z+FiHgM*j1!PyVLgp5WqeNYri@S|Z$1*~y#T`5P*|`<{yx6?f@a%00DO!sHe5^ycW8 z{j%;;B^$N}w6xyZymc9uEDiM!m{=xW>bwro7;pI}+9-nVw3j5hW?r*uHd|=>2_!cHeCMte#aw$#glD z+-?tO{$tk^fn=lb2cXOQD0p&yg>C#^I+48;qIm2^J**vM&9Ev zBGL7FZ{(#7etua(9@oc%bZY9O#XAM!?M8k+yYaha#yfhSS`Kvyf#7gu{b)of z?1iWhL2f0-?MMqPZ+c;QbJiO^nYC_7bv9&(pIm~FJps-vt8$LsD-j22+q}1FNL|x! z1{b^gI-+=>$6{Y_w0DIC4})jx^M0Qo^igkRr+MmuR zPw#$incL>}WVj`bI-0oZZs$PH;D3H+|8I>uZ_S2Ka7FXpZj)ap7ENF!RfUfPzE{uRxNx*nJFMEr z>hvt&qP>h&YOguEpj+e3TVbL4`Q|}~qm9F_=?`HrK=zA!^H!FrMKHo?MPWI2$CMX? zwvheB^qRFgWIfi^Z4Jr5*L^cLw?Y4EaXInbLZcLm=CEbc?=(wcp)$@~y3k8bV{7&j zy;cvGc&bz@SgDtb11Nq{4eGXb_Lw+|A%x?O3jf57fAR1yCC3q^{JT`sx&$MA0H*NU zf6b2yM>y>$4Cj`FRdJZ_*w~PSiPnHUhOAxEW6`-smFmqap^1FYaPAnsIJ_n3JywM& zj~`dcvVV9nCiuKq#y@SdGiAm9%8!%eo$T&m=J2&b7A?WVO6vp5yysNw2?xJqYbE!) zUJCRfq4MQYNhgP8*XJMK7I0c!P5J(Dq^9*x)Y_!>oiHDfcHyJF0&lQp( zj1nXFSh*Q|7K9||R+_^XEhbx^1}$pk;Z0v8Ib0%KHMZo|BH4U8i=|?BNI?l8><@T` zC&l$dz*nxz%~#lQxgQWpz%*`ule{VejQa zas&&H*Mj59-`C=LmMXmDmm!w=g#2z4Fgg2jWK{$r9pmnVhc|;Co+Ptz+7vydbyIsSD=dPjIgPgR)8Q@0tp%$E z;biq+mp>QOQQi(sPl(bem_F&7UzIy94tSS$AYAI<)*l7M>Z{-vBaK%~>@a90l20EN zkRhM2r&w$czd!er`G@Q;Kn8CM?Jj#yHC(!V`?eL~=j+t`*q1JeNl1*)75H~H*4w&< zDl%TkO|p90Gx;A){PaY=kx22E^DeDUU4%(IrQ+4i{N`JzVD5WqKYb-#<)>I7!`-^n z9>Zs$e_s)(`(HwW4D~(2Ufn95T>7yoS2r)15EXLK!&VC_=z@^!4d*d`X3?>PtE;CnKE!*k-}KK7;q$|kckiALlsfQhN3&|@>DDGhMsh=)^(tL%!1KG40ewBH zynJ?kzL)DEAr85JYg?P0lhY4b!)x&`SuB$+9Lm^Hr?t{D{uv*dGCqhRgkiosc|!3| zP6aAXoVPcfqXhSs`V|?fTNTcn+*k_CQuNl=?A_&mmOj`T{O4LitDGt~#i)1i_u*n$ zq9D4|&K1mC+9=mc{_dBYYvUW5%7 zfb9hb2Q$jX8iv;;TroeEA-8ua_2(#JSz-C?{kP_#dV2JQdZub|-=2eMI9E_ozG1p< ztK0>t9M;o7V&Dj^=xx&lsIPZE&3l5;gbHbku}h>_nc;pu&bOF{My%Pf|* zKl11#RKrpqwt8tbI9?)nU_-oAnI6z0ld3uxPG$7VCDr8t^}VXbhU!><_sedr@#Wrg zw+}z|L?d-y8vSN_^$vSOL0DkaIW$(34?>NsJ5gyx=Pv1VZUT1*KNGz_xPtnQLG7B; zd`eAEf(KOjT&SY#iUSC4pM#AJD1-v8Q&*Y|Ja{i+3q>$s8+}}7oxBq^?P>}2&Ho#; zLDXt*H`FWbdjV5nz<8h;f{x7@^J!n2EE%ho3Hm<0m9ruf?+osgAtp35ClH>R>Aw5( zH}4LO_?vX`H>o;Z(w%+6Qi*}@hu}}#y6C$jQwN_oR|5A9jrn|wh+O8bK67$&gPhM# zOFAq@?~a6?D9e17(Xp9o8Mlmbzh<4|z}go!amX%YJxIppN~BsmWO40UsQf`YVerrS zsd3mw$9B#hm*sf{d&jE3z2GM(3A{v|!dz2Z9&JByFtrpHx&7xx(Im~lC>%2YgtS56 z10ir^g*Y35SuK?ulTpL3Y@JAXjvma5Ee{{&ArIl2TX&}?8%Hy`{Z!Z^#eT@?qKXDK_ zX>3K>ueaDyYF!+mry`(V{~<8JI)5f8gj74ZxqT1k4h`OQujXDQ(yImH+UH5P+eq+8 zLUI^fAgqMZQTQjB4!5_jK#K|%l+rL~2YMfXUt!_laGnLk9Q2Iy`0x;^1LB3;*T;Gl zVb24ucg72pK}|50d!#|deOS4nS z{{AGo(^ybfw7wp+70!?Io^;mKU_O2{{$eKW3Bwt~Cxaa+KZUg&w+m|*e)Glk1ZMst zDuZ4ugb*{StflUKMSFK@Sv@;(-Im8#0KC8pveLh z1s;I*FXrbWsp;F>+nKq!BjeJT9ta3%W-1Bo1m{F=y175uQ$(GUG+M}%OBTvN;d#3< zR7vH$Tw}TV^dU6#cn+f$I|>IqGs7*0$zepL(#lc+ocQsj3}M_$7I1cL}ZH+zH{Zgc{IIG^6%M>Y{$5)`h#KOvIOfH_%qS8dk~PN~YJ=sk? zj`kr$_l?TklFHpJ8d|C0;eG$%5B5T=28v+&+%|Ax&S9to%tAK^;m`;`JK&zAP!4A= z02-pHvb(jVUhU=xnDT94vA{?Xj-DCRSK5D%TcdN&Cqut!;8bueDx}af^U`QedbHiB z-+qG5A2wEs;`RA;CTgl*2EF;16)$o{g9$}=b{2c?wGXM}YQ$LIMw9UXIS}AI+tf4N znFUdj)ik=Bot&Do-CtGDGX&Gcz`#JBe};a+{E4342rMetKRi5E6C1GQW-u5Oww>R1 z7}LE@&P7)e^lG0`jltZK&b%MPr>LD=>|}6-OAf?$PKvg^*8Lfu=O$lg==7{)YLCq5 z-!4lBL!y-CQd62`ozCU=&E7rq373$!DX5df+ba*G(Qi|>^qr& zo@l!eT+}}o5xZN@ncEp*&*nVBI9hVz7l~;k_GIv`AEBGhYv25IB4Rt9Q`gjz&!@S) zg~M}rTku_;H)f3_tm8(tYLK;(b0OarA8QSn8e0B><~-xPP!SE&9YdV+zfNt&+Ox99 zteW@X<43EZqB*d2Y$xEnRO(Etm_lfP_QOI0ko9$H zw`mgL1&ZgXT;4*@Mjg1$k+}R<^MD$y}9_Xy}F8Yjo-!phfV7J zc`N;z)+VlNN||5jRqnkZ7iS^?Tc5ZWi>D{=U+5ZNqeF7;`Li6^hl;0Up3?wZDrY{- zsDR}}=A;5YL_B4{ESF8S#|b@h39+*#-Y9?x-O_{d-Itw}+(WFjm7^k$FF_Ay0saui^uQtAzmuFX46VZ@jeEasTZXZR7a8b%0u_Z2lGmSkES9}6dge*{W zlwuPYA)qJ&AQ%G#Z9<3W5C-2KHpm~FY?gM7v{R1a$cGB*aC>+RTL{;~vsALh?tWH! zkn8m|L<=`<)bk*WXl6~bqRe^@(?^M6*F4#5I)B@IdUh>^1m}TC-&3?c6!UG}t?}xJ zhA)dB{6_eq(pU8hzd34le~kAOtR|Un;8|;eXwPZ9zVc2}jjB4=WJ6V^2P-NKbS@ zkvSEjArYep5zkaIn@Pb%N;Goe*UC@{ad%$n8miQUfyL#0LL9XHm#=xFBM9JPtp_?g zAM)`H34+z?%F2qlBgU000eI%h)jz=b$d(uc(afMkx3S6a`g;BPb-i3c>xniFN+yKx zlV5T0bDe}gqpub6Qy`8J6Wd`tr6W4zR=mXJhzm51rU`YH8<%8QouSdh~8h1DJaXhV{|Y8NRG#C2@SQvuXmSwy+2{* zOJ$4JS|UiHnpS)%d^af1!dnliX9q$n@LB7Ff9iWpG{CBV5V?KA&lw{_C$)Wb#E?Ug zgg(60rGH?_0X$qMoRTi1Vw0Jwb=V@MT^TjT9|6PYvt&~~xMVU)g&l0LkL&eubr!MI zR>G-ZCS1dFCcS2*`h{O6Hn6;eB`lRV4Y~Zsd+H-NAle4?bCfJrz^C6M3tM0r0w_oKCLN|82o-^ zcPTN^9AS4@{0;l#H558rvM?>Q`~B_N9F;@uWX@@_R;gKJcJ>OepWz%XC_)KRUmneb zma?CJfmI>HfO6To+g=3q#;fcBFE1|@vWy&bCT2=@rF1+yNYq$Buuk@KqAN{?s}Zvuvy**(6;ORH52ftx znpIwXhZT+Pb)Z%yR?dmz{L~`jg@|~Y9UtZ1+}VBM?y{8}KEcX=w5YdLoCjyoeFX^% zK>wsT*!qPoe{$4;`rB{91)5*1E7xbET3cEsyCh+3fA8*HFPN1=mw@#YIE*&vpG{CX zpd42&BQR(^JKnQ*aQG_Z$v^n9xu^}uf*)y~wB_!!e7*-{!)KOj0hPvc_~VU@_wL=B zfhYsHX>Wgjy)RupR}1S7jy`lrU=xA1yoTnp=4@Yc-R?JVtp643$i}tZ%qS=*dT;rZ ziS7(b>l91t8(RSjj9F%Nq=gPtwUqL%-N1O&%_C1#CAFS~e@`p|xt z6~NHSXq3I{r$8@9Bd;HmPRW@%-Y#Dm_o9yl3Ag^{b+o0`J^t2+h6E4KcF+8!X|xd> zS_UK@m{rE|SZXLJ0F!-Mqv9qB$(H~TXsnZ^QCC5h(*k%J@PpvIL}7RVl6jyRQE(dD z9As5GED?c5jhwt3D&G6A(tse{TGSRu%=p7%h~sq~f}maaV4`8>c+h`mrPH@b-}d$E zZ!a_PuU(sq9}BofrIIGf8f|ZRz~Twky7Osqq^t^hdZcqdL<#+(%Tl7t3|ySOquxV# zx}ADK86vNd%+J><9%5=4ENLj0IusB6UDr70`JL2m9+qnW|3{?+th37~SNl=xb& zG{O_pe@2Q+gUBTcRYP_LewwJ57`Tx<)Tt^DiYtb-3G79lqXWF&iD-!jXQeY9cx&ij%0$QQffDTUq_=kS40 z6hHwq@4iv3ii&TFEQIGjy!`D~4GxUJ`Kun){ro$ym61P2-Bziia)BX6zf-G0Vy`tiN-mmXh{R{&>dgWQER3>Nh1tl{OP<@k? zgqkD;HQU{nESsu25AG@}ZY!|s#-V6Q9DRh1`1D7de))1GA`k=cpV1q3uFiv_o7VX^kfH^RaJ zonu3h1y-dHfeP^{@by6^x73$z#4$9e>wN~9n@zW-YMtnGr@8ONP`WqmGeVxjo`v;w zKZ(vmjpynFvuZ{EB7xPa}Z9?_J~R=rI}AzPNpUENL> zWzcp4k0>xG$PFfpxW47^8qLFS*#c682iR06C*Ors_cNF*I{z;;)ru+NA-DKIerY3B zt(va`sL2<8$E>)x5m1D|ZfV@ZGc$$$2Xx$&?M1M?z0w}90Rqg%!lIO^H2Ur*=G&*K zFC&-it`FRXqzS&{)~={pY$Pc-@uVrz?OjjNkY9NCt7z80q;Tn7IGAp`_=`Q9K1Qw| zE$@Gp1kr`i(QNpjj=L-^AwC7i?DkyyDE$E}a;W3DEr)l(I#3PeWEm3#Ct-ps?d0$r zS~~>Y7?2>*L=X-`Nehcr=KQK6ARuhV?d!d^2&Ek`FMB9gdiGaR-RkaV(pO;nV1M-ume^fU8gQWoi zhMe>5XV0FQl@1@(@N_U6;e zs*R6?B5ajG#DcPY^UalQA7C?NvZDrh(?DtT_*7CBgY~#S zMw@~szkH!7tSRhOUPEhG;CKIZKrg<6+@hui!GJ^@wttR7^E#*TI0h1E{laxc_1_hm zbo^7=*V)0L%KgB~!C@C#j;iWvSVm^+Q+f4zkZz&gw4SMi`V*{Ija>U7ni~z_{b(<$ zaIc+WAU%7ta?O9+vEs(3!=EyJTnHSs>{Wde=yUGESU#-QDqRjSMWimDwg0uMecAX2@p}h4BNiwHDhcA+H?C&eJdf}jPfVuA^#y2T zPis652SoOQ_6lvWG2g;@2Wl+WuU)e&ABulYh+S)a5!+9Xlg!|V@y$cqbDF}y3=VeF zJ|&5lxIvox)kuAOWUl6^RM_m;!$-Aib`$rxhqURWhR-ea0iquo_6*MMW<|vwIPqOu zf9N0kQ!9M=b!{0rw}T2%RgcLnh&G_Qz(8Aw>K7hH>w1}fyG;h?Kx4jDWBFmL?XL&t z@v07*aB<+js0{mp_PVfYiw-9r~|$~Jo6LqbFd&H&x*Kn7R#`XgWyEu$+5{7 zIzNWhynJt^yZ%LLFI0TNqRIKS=fguou7GO(2*g+ZHR_@>;LUKyslb{Y{i}@3xHSa@ z3aKw`2L*vABb{pYZDw4USBCFCea{GnPehBd=S10mKJ`8`E}EVLN@xIZzMHxMM-%W; zDgmb)%e&7>A3Wa0D!@k`@3~3PI{)`N(C>ia{bi;D#2oy_4^+Hzw1gdr!V^57leQP9 zbCUbK5#$abA3@V+y&EEMB84=w8!!huIyzvy!N|aHd~{U3^@|GX1i17_Rh!i}Q5=X* z(aPhb;Qaq=a&L9->*KAh_c=Ji+%twRV|YwI8(dipuI)yCuu z92~t2g-MKRZ3;Hu9lq)O`^le_Rk|i&w}1ZfMN*MOOz~g&CPdI7I`-X&X?v>71_^}L zeEb#V-=E`Cv;doL33%b}SC=h5d-Tref@2I2NSSh#k!3D zjmZq+7u@fQ^Rp`iG}u0X0Rvffn68EFQhO1!8C4ETVMe$}f%ro>5|MBkJKwUi`zznf z`npvaXclm|v4Io_tUKuAp}H|*g(GLFczaOJ)-s!=FLG6Y4Df*#r=rTi=W&$L+f;R{ zCM#dA#Slp8VAHoan4kCrOAJEY21HXh!6x$T#e>K{PV6|qqot!05=w|u$sVW>uC=Nn zzJ#n3t8@+9m_mx)$G?t7^c0nS2bUbAQ;TGP-GAN5_{~)oO*>or{Vuk}uOtW|)7V!x z8Q*KsIe*aR;@_54c3~RowpKa_ROoWp-RKLLn==Zp({Xrzk=J^>Nac~;+lqhFYtbrFihyw!!tIs$>A!xii-<4 z1e&yuJbwHbX6@DR)Juz|Ku+}r{(XK34VfFtYx{zZ|Nn=C#>dAqGv_2%P1)hB99FpPWobW$b<+P@DRL4x=UQogwEysqyQIu(0^c6k zfG}7`x5^$Kr%($)za`&Je4LZRVr^}`larloZe<0`?zyYXntugRPy)oJVfZ9xfL5jS z7z)v41NA>9T-)xxJ|mT_hDDX%drZKM9co_KF?->?RB>%A44);lcEGYel__XqWp3RT zCUiU9X>S^U%}oPGnUaWYjj%SdpX#TX#_Exeqk{Fw*F)wDwWExVEXm9YvMb8e()%dz zFKb$n(cG3jT~F!Zk|p%(F_~2yG~@tmjhipFoWOrAl;NXJetuW9ylZ!Th@6tG??j=& zl}2`G3VfRr<*XzRp&eErF&eUeD2(@h8nEm6w8n;pHhWer){FOv-3e+-BYhb$0#ET|0!*bQRE69v-^P#4|MxP$RV0m*}l{N4MSU z8yiy~-VJ>C?*C|~y-3T^Yh=!*B1%GxwP*W;5_)&%zpDnP%OZPT7s@Kt{??B+#+1Xq zm>;&53~FkHe*u@eHD~VrNt3LrAlw;;c@gQ=45cg>c$UHgH@@>^0tPk)%^&By)-Zi! z7&zqtRW6hn+%3HeOg)*za!3kM2R#6nfMS{Y)~z-&l^+GJPC#rNZGsgqEiHY>T>!@( zMN87nvrZ=l0&0dM0wC&}DNyL|F7+*(^Nu)l;vhC>-RXwmV22Xa>_p+;gYF`HC4eEV zrlqwyJ8O9KD)Sh)`q?J06)*kE*@jaJ4;}q-pu&H7|G;2Z9k7Sbbm}!Lzn0l!WJts! zt(i2a|JBiLd#%4a$V+3&g6x{96eM~Vo+g?k4%zVkOA_x$$dFITNVrM=?#i8SEBMM0nJD+`XeQru8ru-M+|J^DnORamln8b>oVjg7 zL+YnoC5G$({P^aChY0&C_n9YC?o2va1Oi&CthM;{_ zkE(WBi`B^8-`UAVgH3DOm1ggc+pJ!zBMJ`WL7~4O+`We!)EGbza?`Bn0h&Ag7+6)s zf8g9W&&|nafO3FD2W(DMrrKb1&N5>GaGI!S6TpET4_}k}s~7731Y*-e-RSz)kL~9> zAj)%aa7^0X3qxPuQ3^Dv5W>LvwhJ07kfozpT6V%UX>##yXUC2R?Y|%yquf`}+#f>g z04N9;eLi5xvCui2<2r6aXBCnx8|J#j#=e!-3btm|N7;);*eSS5y!8jIKN;^!lLp(jtOp_hOT1xhwv za(tsd7J=r|w1hZ*B6>z;w(S!eu!P`(&krVgo0F`^D(4m#|K_N3w$A|Rjm(9ryp$lFic9H4bPT?QwF%mFhmM@Rm6 zh#&I}O153`t;qt`pQZZ7$lK8Dd0F*k$D8 zKEgs(jYomj&`M8F@4wm_#q5FY?Cb$E*pf#BM&`LwkZ^4! z#4e_wX-}@DJvXTK2QIen8^CrveV~#ASrQgy(S;Wnt{??4)-%hKWsP2#n^TVhLlvYB z#`NU=_nDZ0{4XOfpKj^~C!9bYF;?f{!u`!_A zYB0fZ083jCK_oFPUz~4TfEMCS#XN_ejdor2(R>`#O&OcoQSEc2US+_%cLxs2=q^0E zDljkI&Sw@77#SKm0z{6b9sZLOa5fnj<@=h|HA%&7_WfmMxyo)-pW~a2@#<>_M|-d~ z#;)RnKx;pr7?YVA_nzPH;0|aDJy84)9^fHxycVQX9N)))Y{L`?Qj<*BET}S^HpV-( z8?FE341K!uTr%tymkHsZOS=dC->PBIkjs5xzw`L98ZDuyxHznv$TZWP*!CtH0E9q6 z0DB}~vn;PDDln>S01FGNq5*J^wc*ki*#jK-=mRA`-a48Mdw5WZjrk06sA`^8t2^Of z&?`w36Q&J`?`HZSul_M%QoTKU?9qHkFX0H&=cbPm9{cX|Nh+rOd3E1v|wx+3pLF=wH|q zc!T%VI>;J2^kF&7B~pV29jHtMY^VL8$4g91G_tlf(jh~SzSqSc7d;Z!%&2z6L?mOv zpR3(=4}jWXPlL338{V6mmS)#SB4+gF4RfUQdXiCg_TK*^tmveViceP$f_?PiLqz)= z;7$^hk6_J(!@maLJSmH7&Vn8kPP(2RWQ*QUhK27qk#^!CwHpev68 zlj0_uPRCe>4&Yay*d9>t+fH@_NC*gOyz4b5fUlLSxRB!E;trLV5)TSB4h$&M5(>hJ z1IN$y7kg$PWTGc}V2i18rZEuU?7DKpu|s$iTPINh=Sj2-!z-h16xdXx-Q=D9KO_Px zYou<#kue?0c7WIDczE_BRdt5StaXj5F`$p>isfdeqiZtZ>iY6%28@cp z>DXl{HM&jkaBX$f;%H|YvYd9M1N8r*+i)JjScQH1k6(Ih?=n^ynsx$KTCD&z5L_Pq z#wZ}3FPE@XAG`Zql;+U>=AEc#oyX%@uJTThMx2)1UnD71nLHWV5~?!)8y+7XPP}iZ zO3l5|+NC2A{xWlRFnsI-1t(y1tttAXzP`R;VPRQ1#^^v(ld9f`#Ojv5tQ+#F zj2h&YKqE*%ciO$FFO*pThiAD$|BTxYx2C3Ue+MUNK~q8D_tkUSk`SOYxV`0G<{Bix zMDB#OX-*NS0jk7}CXDI~R0j)Lnr-daQ_{@ zlQ58eJ}0le{&)E-y87UbpMx-+qw+3dW6~2E5xLey2%OmIsY_;J-RvLgUo)|?&ZGPL zZ7TDCG*e&Nj+yf)Rgany^XJdnpBXjG@T-Cpjkn{*MkS)~@@0edBn&*winQb9nypRV z#HRhLRFm!Fk{vWuwSO)x%xq=zz-#0^<%U^^c;m*D1z;0Amh9w*Ax`ysyvc(rXmq8L zK>~Xt^a@!q6m!>k&uKz~A|2O!DL;`hf%G0Q)dcS?5t=o5!y>u zQ`R^n^6#4+JL}ozDV-d^(j!7TRl9QjQ>!7Hk_0c8?9Mt@nm_ee%bXp`!QtL#$<=|L zYqbUD<2@nNp$uNsWBn@5i!C2FAC_>A#rKu)6Ck%l+;K@9do@A9kVHnv;3B|TmZ++K z!Gpe`1Pm@H{l%~^+6aPu$NHkOItB2o-P1a3Yx&CWI}Wxtkrv;|tMP*W!NnG?Fu1tC zY+jvZh=|`j7KG0m_QM_bQcIh)^;LBsH=a;~=k??bd;Lg=&>TnP1%iBk%Z97%@!xmj z!GHEBv4&aTdKo&q`XvZ|nOyTT>g1gy{Q`%!;tcS}Z0_rodyhqgV(K%}IgrXB-ZE`Uq<1q4)8)WlrR{MP>cMDCNX^(%YFU_#Le zL<~R+(XNhw&9;Z8R)d=l+ISGs6WEhN@iBFoF|VK?X!}^olOn$x80uM?`%;_0)0mo? zf+pu*-be)swJ<-g!TJeK{EdqG10A;&hP-??BtATJ8^W5O%`c-a<$p4W@pQIFW_qOU zUrfPGXlQ{54u^#wa^CpZSPdvzp_Q+A>gMJKVqn0li~&oCK^wF$Yt*naF=aHo2WQ`M z^<0&9SsOB3wWv92p4v*ShxTLo!TUR-!WPBaUXHFCtO8bptWdD8{h75a(!$zU8$NuP zr?(fox=d+RHt$F^+T%ybj~aV-#RrA}xWo(vECNPtByXR-KFa^UcTppNeW&!`AW2C{ zkZoJq+h@pf(FT4Fr^}c=8oBcfL2Q-_3>n6r*%bI2U(n}F?c_lys6$!F>O-@IE+58X zt0a)u?Bp(k+!qnENerlL21j5BuN7(Ms%VCfDHF2|_kaOcr>;=9ABZ%m#PT4&5_AKhl_X-|jrm^W znObmx2=o99;aeFkML8O|QPQb*KmQ}~zq-2acq-$*eG?)Tk`)OVWtY9nsK{y<$w*4} z%FIkC6fGeuX~`;OuOg{rhisATnacKFr=IuodEe*!$GOir_xO$PxP}K{0NcWa`&P$V z?DEj_?=q9basi~i&lf460n(18>d!o*mHGJ~&ihCz zW!&e`*<1z9^?+S;cRoI9AD{G){CIAEKuKhuYseUrFF$NrsViGre&)5Cmmm#3AMAz3 zLTPmJU#P@*f4_eD->betwPk=GJs(3XWMZ23k~>hO%t5S1anQvhPiCu2=qKGOzSf0Y zBsG6o$yMS=S~HH)hv+_KjI3VS>UQPr8(#~Ak*WG0a^y9R0eE2XzIpr?uV&kK8II|K zk61=CNY}Dm1FkW^Ei4yo2SCPK@)qyMFHx;khW>p=$ZI89PyLsnpvW%7%J#5QZuPLltHP7-l|{vEy%oaTLg3p3zM1OnfEOAs6y?v@k46aUi(v<^|hJ`@r#6r{IUGWJ{pN8fjFBN+V$(i7%aVAn! z5b|)4difCQ`6ER}74^FjYS{+_tjejyez#%N-2mjCAnlaX-|mOf@I-e+`E^86oeCkv z56g>LjH!t2lUKI|FXeS%ol^=NYlm!(_RbTXmv@q44eoOr;`$vgw{5p^PN5*04-V0H z*8aQRHOt@vO8;)|d(tRv^x$SbjfPm0>PUQlUB8JT0yl8IA*>d$UKXIFZMF+Dhn>B) z#7G#&d=Nx2P#AR}DJcV{3)=1L|Vcx zc!!tk&QJHL48Bb0s!)-Lkd=ri&3D&+Uv(7#_IxjQ(m{a;6|Ne0-YrzOl5f-RD14?dlAO`c%yO?)Q_+p<$NqS8M(D29 z|I&yiHHxIPXFNiIvQ6IcGwiXB=><<5uWwdGVPSM6KJBZ-SH+|BS? zGEMo`bgF{SsS#k3ohvvPQ;lDvy%Oe3LwPdAX5NU+i!??(DoNLJk3B zxr-_B5|Q-|XKn}+sq{UJ$%0A=db+y#&JOrNbBefrd$#t=W_EUC4MIlBqE_5%bhYOJ zg~BplN2*ZGlfR{~s6|=;J>$gWWOKjYdL9!2$bx;3wh5~8Zi|n#ZOe>D2Wa&y%5?K9 zxPWDO^xO?7;6El*QbeQZPa0HWxT=)JrP6I1gGhf0;LrHIVmc|8C<#<6w^ii-EloGlZesKk z!K|o4uDrFXp0sav(DE;s{C>2m_-}M#Z?|VOOMHA(5nVn3y;HBU|9vu&meJ4gc+J!X zXruTnw_kqa58UquCvloqevi+(PUKyrT{Kq=<1nm=+3w%i=Cki@`Fpxc>%XUzgd@iN zI2II2CPXJHH<2??9An_w#OiMF(D1Z<+OaDUt=^h9!j+8&hiqGqoXFC7Z8%{ir9WuZ z!khm-MbqKNo_(|1q_hUN?9)2Aosub;s>tGZK!RL_if+;ri>FasXa`eogT-gd`tl3) zroBlLuOxbQ@Gopl6I+USAx%(g`-IQl@hy?VEiJ|{+!?kjg^okHr9=c_Q%G^saz72X za4kg4EIs6+E}n=rz7fV=Ll$&PXlO&=fGkYAH&;LiQNE9{ztmpiRUVF`>!`lVa4nGIPKN9+QaXd{IRa}+b{b2IxjexGw5kgcyO`)K`2 zlUG&uS!te3)7VIQ+l_6Cin>J7vG(~@8dx?31rE?=s&CzTv6(_l_(ZeX?HAH5*ZDHD zuhN*in6BOA&6KZkg`vFe4434$rU$`Y!m^^?jjfWRW=-VOy~1x_iQJiD2p-(zB&{=( zJfRa@pxy9?-_m#A7Tg&yc2BNe@=~?VC_7Vw4blZRzY*N3r*Z#PyjyOViuDc?X}Vi35xhY&v=KUTE2*9i?>ZX_YQ#Gv=g~#M!rq{ zKuupYu3sT|YHVq~| z&#oz!9fSXVbUl=&cXzQW2Wb)i9x49TS3U5T`08#KS$>d=RFx?hOI_Lydu7H{6I|0u z#+>^ayXr`zYi%RmSXc0`=^Dq6FS2USRL-j0q^+FwA$^L9vGl<2rArF1xBv71QO@XwjROlvAGj@$^Z8$EwOLq`CB)|l zgE{s89JiIxZ<*C8KqvDMq0d+G;X};7=lGq2f8qDHY0SB(K~h5K$HKqh$>o``wsed* z^Caiq_q?|M`M#_&}ps_1~MOw^4h>@%_uCjKc)M z%I`8F>0E$?mfme6qu=RMySBgh?+0EqI6O(Uum{7kplcCQmuE#49W2R+FCS09M^H8O z%#73w>71DOy#;{(SY7$qfLL#MeJ#+QtSyRN2E~|*1NpT01v??_8Qtzu9{?Diw~o!s zxR_L$=$M}1Bg6wXRU|h-B~3PPt<FxvHP*rmiHy=0Ui(G zZ<+76E}p}$KRe;$cr5_^T_ik9xSdIB1%OG($;Gdy`oNk5aT=N)CM@r3Y7~;>(1cvy zvvm04MM%mB*DAksbSHx7d2Spzb8gK~Zud_kTkE-$#(U8ZQ`+7aW?{eLt{ijY7}hi+ z&YWs%U5M}v9&C2O>YFW^$>ZJK8r=2SD=(e-RFms4u4*Ll^A2iac_vwexJ}~&z!jyXrIlYAP+cJ- zK3iYBW4No(yJ9NMf6m&O2l>$?mZN#a+GU#92dj`pwuMWa&@y zYGPF7xTuokjyr1|uwOwT;m05%X$0NCMk($qKAuiNz)CpnFx|+$C&T%iQBow{d$ec|7V4e=k!FO+n} z_s^dyGtW7mzOoX+K|xVNNhIkC#MXv-#Qyc`sO+5h*b6i0>az^_G93=mX-%x#xAQvG z&bv56C8%<%a-~V6Uqm$j0%lOn_y|Y(IV?d z(F|46G(Oy3z_YJ1EuAmDQL?2rH<+=s`s_DZU1PJTLEdfQFN)n=&|T8|sw?X4V{4jz zQZTc))$Z5NpJ^F6WvpQg-DY=R6wil$!0tF@x71pPLtQ-dmb6@qUI_~oW()W-*5*nl zVB4f7kf@AyCd?e`HVeXOqIl~M#ov|hkSu;5wG%h1vAD7IbF%t$YX%F^jRr61P1!$O zOZNa8M!mRWm(C2z2xMerXoqX*=!ggkA`9EMZy(YB8XpV3u~Q)*WQ18&Wu;#*k?q^! z1#_X~5a?}&$j3PIEkL#kHUcX}!^u{x{6az~b}#D%F5Q7b8IqQrVs507@s>4=ZQt1t zD@`@EE2^pJL?rtOV`}j3z{)W{?_To}LJM;RPR~z1hvZX1T%T=ff*WaK`p#u7^From zp)ZHDOdJvK+wlWS?)na3e{7$UZ&y1N;DP>eKH(phVm2XV(VVBlsi0`EM3lJbDq=*wvblqyB(Bl*RzL2 zoi#8p06L8NnJ+57=#h|ZprU?u)Iy3|cpQj|5%*#g^l6_RrN$%unqUO;4b&3QJ@BzS zY+2GHIzj*0wHr*9Y4klUqnDmK^Dua_l6qFAqI%le7~|zOre6g`Ay)Y88U4y{innIa zcmf8t1m7)PzWpV7g`tin1S&zH=*@rqYE0j221k(D?rsu@AW2$vv=4!9t5pfe&m z=wb(g*9X8h<;g%4aVfs2mWMu&j#Kn;L$_tY6`X0s3-uhcgUQPNCTUH`%&u{?J4oA0 z-beXAk|UP>_N`mD;^X%QbOQ&rGICgDn~IaBX857BBAoL=ss*&1(Rd2q^#^*!J}`@T z7mYR^&5c7Xjg4GxrkGw$Ho%Yzx|x1gnAt(vbiYtq1|#$cDdq~JADEO@q#PR`XWA6{ zqU)k!YOCF(r!V+N8d3Hyjlmtj`Wj+3vHdzd+(g+i`6itB^(3XGwRpE#vOH3u~=35E_CQcGV75fN_8doNZpF zC7^K)u-@LpM2^2#IXN7SxZwlI`VtMncXX#`vzy-Dg`Ho4ojghjXUdT&p`iCyM4!C!FePfAZn5#CO`_ zrZhWB?A)Gy)^;ebLauH&6r;kz(2#jG;d}?9fNm*ThiK)pq;A0r@DEf(P`gRv;}Q17 z(c0Rdp%O^9OBsNHQb1I`d^zc=3QS(#$`2VJ=-l0^BbeNZN?#4;FT~t!aS^Wd^kz4C z$8=G`o}u`cHC589;+=yLMe%vY<_yxW@^k9yzrMa1jcj4-Vg((mfNF|NuQ__*VAbuo zVcq}cGM*sM>u;nX|M|1KUqlIYEz+X;WTg_AYhz+a$;jkDGZ~;~Sda3Apge~5{6$Ys zPvFFG``R;4-HD6aGdJeBL-(`vLZx+^yy4ZWQ~U`XjO|YfqO3nex=@xda&nqmSTyLh zwYJjuI;x@IL}ILuTdZG#4zP3SELD{_P6357;48wTqCADSb7vni6&WZ)Rj-k^-U%xSxp9;-oOfr(4Qj#UC8bnG&st&vSK*2$WPa zze>}agpCnuYFAQ9T?Y^0tQDjN(2+z>8?238j~_j{))t0|Dps+wUy1ZuMfV@w)z6o=PUd+_MptYyO3&7wX3!X<9I$vOYhY=FQH}KL$n-MW@#pB(6F)&JiN0dVqkz0p`Hn z^D^<;0Q_Jo1OVoHo9`&PayCWOso=3=jQRp~hueAn+_}aST@O3JGE9jhdV0f06T`-5 zq+eeX`eEI6>DJzwy3O)|w=}vwn}|7IaJHT_NjS@?c~k7of3{^=^?%#*-?mJuC^h5% z-#4_8=e%^~xLWXs=Sv4^^;XxQ23wN;6l@HX6)+pobh)LVmXd7(wkh_?V+R$Y>#VV|nj^0m4>T+HD*}HOh5sW*j<8Bt zUmsNTEGfg>z#UU2;{F=pkQ%OBTwMG1xuUlN*!Hid21Z6LYtXxVVC4h(obPTvsH=Uu}71X{jNkhs7O6<09Di;2lI-#%M@8fHH z??z9R{w+~)wP)&0`~)H?P%f>qA#(J+GoyML3c8P8!v-GVWY*&~@I;d$HT$@B4r zu1`Z=gzN!YFzM;x%KT5!R=daRpbxDeNT=ZjWsE!s;{W%x`BksA+L`6qF%h_F zA7@)+qawq@X|`-3fs_^v)Z}z#Lg?k`Z$tRrmBU|r88dES9oP`9$H&Xe`&^YS-;}{v zjC!J*w21(mYd*9QRw(<3L`9$5^xGI1zD-Xz*m`u@O1^Dx zYis`4+D|`}c>+lb9pWzZCUa?JjdiwX-@Co#N zX}CFR!0JQ9_;YXk!^P_rXDctp^cx7qsPlZYV7!g)TfpA=OJ##TV+cZsekk=_Y%$Ab zuBp49Ymz1T6o^)Q`s0w>Q<`M@TIo<4Gfx;FQ93;akM9}a+& zMg#>2q#=Gid~{@J5s~lQy4#(8Ebg}M0j)Fc@;eC7NWf64AkJTXWQfZW)z zLyJ0I%DxSYPQ~I6b^;zCtg~EiYp@Oo!hL3*B2uw@Mg=6DOO+%K9%h)3lRC07pi8Timfsm;fVSx^N5^xPEIf{O4gh1Dxf1U z7zYCOcJq20^$gj$c4wv|DkmdVlSbNEmL_X#g9SoP8XAh8$_)wQ#L84QypNan9MCB;YX*+QUswy9^!XUg!Wh73Jn}R?tw}HSulNvhz zaFo%$$!ZmH$ePZkC<<&0!Xf~D@Ld55mbgpUgP}&Gqpkg@Dl7X_&jXbmZYZ36 zM@;~7;^}v#aQD6Qw#o37S&2N1Ktp+T{zRF%3gmvkt*diggA0!f2Oxb%rn%CL+IBWk zaImtDlqZ!U*F;f+IHIddMRdjqh*J9a^%V}94P+PB)(r%8N&WgHlmmJR5r>7RG5s zUyCkY0EiGxL3u?ErP~3(c}os8I&afc-x#qDS6<^R!#f{eqfv_4#mCQ&>eFk$Aqkfs zpV=~6=F8H#@j_3O_L8rWd#+;Vv-|3O;cj4Zp7>)w%I$h|jT zw696_YBh6Wu8AS>uaOyUv?*U7h!jh#_XJ)F*x|<6F#NOGnQmM7IeZa3f=5V|89`Rw%mYKaC#HD6^x!3fXYJ0IevVL4k}f*S)t&&_##gMm{x$7w^0GDH2LZVLkkPNF5wt0@X$Y{r_+?it*E?6yJ^$It0&jO zWp>u~^^pLcBze41cQ8}dao;=_H#3$#!IvUZFP ztgB5a-;ZI^t`5i!O;1~xnkxGKcE~bJxO-PU=kRN01_ls7Wuc4)XgospM~@#PfL+8e z24>xS;=!fg6F%53NeKpqJ4p`-Q7F3iz&rPP4{o`nF$jtSM~w(T*t5c*Ko1cI3g{!u z^}Z>+WXI&ECgzuSshSakA1x`lEg{CZA2|t<95J)vFQ*;3LS&=dypr+!u-HIc>5_D} ziMe?%sYv?i(@1RAR6p3As3^K?mxH5%Tbm>wu(sT{9)KDua4%aOxaMD1NqFGFiBO~B(<52*6h$5eNiCI$Aplpm*OniKL zlZ|N|JTNL21|msCi+MYDFMw%w20D(=B7q`#hM;tLsf3%_gO%gq;ST~xHvm;(Gdhsv s%26Y}`M1iL)_>KV{<}8$U&+sEpv=5@57&W-NdhlT4PEtY)ic5W1H%maqyPW_ literal 0 HcmV?d00001 From a3de8d5d703342cde18dde115464907fbb856981 Mon Sep 17 00:00:00 2001 From: Fernando Alarid-Escudero Date: Fri, 13 Nov 2026 19:40:40 -0800 Subject: [PATCH 09/12] Add the dampack hex logo, favicons, and the scripts that build them from dampack output - Sticker (option I): CEACs with the CEAF above the expected loss curves of cancer_psa_data, on one WTP axis; both frontiers switch from Radio to Chemo at $40,000 (dotted line). viridis option D colours (col = "access"), JetBrains Mono wordmark. - man/figures/logo.png (240 x 278, 17 KB) in the README header and on the pkgdown home page. - pkgdown/favicon/: favicon.svg/.ico/96px, apple-touch-icon, web manifest, from an ELC-only icon without text that stays legible at 16-32 px. - logo/: make_hex.R (SVGs from ceac() and calc_exp_loss()), render.sh (PNGs, favicons), README.md, the SVGs and a 600 dpi print PNG. logo/ is build-ignored. --- .Rbuildignore | 1 + README.Rmd | 2 +- README.md | 2 +- logo/README.md | 42 ++++++ logo/dampack-hex-600dpi.png | Bin 0 -> 83048 bytes logo/dampack-hex-icon.svg | 12 ++ logo/dampack-hex-small.svg | 25 ++++ logo/dampack-hex.svg | 39 ++++++ logo/make_hex.R | 130 +++++++++++++++++++ logo/render.sh | 67 ++++++++++ man/figures/logo.png | Bin 0 -> 16772 bytes pkgdown/favicon/apple-touch-icon.png | Bin 0 -> 8386 bytes pkgdown/favicon/favicon-96x96.png | Bin 0 -> 4035 bytes pkgdown/favicon/favicon.ico | Bin 0 -> 3609 bytes pkgdown/favicon/favicon.svg | 12 ++ pkgdown/favicon/site.webmanifest | 11 ++ pkgdown/favicon/web-app-manifest-192x192.png | Bin 0 -> 9825 bytes pkgdown/favicon/web-app-manifest-512x512.png | Bin 0 -> 28908 bytes 18 files changed, 341 insertions(+), 2 deletions(-) create mode 100644 logo/README.md create mode 100644 logo/dampack-hex-600dpi.png create mode 100644 logo/dampack-hex-icon.svg create mode 100644 logo/dampack-hex-small.svg create mode 100644 logo/dampack-hex.svg create mode 100644 logo/make_hex.R create mode 100644 logo/render.sh create mode 100644 man/figures/logo.png create mode 100644 pkgdown/favicon/apple-touch-icon.png create mode 100644 pkgdown/favicon/favicon-96x96.png create mode 100644 pkgdown/favicon/favicon.ico create mode 100644 pkgdown/favicon/favicon.svg create mode 100644 pkgdown/favicon/site.webmanifest create mode 100644 pkgdown/favicon/web-app-manifest-192x192.png create mode 100644 pkgdown/favicon/web-app-manifest-512x512.png diff --git a/.Rbuildignore b/.Rbuildignore index f599bbe..731edee 100644 --- a/.Rbuildignore +++ b/.Rbuildignore @@ -15,3 +15,4 @@ ^LICENSE\.md$ ^CODE_OF_CONDUCT\.md$ ^README\.Rmd$ +^logo$ diff --git a/README.Rmd b/README.Rmd index 02b9fa3..258e6f0 100644 --- a/README.Rmd +++ b/README.Rmd @@ -17,7 +17,7 @@ knitr::opts_chunk$set( ) ``` -# dampack +# dampack dampack website [![CRAN status](https://www.r-pkg.org/badges/version/dampack)](https://CRAN.R-project.org/package=dampack) diff --git a/README.md b/README.md index af7f2d3..2e23250 100644 --- a/README.md +++ b/README.md @@ -1,7 +1,7 @@ -# dampack +# dampack dampack website diff --git a/logo/README.md b/logo/README.md new file mode 100644 index 0000000..7db9452 --- /dev/null +++ b/logo/README.md @@ -0,0 +1,42 @@ +# The dampack logo + +The hex sticker is drawn from dampack's own output on `cancer_psa_data`, the package's main +example dataset: + +- **Top panel:** the cost-effectiveness acceptability curves (CEACs) of the three strategies, + from `ceac()`, with the acceptability frontier (CEAF) as the heavy line and squares. +- **Bottom panel:** their expected loss curves (ELCs), from `calc_exp_loss()`, on a log scale. + The heavy line is the frontier, which equals the expected value of perfect information. +- **Dotted line:** both frontiers switch from Radio to Chemo at the same willingness-to-pay + threshold, $40,000/QALY. That is the argument for reporting ELCs alongside the CEAC and + CEAF ([Alarid-Escudero et al., 2019](https://doi.org/10.1016/j.jval.2019.02.008)). + +The colours are viridis option D, the palette of `col = "access"` in dampack's plots. The +wordmark is set in JetBrains Mono Bold (SIL Open Font License). + +## Files + +| File | Use | +|---|---| +| `dampack-hex.svg` | The sticker. | +| `dampack-hex-600dpi.png` | Print: 2 in tall at 600 dpi, transparent outside the hex. | +| `dampack-hex-small.svg` | ELCs only, for small sizes where two panels are too busy. | +| `dampack-hex-icon.svg` | ELCs only, without text, for favicons. | +| `../man/figures/logo.png` | README and the pkgdown home page (240 × 278 px). | +| `../pkgdown/favicon/` | Site favicons, rendered from the icon. | + +The SVGs load the wordmark font from Google Fonts. Print shops that ask for outlined text +need the text in `dampack-hex.svg` converted to paths, for example with +Inkscape (*Path > Object to Path*). + +## Rebuilding + +From the package root: + +```sh +Rscript logo/make_hex.R # SVGs, from the current dampack +bash logo/render.sh # PNGs and favicons, with headless Google Chrome +``` + +`logo/` and `pkgdown/` are excluded from the package build, so only `man/figures/logo.png` +(17 KB) is installed with dampack. diff --git a/logo/dampack-hex-600dpi.png b/logo/dampack-hex-600dpi.png new file mode 100644 index 0000000000000000000000000000000000000000..fbb9b12483dfbccc472285a5791a2329c498a39b GIT binary patch literal 83048 zcmZr&cOaGR8-7(%Nh%=~O3@dg>|~XdGLn_8tdPCOag<7;RFXZS$j;ug>^)EBu^oFJ z<8b&r?`eFF@ApUki1VEHeV*rj?(4p;>wdiN%1cumI&}zwAPSjV689izA0_zr?ZAHU zKOcJaT_NZwBqMP{*(q$g@1UD9wrG3X@qJgi99I|ZfoEs`sb)Ix&(GV`$%lTX-!h*v zr@ox2YMnRprD&!DlZ;w4+vdQS`y6_dYEG?~aI=p+JNoI<1FxPw6FY0rB)g)XzBzul zn?}99)Zn_aFv5+|cBj8UBl7OE-KUCB$ zaVbqVQ3s;li1eGiFeq?vDtwZc+Mi{W13_XX5>DW!D9`c>!<96;npMUmD`XR@d*yj* zj>f|ek)tjTd-~JoW;F6X-+-Vf-E%_VXJoK??n9kBcLrKQ@{PR8M|?AKVhe^WwU>0! zc*A*f2`_kDF~Yc4&1)7Fy&ZbG5dpJ%!lQe6i~~+^TlR6Lbe)ADJ(p)>5EK-qh&p?{ zAv+)Q*7ZG8(%Gu4_@r)8%i(lMX7C!Zl3*MU89IB4N<$WW?B`zY5^j46`Hy|aV+xZ- z0wkIHPiG6|epQE{Qo93S0k3>g&q*naEgRo%sA=aFP)~VwA88XkEUkccXu!kc-Gs;U zu3zi=P8Aaq6De|cv_^%JE+|5)yT4z=wJ#iHK?6o1E#C`4vU)$B7AQ#_zU^#ksXo*G zMyGl2Io7o!`nMcvU+m?LVB_y|RC*5HHBt(GS6$WnkAhE!BVC=%zbk|kZauM`%W~N0 z!MMV&;Adpu8#bm?@O|^f9VZrWQBB~cMM%c&SHU^6w5Z2N!5m(Nx7hMq8UFszYgc&J z7m5tCl&PuovyO-F_GeCOv16f*kzyBcs7r&hC~|uWl8Y<7FIxuBBU zAW$?3FGCT_t}enVFsNOS+Nyn4;z#N7NUITYpjGW7NB24KvR?>`ee?S`r`(zFNNm+m zdN-Z<%n81inL(I;M#pnIk@vMW!o%4t;^F8&4!@Ay^Ifa>jDvdaTbA(`!N3%g9#^w@ zGc+qWzZBuWmH0WDBg$3OLF<&1lzmPeqe@q^@OtI`MnF@)1J(sB_DeXZ&)$r^tP>F! zV5*suQ!jiQqbQ!PeO;)(v{!9TBUclG20k1E3uE=Fl2eO8?d^DO35Y~As+(o`i0U2R z#SATl$32=WMcQnEJO>~sj0fIo@*2|Z(>;uN{cQvItb()(mgU4U945bY5@#2DOUmXc#~HOKY^aDT!e7DYefWEf_nYi`ja45rKJW$4p2KT@ zMK*t-zy3*ljo%!5VtDt1=tiF-iIV6-tIVyLXzNeoaylgpuFjzmKfh@kDZL%;R#u&f z%(3o@J`P@5o^X;;z3Bf|FT!ztLX9kOGSzZ{OcuMe^fcqpGQRHOr3O03_Eh?Wc+~dR zx_MDno7U3f0%j>^^%^#x+lBm(ovjozz^QW(o|k!2Bbt;TRO@l4%-EE!b5xhAtjc;ayrQ_)`<~axk^UIsom6gE?hVK{^V3sKVRX z^#0u$^dsB*G(u{*dRbs&kHZgu?|v?du)3fAJ>SSO*!6?#1%ZW*aylyU@DurUjJmG6 zNH@dK5ClnZ&A?$bMb9SDsems&4hJ9)cla=9=6tQ2D`fFVXS#OCW|e*gj7s0g z)DYIJWRBz%g+T8U2631%w&KsnJYCRn*xHZ6E|2in6B5kc2OGp)aJlR-rSdx(ntJmLv37miIxzmre`Al+MxKm41kS^v0s+K=FbW61w$akgmD_24A zd6-ZXJqSi<8MRv8t`%NFuq2r2tyQ+SBoa zTbP-BDd$%*4LS;zvh;&dU;bjq{>W{9Ui*e4!PR-^7ni@MM#h9G$Kev_E2uSA+NG!i zf)i(>*v@rEhGEuP+H$n| z{ZXmtlFf*S-f!ZB0Jn>Zoi+Qtp_i>eWOx+A=Q+RN_R3=Si^zQ!qg%M%rOMHLGAkC8 zZ*)?cDHMnm=f%1VG+9o}gp(Pt{7KlFY*X%q<69LXBv7+79N@U5oe7)kT?=VfzH{RV zidtD!WQM(C{H(KS%wbxXnfxrrdR=DlxwGIs#3F466&TNqWom4v!zqC3 z>Z`Nv$T5jR8>c6 z+`s1b;$M4sYj9R}CF43X({P=n`p=LFmW!qpD^H^i%(F*#^Jf)(l{Gp} zluzcIKzL&+;Abi7hIckVOZnMJ*=5S4P-FES)X+75UN^D)=4(iFnnH=IMMkY@y$a>I z^Vf=7#=c#4O0g<41ygwko=TT5mwW=p4dc30Y3Xeg-g|$(STW_=cJWd7+NbiG_2I;x zWiK_fnh)n$Zo05R)&$CZ9In`u=LMvY7^gVEI~WzB6l5HozuzHIaw+yrCqL&&cCQBI z>zppn3O-5Jq0Bq)%(*jhr~MblF2hOGLq~sQCedP^#|b%O@i?tUBK8e^^juSz!DR(n zCNoqSE4%BY976dXGq-hRmu8#xLJml+Im*wU&+LJsya;d7>l-O7bX8CgH?dbGLM!;z z4To1WO~KKM>%lisE{1gs!ns@YDk$C>*>t7j!YwTy)u-I88}@^V@PsE)*r@gOs?iZy zId?Y#ygU*tWfLt<&Ji9p;s1iXuWHJ&gHacuo5=q-5G|ppa~MV}h8Dm_wP2~c0nDNg zJd645@ZOdlH^!&sCOJrH7C0X|ec?x!J+@+)XXDQK=l!WQQ>Z!J?u3iF+hTC;y2!sP zciF)wjTq{m%f1n(xt(MJ3MSs*j3OmjxNH~cG^}czt_{=QMB98z(AnM?=g*oL_T=*a zKbY?26?l{%+U^_Y+cXhRs%lM7zc*xxAy2m-PO7PGayW?S>FoQGKOX;@H8~|P_%lm| zv%Y>Z8u4pyAH4z3#mnGyCubei1oOUruGN2nZ#g6=p(C#;>Yn1ge#iWj zK9;X8D)3#UePF4{#c(1TpxLu~;2jb7c(Z)en$-Et#vS<32Kme9et`Yl9?}}&0-^^lMl9b z(%@cMtjW2<$R3>}v~Z64Xch&R|MUk5zY2?@$*!Ih9Zm=;hbzhhb)QeqHJk|w*<9;S z?~bP*n<)I6`Pqp*OyoveB_*}q+IKcM&U0#}ECYfM*Er0~JMFdAEixOl0MQ`>Po+Ga z%XhAgT;I>S^hpUN#TATK8k#6WnWf-J&*3<89z;4l^D&YRY=g_AKBJ;SL%;eTr#4+9 z;gVHBrf!p+lEV~~Qo8Vz%s(fFXV;Lj?q7Xw#(kY!h$@tj5D z&OX*x&pG~Z0nker*_K+%X{Ng=NB3nr=zb{>W8o60`w}`H5eMRaPx(S?f$Q=-uHh5) zL8JGfV@dBHjwNqY@SffWm0pEY`x~x@Wqp^XsKS0N$wNkKktA1ObD{3E5vzr zVvqbs#l2-;pU;oHpkECA${YU)6vPu^a(yViK1csN4S9g8=|FBCh=h-Fsqt;cKjmg{ zHT8I8{G66h<0H+BXOFY?Qs8DXLYH%9HvVs3gA-F}97hTM-9&P6slH(rMGe}aei*MBSarw7@7O;%&bLi{7K9+hMib9_k0-$TQJW@p9a@Pax z47g@s0k7NMU>p0W_nQ_;-(XjcNNd74T1DGG&!Ur3R)GT)l95HTjb{AhP_re7d@)g8 z_rb;7PtIv=pxMJ2H|a4lD_1$ia2Qy_;Fh+pXx8LuD!8_EAQ|P zv+>_>4Sxj8#PW4|#SBa2h>L>fDr&ZpMu{5A-R$M&*=}91sSHhsxqyoT#a#YsJbxzE zU!2R|$tziyts~D)uBTs9<^E4rz*WF?G`^Y$u!RLj#)poF#ce(g^ynX%@jHUaX(~?M z8(`1n?>n+475Y7WAg`P>krcwvf_RBy41xpSA;jFi<=pCp;9IJ0uDbDcu6e9-jBV4=hlOQCu zW7OCpkZM!?Tlb9~{;ry{iww~{SvE2NEycni$HF_+8Gzzh+boHXH{(+w@$jjn>{vCD zRSw~n7t>YN8tW8RDEdVCip_xKzh#&bTn_mMFJyRKl#-SfMMTx=Conw!zW?*Y7+dV4 zHApGIOiXkIw^4!X%Wp{d%8P2ifv{Jw%QWoF*VW3n0YPnN;So-z4a|Nx=6NroTT}IV zR*ZP;gQP3D4mb&Ix0SmjM$1Ufk~q<#mEuBoTVBq8s;$jg6IYsgavv*t1B}A}FQ)zT z=O6Q0)^h?;e4C$Zru;t2{JeY>lrmrMn(vLNB+)a11Kz5Sr6~Q&Eu)8&Th|+jQyL}% zUJlAoz)(e?d-9`O+diD7T3^JJ;mSt+KTO;{(HuG5s|)bEW&5mW2N#vv%_PxlEekPm zmIb&c13d~<$*=c#Y%axVz~w-T2-gUyQR73HCLNbmD&fpo z`VO9;DI)|-I?#ut{odShAkM`Yxrk^Pz35IKnFWAdC%W>%?%X%_O!Zn+D zrjKV_i1Z@mp81M?PS!SrLP5Il5DiOZ3_-0zn?>zZ1w z86Q63F)zF{CNI2LPSe2~JY|F@F3g&z;JM)Bh6DYnP!*neG-6$Zs)P5Xqpzi)e99l7p$Vm+X0NW<_-5 zRmE|?8d{Ip443am@aVd^_?oG}-8s!=AH0~)%5T;c25lC2IvymD4NFBoA7o%cX?4{=3xPop?!~cl1{P)-D`W;!>-DhGzySloF%)sdGu62n_MEh0h<;=QeL{k{kD*QVR{ML8~!YGk>5 zcfe0{%q;F_43)X@#!N{)3)x?3=yQ7?gyDi8LFGS{$Vq#Oln_4EG>1fJnI1$ruFp0U z+ubS5MfIC?$x&_B&iPf8F}(o+nqNvpY_TkUQS`L$L$O?H3-|{@pc?Ti8Y|0EGa~od zxn<;DEwOZ)qmqmCw^&IaCYUW?dIWE3_4sUi*Sg*q118T5t$gbhgV&d;X`m&i!SFs8 zCs3lBKYAB@m^gVIJqBLjD9JD_haX$DZLjXV+2A=dbAR;<6+QM?=3#7B&G~($VH$2E z!Yj3W_=&7lLl(s{$2{?LB|dQUTsIznrD;kE2{40EtZx4Yt4(vhZtu{gk9&5%01DCt z(M`FzZf?G*S5kkW|AAT4ZF>&ww(M0^YMPKljSG<7d5}H?WVqc|=6?)0w%C8^z{XB0 z3j7Xlid&9{#I~2*6@zEhmtNdvoibp7$v-|X!!q0F+eogJPlJO z5i>4Yf8g!DG2KJJ@4I-HH{M`Nym>$%{Ex^ZGXIt$2KmP`tturDhSNt$N^4w(QhUT_ z=tB$LgKOSCM!i`)we9=5jwav$WVH`W<(e9NT*oYJ7Zhd>UM%CUXyH2O1sEtD0S$m# zKMJp%gydB~GN(*NN+Y`p?-)JI`SA}Yo3o*!_v(vZi1q9<@E{`g>ob0`mRM2-wnXz= zAHB7%SbT-?L~EG!9*D1MKw4vq>9oN)blqs�TEKcGDHas2Qx_cPMENNF;x8BnW0zOtk$N@5YLB%Ie=r8?bRFwRz2x9?qzeh$+M`^GuZ4hl(d2`+o@xBMnaO zWj1d2V$+27?EK99er2=uxUg%tP)Ph4!~El{Po#npz z3!dS@!Rz7v;CH3%5(|q=vKKhd%kVb|E9f6e2OsZP3U!EX%SJmmDjY(~CM{-CO z6jEYnwOtxl@Us?+g#uc?w!-+BeUMQnNaVG&*^^CWXUk-hKAs#k&Wre&oipuPpvqx# z6k2`&&OBTuG2Kwdas;&+t8n+;)A(9mz^B7e{)c1!L1%0!=U9QxA1f)|s2k|91n_Kc zb0l~dj6ab$X24`TW;|Kh<*)MApB2KS`aXPBf^hIBP;z-DjjWf9`hVfn<|_EZN-vsp zLj)Xdt%z5Q&RXu%=rN$7t2wA4p#PLZpiWu+02KCK6mMU!HNWU;xy0Sp?}O+(=W~ zu3A`z<=)%y=+M-)_dvL$h0Y44TJ*(ub(X~3R`9PkAeIlB$>ERN|A$ZmWw7>j#8ZxN z0a?hyf1EsXQX$nGMTP8RaCuAva@p9fTvl`dehEWjI?Xp^|F%Pm`l~D;d%irZk1bmq z=bK3FS*evB`j>RDRC(~px%C)?0BW^|&I1DJKu?cfe!{AdB5SZ%eItL?%>RS`N zJ&;|;-S-iHxPTItR9(NKLOF%UKRLL)z2Dq(e`AJUDWXRA5{j~@PTTknq%9xuUJrCd;YhaAqzlj^}Qs_r{x99-F$b~vi4i&*^Bt_}| zNnw7Y`S+}jZS>d%5g%EmMvIOdPQg29u+d2%+#4KAs%r5!m8Rl6rT>%A>_|X1(H>gt zVGQn{Bo|Uu?a@of+6Mk+ zIFE8YRaoM{pZ(L|I3S+UYqgJnJl6853W98H`7<$P*ih`y^~ma7k!p1aq_MEfUovMw z^)H#TTCMU(OLVaR&BP;sfT&CH1cZ6mHL!g-lpjJeCd1TOhK;0?aX?b17K zWJ$X&<*&cuH2HB>$?zPA)%hLLlKOuIuCs^1$t7rxp{Sc4=6dgL|978@0vMrT)zy{% zxYddTnx`XZOf!h&{%_*NVu|%kyP)Sd)n0;ZZC6`gPQ9d|vB$5oo%gx{6FNvxLrt5f z)9F0WKHhw3ZTk#ZLYdRqGXcQ{7pnNX&wM0pqUN~I4z2A04h9l62V`#d#p+PrSuo`V%49%p>I?w0ixJDk-t7FMwZO8~eKUJJ6hNS_dF@g? z7i-S*Y`?dL87Qs2pz;(@R!Sw?+Wgqx?<}@k&IZu;EM^B*15_GU{#i|!MZXaoTMN|L zqUN2{^NEpMz5tbtadWNguM!XrFi261r@`j#Z7%#7@He4##mnRMeu+$gs+U*5&xi}- z7FBuqHVkMSfM%cmUX3Qe$@1soK(rAFCh!2Ihr8*H6a1>2U%3DvpFv56&e!t=LtKJ2 zWTyu{5w1(;iL?N`nANEV5vH|}AduxT#yh!IAP*G_4)OrrS25+seXy#hPwdc`V1D$T zmr)Wy4rsOn2NwQECUc!MbGNzKDzLnOtUmI^Y$2@Xl*n1&6N5rfy;Er-t}?GMm+vioMj%*=Y0$#~5wZJ%tJ~AWm+`^=AKsk- zOgS!Hn=BD*ael0#Ou(<){9F2)ISRpyoa_kku7ERp1vvbb-D817>K}*(nSE>VO~O=& z0nODG#o8w8{E7z~9_1tANa~cLF z7M$lHM^8{&8R5-c)9Nn}C1;qD^AItGAx=QIYM?4AG(mds+WGyHxYWo*ASHvX<>zoh z4HLm@`A9cUlaQH}84qZ!5P+vxBvAd9uPV`SHr2^zBy$LkL|mY%k+Yo~yQairaX$t4 zK$d$o8O=t?=Mqv{*fPGsa^Jhnk!-{$xkIv6>sbh~e~_eE+80;@QqDG6BY!ch8|SEo zrg5{9!#v}4=YcQlP5TIT3UIU?{;9~b1jUcrpUOhBPiONEOdyG+f}OQYl3}T9?-^IN zVdkjs6MqP*4|97#0|0!Z6VSU$7l2R!dqw2Tacfca7p^n6=%;C>x~ya;qQK`tp}hQ% zP^V)sm^eyhXW6WeNV^-^7Gb>VF#iO3H1^~dD6a}-DS{%NC>OIH^Uwi>6z2JV@QuNj z3zKAl3ax=s6hP|`a5mG~7V|6Kj0n+i1J;M#W|d{uXJlfQ_W*>a^+&G$sbw}de6uvM zGvrkmP1-lWVf4}$&gVkg7GJ+b?;?@n*jVyk?^!kn!*E^&rMgf_j=kiUxJx%!Y!-sZI~;8qf$c0I#_d62>U_=CZa}SuO)sCa{i@#Vt_0!Mj1+6qyvixdjz-oce2mG z5o1?WV$RuV8QGKcO2gJ(>@r}91@9`YW`PJdNU__8(FePU;g^WAq~xyaRxMlvL=?8g z*7rl$qB<5#bcPpr1iO+d9cKlk8x{dJIFQkHdAOENxInn0;NUDE9RGBSK8qnZW;q>X zrfj*9cMGUyu*u|A)eka&*=RXz5@#fwX_2ARpDs%_HBdHmkXRE#@O)CRIVlLW%zOGy-hZhm=UL}d0g;!0 z#OrF_r=4JEhE|p_46kQm74WA6hu(wnt?p~T;#?cBS(uwJ9d$Z{wjdog2oU1@QrV~V zz_8@t@d%(_H9IEZH@q~ktsnt}#yu7z4s#xbPh?Sn;Ka$7AhNzxuu$a;!wn79NQ+gCs{E>@3uU4_yow5fk0S8Nf^wm`TJ2n-Hv^cIzm!>{3F`J_Iv) zfS{lLE2I3?l5N0u*N?xZ-&pP$=SxhA)iHVi%OL01Ys@!lX{QPHR9F|YdiBfE#XZ6Q zmumLQnkt|T|B^Qj&!(O`(rpHT9}F-zT_81}ft3&=n{xBS9v@^ptbYoaJR>W;6l6bD zYj(u#j?9 zeD*crgqSTBw3YKR4O^5h^1`Hey z)ZMXqj{(Mm1gM*evF}ucw^JrV-2P9XyFC=!2Y^N7C1Q6#yb%hlDx;Q&VsMHCaIHz) z24*5@o!;nQVk7UR#?VJO+Bqw4))V)^l-z;*%@yg`9NZqifDXP3brSoH^X!&-mv~r3 zmB2%2kM5Koh^uTdQE4X+cXML0S?1u8hLVxT4MMAS;LGg-b3_HB&_YYF_Tck%z}JLG z=p?w?@U27!TG@7+>MN@tH2Zw2F7-d40-jU_CmSF@Y>lqo>^W7>wAP4?B>EAr}ppo!lO!Pey0nA-3pwxPxiH6WiLr(QdyKcIgy9 z47Z^oxt33-`c{`iVQ&HbQ8MG)S0IMeQ!i_0ii?Q?G}0DoH(V6FSIIx(5+RLy{|Zg8 zmf+LPJ@_5UgU@e!gC$yXp@Cp4g?(*7oz15|c2yZ(z)$L_;IZsm9`Rc#y6&@dA;<~X z<^h5O!{+ux71S--NEiEpLXxEq6js2#LzK*~oC71ZCTXhav<)n6#{$Qua>f+w=(6rK z7&R^cX5eNu6%_xJW0QHOndWLW!M^ivW$k}X$Rl1>(=Unxln1|h1b`bLtO2+Agw+qA z0yrsea9^CLiy;H)tn{+OY-+2p!j`tA;n<{y00{T*!xZ%hf~P46u#>XCNawM*&EmDU zt}dFcUiD0!aDja$#!_2A70iqgc5|#xV`x@P^~4y#2K!}nGpYvl zJN6FjJ|}X0fI%|M7Dx&DYt?qXw$ae-m)MIJ1>KjI&DBJYhPelibI{`G%l9s;6B8DH z@pmb3TY#M^I0`rm_oaH4qV+*{;&5*!JH`;zls6wodBj#D8QDdkD?T$PC z3Z4O1u*(ny3BMe|pYs z!=;Qx^49{3CcIC>l6M)f%r$TP%Ik%nYy0e^5gAQjg_o*ErpN*#V|+)Pz?%&GgxxgM zhmR60u`p}=Z&r7?24=z(&Sa*pu2bFV-7gGwn!CU*6C2-Yh20hKoRUr+Iwr)DX^r=< zfEO5t1AbuSF;H{>UXHzMEt&E1P-FcIApic%D{*LSq|RwSgfRrKkhlx7+EX0;_#nj zc6YD3l`_%DD`gTz*MI^{0^HZPVSEGzgOVfY?Jc9bEi=a-vRsx|e9YKJa*h%#t$ zZlJ!#_)!`2@TjjN0BMl$ci@+c)N$2S*t9R$WJ!n_-C3!_6ngV(FsaEz)#%ZfEv%&L-N?R!gKSh2zTVy#+1X z^}u>WId<=G%fb545kLMbs~rV2voy`XJK6k2>T>RNLj(8PIUmZrxSXuexNkSY`VuVV zPA6JwUy%ZDfjPa?Fh>J>>$%@Y_S!bPVn346@{(hiz?7cao!TXcYHTRV@?bH4M+_)t zLQtZ+h2S`67FOvu{Xma6<_N>q)vPs!)MXy#YKgx-PLFuO+hApu?*LOgwgl;}6e=R} z4>zc}RAsI%b>fFYDE2VD1fM6kT0c&IOX<=aV_>F|*;!-*HmqxBIrH=kDmLA8yO1E3 zy@bm5hNTZK1+u)Me~-UwowsqCget2%O_M!@-yZgAKf3mq225yaVT2ogFMjJw@+Wf& zqX)a_^NRx~jxr3Ku#;%O&q*WySXvXhu)kTgUqv>wOeuA)g1V_B(cThyd6yD=5tLQ1 z)$QNP6cYQ_#0l@tZ6f0gv4Igd-8{a@suv1z{%K#5{Sk09ku)$o894W8!E);T1pexe zsXa-D9AjYUzW2-yz5=dhVC){wwc8zWrT#kd!`93Y@a}&*2!>fx10)>4X`cw*nJv{| z$)twvt7rSxODD63G?ni^icmPL{bC-{eYBIn>6pg#9&^SGqM8)t6JV55>aC38H(Sq_ zbFshkZdZ|hQOwC>IE2_9QB!xTbx7Tx*CL33dm%f|@D0FMOuX(!eAC!Ppuc%C4@n3A zPVbE9-|nAxCyh}ZKpjfre0>5ufEBBblW`7%&h{JA(;EMWvKY_>3kr;F#E#aI)4HKt zH&Jf(VPOkLz^;JaH4l+v(gM+Kky{{@?Jz2DEDfCbr0sElaz36|WBS*w4a#Jwnfgda z`&3SoMQbJHN*!@_Z=O5Guxv}rxu29ginLn4Aa~97E%>y@?EFqaov2R+lXHL)7f(>x zf+H64mhOg`@DLI6Od2@x;(vTGgj`A)o~}Mj9jTM{@QaAj%=_hDTFtUV&((49>F|-!vb72n;B+F_Y}G z|4q;d-1ODE3H$yS-8`fBqz0&>{7m>2Z%F6|lyQ#fo|%l+^S-K5vmYuFlZ)5nB{TCA z2)r(I=OBlYeHtmW90i=3`F2BA9sKK=<9A=zm;O_#qlvarm5&~l*5?RAV1`kVT`WOI z*eGR<;KPApYe}2_Fh#2&(+<8Jp#izjsloBMEo4z~XR9s@4QQwS|c51Hr zZ$n4!B3lgB5^O7A5{8=T;dv9T_?-n4`KRAI;|MPW5xJ$?95IKNHSn+3M zh@U&r?>VYqA)B`;8{HR1KMRsVeXtdxxeatC!WNv&2R%B$GEz2fF%1vY=d}i(cHE4V zaB3!N*x5lz=MS>7amBB)s-rZh0THlv*p@9(BR*`hKZNnAXtGzZfBtznjkuu zv?PTIxl;-3>VWxxAZhYOlI4xS?8zE99saofct$8f~1B2(FSu!dxgox3)z~2zhsOA2K z#jhSzW{}4`OyRgjbboD~bRif={sjgu^x)f#);7O*$9NfR#S=;Z*a8n86StC?_G!fj zeg*kHkjk55fPFHSCbGU9ZrG7p^zT^-;QR@zjm>5tG--&ff*I8k9H9bQu)rmAFn01E zIFT$1#JnAi=Y-=F7~?Oa`jYdB*5d-7yBcwBP-z6T#8qvoc)oKK%G4$Om7M#e1h#a6)5Q z<9$WlMHAHp+{2uwW9l~#lR)+71qE-b7O3DGeCFJvy--ClA^Vq-*8C1NJ|Z@kY8Ze{ znr6xW_CDHgjgjixb!4W!n(;zgkp%V3Hb{13J0s3F$^~6hRTt+1f&pFT+8&qb0P{~S z!0NI~I+>;d#=I!>cm9nk9b0ZMtv z-5St4b=_0P4jnoE5Ga3J07XO1k-XXZ89#FVZ~<$Xlh4^THY(+VWp#@R*=>V1EP`DfcAL8<;cE%1KJe># zUeR7_BlBXuo@BQ2F;BWi0;Fiz+sxxUauxDHpS6*f;v&PV#5KsW&eP zyEi4rw?CJ+Ik|%Bnu*XRs_ojfFYmZMplU%!s}c&+AwGp}(dC-BM7J?`?A&QfRJ!oA*TNG&Wi=gj~$@gsvsuRlu@M? zS&#bL2|#4@Td?Uk{30FUhjBpM1UT!~ZrVIp(BUO; z?7eUKC;?B$QInmHR(~F2tvgS@uY4zT2_9u2^hkQ@K8Nj_6I}0}PM-jBh-A#-gdFe% zKv7dOI~?o8rv+9Y&nFZ_1iQaW@bdao(GwaLmTly?@e6 z_qy8N*6=-0m=FL#5L04#U(_}N7~p=UfsSvuJJ>HvcO9nH)kIr3%RS-J`&EW|q$s$u z0RWCa)MLGJB=9w#gontlgdK2m6Z~1Jz3kj{J1-*Ox~AksM&Q-CL`PREyn5Sux*L>! zo)zg#nbWcRI2M5f9OOH=!B7n4GS(38GV+`~!Hi3ROk#Hc7~C5>ObKcfTm2WkiaJUq zgf0c|dZE?jyElG(-UreVFwBE$2=Yil?ew+75dGh4GOQBJqtiFNf;ohOh*}kPTGF9> zpj-mq60x^^N4~9=kfmbKf{%l1$(<$B1WE#y3h;DP_FF-2Bv3vLVSqWfe}%uPkopi6 zaU5_x5n*H?aO~3qNz@eq5p0B}X|D!gS-m0!+qZz}i4vY#mq!274xovpBo5e~c9-J0 z3bZbmg25Z0VkyLIMk+giJ$Tq7geA}K|3dyQ$4u$nmT{->q)c4ob!Ab+Bny1t z2^0@;<;*%yQXuYXo#Q6D$wQT~EpGOw!e2b1J_CC4Odp2P$ckzP%ARDg3t#Ev<3;f& zUd-C(uMEb5E*hP9rR#u~E!uKVbI$=Lc+-X!k_C{3_*rj$n# zTj!913bd~oSu-9j;$nP`bZm~R6?M(Rg>QV|R2DC#(WYC6a=f{{iG2|`T6b9Cnu|ob zb2W3^v4N8^#YEi>8@dcCEs~ zeBvSD<&b$jA}^AjRkNe!Ee6kHC+kZL!a$C_Ij2ug7xTe&m|A9wKG_ES@Qvj|@Tg*z zb_-qNKo0Z7f+j-X%(*_oU&gm8HgEeA@eOxcz2=~;6n^#m-;Ui$QWr_5D}VZ<)<0)f zo};4Eew!RRF540%t7QLyC;aQlOM>UpomG^R;ukr=S464V{JgATHPIDf-Dve!;JN{l z-SIQqw(7TMkrLZ(9Z7_Kb(i~oFUbCC`>QH0Hik^V!~~qRGM}Pf!c*&!5-@RFdC-xw z{msy!zKk)788*kd!=S~~_j(I5+_fsm#u$76YMQ)vU6)1XV|gZi3A&vH#!$hu6U!|HU8T3?h! zr#Ty?Z*c&8{|S?V{ZqnK4sR4Jo|Fi~A?$>5ukp0_ZnFo3--$$HCos)}mBHSxAJ`ebfA8(d zzw4V{4Vz;)gKRt60A5|)r}XYb@Uw>F1leU96xNS^mhya>`zmrbrbJ|=QgK|LD~6iC zE%ufvo26PEuJkIBopp_~w`WADrtzNn+IOjW5RjeM(Q<<73;CLp-yj z!|LcGzU&K(r)dBVC{DvTL0-_<9B|YOB~6wvLb;6Ze6!zCyiQNkcNE-vjWQIxyNc=)wjAA@GuX`d*eLd!jo z9rt@auj+JvqIF+QFg=nN5FB~Vn3an9S3$1lety5GqIjzdVKiBfn(vXrFn#MVZoDjZJ2*1by;?mv|E3+JzYrfTfHylWHAZz9)Cuo}~0sw(a8ofX0a< z3|oS$Z&X5S6YxuX9GZ5u9ZU))qh&DeE8fgv+d3+1;;yNLew@Ja)3!-e$x%yK&Umb? zr{}WEMy;p5iOKSQ@n_mD7F|B9__eq1Tf=TMN+eT6C4O530lj4cEVyHCCZCf!Pl8(l z3?J`(OP71PaBp->w`GLeI{(rcaZbdS$94IoH9Cn)8A0a(PhE_EHziEbuY1~@$$)8C zzOF^iwvk+A;)Tt4g5{XSqriCC^nAHDJ1m1Ddtb zh376F5w?|LWnuNl8QfTStG?7(<}0_2A=UXzt>L*mmNlgp@KG3R^oaQ>Y~1mTy!-1; z4siiS)P#Ed&a0kkBz*V0MCV~HYhUe99zA7s*od4`WLWl2yYFd^PBmJcs9f0b2Mk4F zYT45*pcYUBy}$(g@u#0zNzzzn`*}tXQTJN8E z`?fLra5UHJy?y((eCsh7`0TyHEmhE9HKS4Vpa7`8|I*@}wfT&5dC0UmiB#bSWk296 z!B{vnkfQfhL_Z}9N3%1zeo)-EkYdK!V(9;(?t#y(?+2Gj5NUc31@;tsh`8A1=&bc6 zCZw9&?f?4Z74u1ee<1l_W)=-&iiS?OnLY$(L^!}Tdb(o(mCZ7GLC~FNth~!h#2p`O zUgbJ-)%fb)d;ojBIt)bG0-`_;(Y<`|&Es!;~Vn{e`wX8&%vl z!%u-Hha@U>`DR*Ql@zi8+gg2jP|D#Q;D23O(8j6fCMa|JpQCzvexA6mJz-KCNN72l zEnImRew~Rkvwa^vuq%f>>{g8z8FolKHGag?vz8&`lA0=Tp#wY;(gs?KD)ZtJi;Bb< zXQ!0=%0$Zgnz;ln^{)(jx24)m3nA;Ly1#uAHvL1h={?d*IAr!Oo%+#i_}fV_iKIDww$>()cY)0 zKY;fFERKYyfoohoCL@%XL60{T+(f>ET@{O?qrw|u@3>~4m_ng(x}!(xT2U6ovbtMi zd9Tub!1`lp4IOYn_+8FprcJo;hYPrAhF-htSD33p(I68117O5;>>*2X8zK)8^l| zO?L9mjmqVPZ4!(wuJ)scl%sz{;Z*S5m)|Bux4ur^v&E*uC2xLPRJ}pp>HG!6JDWD# z*y}kFS_U`s+Lz6(e(EghxF=hNYuCi>@t&v`hp1Ob`;8*E7DLtS@YT#NdV}9T%}~HE zAm2Fh&J$n>;lfdo{$pJZ3);y|cA8f{I`s+}-~K0!L|)2Ovp~beiL)eavGkZRrEq^v z(FLBX=vSH*&y(HEV;?0lqg0S2w0TN$!GhdK*63 zPH80H?i{n9{7MXqg1-)8WX{E;V!z4>wFLgj1MYz@Ti>BsrQFsJYp>wRKk(Z`<%`>> zzJEx?)TcxxK)AjRV*z?*?a*zOlb%C|utr6ELRa!x+(6qsV(RJDJvg6aPIHt`oo-tv zZB8Qp!RKU(lZyPp>9>@|V(NNdF<> zC3wi)e&|XEPv+O$q*XiNDeA!+8utnTc`M0ZACbgkTnL>pN>jf{f6A!e$@@K5nBcdi zw;@w5M$DkU0F2QrJoJ>mtwhM>8%vnTUDw-Y%5CJT=T{#&>1F)D)RI?XvgfqE4(@H+ zH?a2NO&w~PQRj1g*@R$bfh?A{uYeYAs|C!i&P2^Zi1qA5t*lJ5^afkU&|F^fVH)gP zPQOMJ=sfuE}^{=Bt&YZH`Ue^}JCt_LMktHs-?QVaF9y*~o!%IK;2)L^u4+U+m z2SJULjh`5Q^9z#)`xqDwH`)-i24#_e#3xD*dcFB8T{$f&pC4c>S@+HlU1ITi$eDiW zy(zN)GCGk2ey_EQF&qQ10esY70ktiXAp`e{%(esgB4-QQ}dD!S#3YyN-rjin|hKS@ zIRI}2N}n5;FWqY&W??ohCXV1y8QUpoZBp_4n%H5{i7(Wr+dE%V8^V)$23VQJ zVmMF6N8;Cw`j%!nmjjU36^ayN;$$fM)?E9XI-8b3j)3(Bu~op#&r|M|Z)&kG2~YKU zHQVwWjZb%p0qmo&+mb$R>l+f4zoGj+r|Tm4b4VN9J6Wpup4)#Jzg?uUTC^%ExRIVV z`5ItZEJlHm^2Gts)?vpa6>+&3_|IW`B2*P%tMTz;lFpg#&Sp(s=aA+0ZC80T5M^B;dP!`3isXlHVxkkmaA!=kgKzD6V`PIh{Z#)8j zJ8&yW2-0H(guKgeja8v_6F4)287 zFf}zbGyLnp2cw`(ldW4LX?rdKjj2o%?O;pPQ7eV-V+2VCtWy;FAGU<{#nTKOjv2_W z0is);Le%!~BFqDe2)NCJL|3e7dqoA2I$)H%dF&kFH z8sE7rt*Uu^j}CtEI^uPJ)@@v<0p~||yG{;{xn?vV*h|MV_dh*X{))n7X4MkIw(SsM z#Cc8v!})*8PLD0$5YWheM(XAKVl73!`^a(jlC5JZft@h_f$;+mK0uzCl|2z4y%o!N zc>A6E5@Urc1oQy~)wl ztl*SsXuC4%jN8h33c2DWyKey5)b^@j(Q>n{{ZME^c3O89kGl` z7HWfr+X_88UFYC=mc8!`MWJrX$~9A$&6ob9na44r+rd6po|v`48ZmCo3^$(?VpZU_ zD7-ITc|_IDu4y#0<*H&WM^wXE(B{G^vZ-e7X5;VD%1q7)?^04o7ue&(QjN1Td1peF z#P#dQ=ce)uIbX0~in@yFne≪;&v!c4<%?OmRvUv;voN6)BAN-$sDC!lktZYWn&_cGyEjqrmvpnFbt>CL^5v@_&WSB~7a;#-Z?oH%6 zT5Wl;7S*AJz6`bovWw;Z2d6CUcOyM#?hePelikFvNcLEt>!uV13UB9Q3E2KDbw8on&!9#PVKDfsB1w z)ob3e@hrIn%5cm7vG-O%bwuC#;KAMP;3PPN1Puqb0Kwfo1cv~@-6d!eB!uA3!GgOS z9D;jrcXwx+{O*1E->IoP^E|r>iia+`d-vK)zx6G3yPtabSpcVR&mKD50~TP<#_R99 z%XU|7hNsV3y?E61N8RJbir&d_)Z-ntjVxx?i~(qp+m<>V%`U}C=I>;21au+?ozI;yEy>Mpwr7!x+^_bTma=4oq$Cj)**@QbN;=eEHU-;M5?)G-@ z$L`pXQ2|LJH8$QZ)KeNr6>XC@4GJUa!k?lXm7+BH0u;fS0tb)F^;65EGYuA;3|-0-6jFfwhgZVRF2r7fQ0%UAnd)m zF%EA_^#nbs8|ad!tlvz;jUm^GvjR*tevw8;#eQ!f-q$#+cR&IQ{TCe`;|Agw-Gn zknYxxPBYw2Ct4SiAx5Kra9T@6mL5pN5iPGaUmkec4C8Gv{HrI#so!0dQ@Fj#?9xpUOxm z(yNz?%FtqVyOnu=?FR9{gOaELX+hLE&@^ye;7;|M{3zr#i~;zOM$h?_meTrC9TB0Y zR-Oweb26tz7QQMzVp~)4?ypJf{l8a8k7i0wD_Opj2R%3hBEENV%9B2{n5aldFo%rS zU~KS`M9Wu4otgs#|D9*W+8XZGz4NPfaB3g$Zn0vUGcRzOaw@W9FU)}0BCIQqrnDs!}}r`r*NIGmYeYYSwLHajLXsXj`o1BIHV<<7?uttWg+^$v{4>Fwn)G zlat=r&+!-f9$H^7^H{4F%m38r_r!PZ12DZU0?0KGKeOKe&BA%_O9PiJqRa!#GNDs9 z9^R_nS5Z_~We<->nI`gvp;45C=mayXBek}ku95eHLVD4@W79oh5zjc%v*lD8P;g(J z#M44!>BM}xVX2nZg8^VkiN-a^aqJx3?TsvhYapF}CG;q|5rYu3&D;IQoq`vwx&~uI zpleq-D6O1X)1fF*2ma89$;{uyH9+oex%{?mrH8ut0?3cIl|R3_3}*zZq3@a&&)RKd z)k)!?ZtXt(!9tCa`MTYu)n4SBM29gVQ#K%YmXRGuv^%T*hS|<;9_*+E57H22pWm81 z-KeQf>A$`F@si8U083{I2(G^xAtgb{n?V6aZr6L4e9fgB3!nQ-s-HhOIPUhIn(@d& z108PwKDq25x0GSQC(H|t4hG_2WY1VA0#9a5K;~t@q!c+!if7_2v=k~lp*g}tXDNS+ z<+U_N4F$Xr(92k0gI`*zJDl1^bzC4k<}n_5Yd*m63@D=5u(zOXQ6s5rl`; zRPvK04oCVy*}p{Xelk|^XVlDRGq_dm+C*h!yr}8|%)~i1U`t1kqRbwC`1t9zqTxJh z0PiX%Ail=c-5)&8h%-sLg+0Erm6n%-xrJ7C@>oQ@-RyUBon}lJN#S;4`NCVO5hm)e zYD9pEA_x<0o8dP`MC>lV^{?Lz0&ENb0R(LLyag-4l*sYtZIA9{=TzUJ5#tPp}Ds$kgoXcba3Z0)%l?+3!yW3$(lx!a&Q~tKV1S zlfKhfSU2IT5d_h`;7)lug}k_ac>37%G#i##@;QBLi9-1Thm1((;FB){95P@yMPqGF zxhY2n#k#%Y1*SnG8&R{|9n}g2>HnPzV8oj2tC(ezegEZ2n&vAmItG4x%FmqAkG7Hg ziLOcxwl)O*z0Vm_z}GM2AW0*&7Qp;xD?kl;$D!|y{XQN=IKOiZVF=vS+tV2PEb38? z&~aNgmN>Tjw3_{+PU@B)DBwYb*3n!p+4+gDlkZP!uO-+p;hZUdb>z*wYurD<=3708 zkc+@t^r;OE6`CS$Ah35rOc%gT=ZXm%CXZ#yjIBj9GEKRgTeG(w=hBP46-rLK?7Nc^ z{WhW5KjsCvzJMGx=tx(p`_NZD#F+>PNL}`6N`AL@D!R9?rE+=f2u+2umW;XtkuWcg z-3ZkWYZR+V*o>=e33yRv>jAt$xiP&3mJ=k|-kB)0mGHAY%&btuD;wES_<@B!>h>YKXs>WV1M+m+*! zxt?Ug9Sc(gnyOFR5NrxHDS=K9~}vZXbRKUQdbS=doz8j7;WkS>_u z0u6^wd=2z=(Ub;XAmALbsC6yMCL9zh>!^fad#giZ-yRCD^i-j?ZJA*zj@t|0YvddG zxs5hkA3|;w?+1}0&A#m3&zY|A+x659u_5uYFtAi0$u^FcUC#RqoQY6*Mz)uiuD&^~ zy)%o(4&{cG716VR7vJe|kMgL*SW4%YnHIy}HSLO4gPh+0vIA;(FRd5WlnZww>+N|H zf(2#1ED!_}QlKLvP15sN+gp%w;NlenKD2ByPC?^Dxno@4j)5(mLqqXNIJwK5Qsev9L{!=gRq?zVKBBsqjk(31Tp;;d z&3O=Ag?u0fPuWvCXZ%I@f;pV0D+^B|EbM|~Na$&vKhaulD!l1fm}>jz+4pyfqEB3W zo31;nw{CNVe3Z;x#4%EUR$AH7FqGJE72I-k#>-uVJyGC3=S$RTgimo-{l>kRZ=HBV ztCMf-v3zOiQpd-CnZte^w1N3YGmuqS9dQH|InByb^WcOM)W2@LI^|Q^ ze8_V#{OPs|n)2#f(>9*)F4lXN0^Kg(TO5>1PBx$B-1IZ;fPg|PeJSG>=U4F1871m= zGaI#~>5Tx(qA%o?W$}hhc-jxZ-o@+xQ2negi$xhU?DRe1Ejy3;{m}`Oal+hAE1peu z^m2O{;;k>Z9%iC2pSE;%&oQi*?T$f;W@O+V0Ay%2V}F!Rl~1V;{P{^fH5}Q4QzE=+ zU1+~KByiaC5xWl;ke*se=v92Wh`5ZsxQ$lj%}PpFI2MT)vw|m;z2SlHJy4=VAH|)AL;#8 zTvji;k&kZEW?PTe7IRq|0jN@BF$kLOUtnIdv>HVx|g{E?)SJq>Gr_Oj3FwnP7gu%%Ypln8wBQ;fu!3FPA3)cmjN9kG(>q^Ft1gLZl8;NgI!p)p=`O=hIJ!?xySz{@tww>)#-OzuV?c40j%=%f8PMF?ssd-QKBIkO& z?ziF6UpAwM3x1Ds4}J})>+_9?H>~)jf?|WAX?55mMu3EDJH{pok~Psft{VttZ;I9x z;e3D9(-LVsA%q%s2AMvo3FX}xuCfR|{0wHr#ean+TgAX4%wUK>QEy_6 zQP$_g5E{DDCLLC_Rwi)qXhXf*F)i}FEQDoudm$?;D?Vif6B&V7c2EHg39+!g5M~mL zR2Y_S(MU+?R7bOrYSb$I9M=akE|TVx^6I|=`Pct@G@Y1YQ({DJO`2c=Dhm59QHcz( z74{l7?{4j>ZA=4R*~_(JI!M~+kV>WrF~V>!TJ)dNU&Nk|v2567z6DuVmuTDAjxvs) z^B97~1KhQmQmnqn_Pw!n(;mfDm^RJW8BjGc2YC%TMoNBO=_ z5si{S9}V#1o5_KcLY(G&1;EhS7ik4+#I2R+z0Q18jp>YI-Hn8Jvesho{o?q{-6+dU?;K1H$#jTg~iS|V!zxc zza;%+Ymjp4W)*4x!=q*epiR#lii~U>uKN$28jma?9ab(ZF#NLasXW>ty)3p&MNA!z zpQ#NoLbJrPoB@k{N%5p>fk)oV=kwZMjDu0B?dK^44@ZHdZ6IF?hlh6h{D^c5k+7i{0u=ac_ z5-UgG8Q~DEFJEz?7hU?El$?P!K8y~SXK&LVu1FHWgTuC?X>1b;;W>#|B@v3YYIH8w zFGp1XpZOdKU{(ChB)PLUyVD=$B<}B(!KxRK6HnmnkGG zwT&vn*XQ=ML*!Is2LS-A6xF6~4%iI;I3hOC1Mse;LeY}ah$x{f_6hI@a3LM26fqkg zYhHsONN8`ZGSpP2z(8d2j*Jz^8#5h_DG$7j6^Lo)F+FoniS0vE?Af^EPIyRFw!F;zJEW6kac|jTKX7FP8LQ@O~pB+gS~{a zt$iN>fB4)2_6{i0j3{A+4+^mevd4Skgb-0-Mo|`g`{E*7naezwmZz>4@%>ft%eL~H zhZ276Llr9>cBX=YaoBQsQ-$KA=pA0!8M4vz65V?JMEnOeb;ZDVy!>rAMakSAN^F#N zT*|>Lao~t~0DDKRHT(HJHIGlRl@&?wg`-qJJ5JSE(lka-+YDSLoVQ^~zD^hr1vR~P z=hh^4S2vi;QCnG<&%?I=cm%z3b5&BtGyS4s?%Pm-&Y2*@E%CRT;~b)9g`jq^E-i3z z^&Nk*CPmjS5Nw(YNV#Y*phA4UmJR0PkPDAph8{pqHNu8AdNzw_gj>g_dtKHk&b$lD z)DI{RB4Sn#>Dx_H4pQ28cYTB@0~kcT+8^3Bq%msGG@r$iPNdo5`l)qVpB zaI67RUQz0^8n4R5&#k#rm$Zz^)Nj4PfKB|Q^h2nd3jHOvh?1HjyK--Z1wKWF8gxMc zmXV=*htp7fBpQvKFkRSu=A@~{u1SC?NXrW*To379E)U{%pAhY0NVyN@eCHmm}sP=p?8d_K|QFy!i z(_*(`TygJwFLM5qo$y3`f@00`G#~7PhBv!qI0hAgt!5Y5aFAGs2Lg(9_#b*8>Z}Ip zAHvO5lEk-VA58g0tm+{L?aSmR5Hl1B1RKI7YzJ=uGEK?kQgkh9$HX+_ z)>abQ{Wy6$jAt>Op@P8eK9J!TjD#SK`E_#2=gIZe=xD!;b7=S0&8)7*)CgG9!5A`W zho9c(5~{VAGkUalZu08`vm|H|}4%2PiD5RxqpJ4Z)`W7_Lw?fEp~0|D*q<~tj8Rb6sjgOR09#vW$CRaQvxNbHs; z&jp}J1-z9#+T0L6ofX-Nxec_%QTy{$-LUxi&Q8GmZsU}(?Vo_>sM=5V4-Gu4-Ho;# zJ-t&nS635=Ns6MbgFAYbolJ{@%TIwA<(>dIi9KLejL4({&(Aai#9RFd$ z9}|cjfw~76RME*u7eZCgHuGo|*E}_s`qw@MX&Mew#?j4|@k43Cml%6Q=VnhHiylwz z`cpGtXpuv~%5fs<+D&FC@+5b9-TPc21_sCp_K~8m1eg^^k#5u+XYXQ??de`0nw%`L z<{p)d(Fc%A5>a$>&_hySzq^RYi?~$B<0<5Xkb&k3fXV&*D*_ve#zd8th>Z;7bQ6$$HmSAKCZn7{3Geh*!4 zJ_8F2i)?xndCtO`fKdmnDpVG#abuO)GZDbNYaf*r)(uW@bk_cy(W00TJ;*F0-u@i7We0W=K@R+nbNP~nAL#YbDf?D z-t7bHDcN$UM4(ZXjB>ynul0*3fqIcA7eG)$PHD-PVdo;CKCgNoD;^Rm6z#&~44V$GS3&FY~4S>OPCmViErhdrE9f+da_(*Rh z%1+mdO@*>qM@Zszzpt0X!TRI2_$W%QYsU2ViHnKi!|hHH@}w;`_{B zpQvnIy(I}Fq5@0NMWGtn{1}p_$Sm(OjsGZvfBZ96xNt=8=n#vt(MuzsJj?jFZDRTT zO%L8C5p|O(l^?#_;L}5j)5P*s5eFn0-golGJF@{Jy4$qu&G~&Ke5kgKyS=Gy;Is`| z6$`%VhsM0+EJkprySRz`1Bc&3?GFQDP2fv+I%~hm6i-Ie)U%wEwAOc*PqF{HpJO_Jv)=?~iu{cWR^TjQe*#c}&lm`l&;=F<9J zUOpcABO|MyaWXQNGg54!tm20cfp6Zv7HCu}%OeqZPUgxmhTlPV#clai|t+)p@{Xn4!F0IMhX9N*&!xg@!&v-VrnLlwLmtD{2^GZ zyR3)QB&BZvIRtw3ul5126C z=59zLv#ssOUQNr10U~Umf=l7SoOyr(w2u3iejxlHAE~U@wYxbqEdcoz&tCfqT*5#L zOX>{kn4a|XHP9&O)D$T=j7fb%xz)VMIT=`!LoU)vS4RBP~>df zk9L2)*g$f+LrE5IXK5=n+GkvGu}$^f({+Z_vvpnHE}Yc1QeRC+)4{N^`?naH`(>H6uWB$K9e z8E~H$H=E&Oa?ATj(9sv43e+1TSth557gZ02B4R5aoHWGP*}>d2^b`De>;(BLU)dq& zNM;{clF20W7@{W8C)vXE_@@xlwYYhig@g@PP^%=(H^jt~08nfK(zmac7;mbM{a{-* z#i7TwJ}2pes036g!V;-wh@#||M;IY^HA-57+7b(FAF#{3QH4bAXRK4WsPsOxV@rcF zS^+L={yu_Z3L*-yY!kgn2ujfEX^eZC)iCO$I1?)NveAQ&nu?`dJ7br|F1nOg=e@Bq zdoPnUq%C&zjfXbGJ_peC4aW(HjP-G9 z3V_Aht(af4QSi2mQ(F*T3Thkq)m;EnBOJ8FVZrp*GgKX#SfEWnuR4!q)QM#_Ag0&0 zdbAEXDyUlMWS1nTe2!WDu*%I{D*9#W9h}_(Wu|LHV+4yt5r(X+(C&8lJbI9Jn1%;} zs<;Vkxk(+X-z8)SZrd|IHz(K{85}2{?E`h(YObb*1iEb_!2F);P9+C+nL`8@!s)vY zk0Y)Tp?-oQO@lAz`<#Q8^53&ZFox2OQ2qQ(H^2JzQB8m4L?=HohuzPJ7P2kRs+cVh zj>n*#6`g=H8fGJgNKHPBg^;S~<2C?xNmL1U@=KLg;iovJcDS+chW&;6=SeL6)cRBT{&-er) z2g;78l+^egm~7rc8Z?~Za}rTsd?5Vfx?nqTr3n^p2L7g{+L^T}1AAM+-6j>TmwhHU zNNC;G=-I4a@H)NUMtyoAHM5M_(v6xkQv;XrOp~W5Tbqw^1fB`9SBz%?<8pihJ;tL5w*KjP8IOJX&x=M@9tHQCfQ z!|Ed(nU)_I#EKpU|GAA$z>EgW=7HT!)%%>IR)BCLbzRaH28Wp?iH%HU}z z#-^sGPJZp;LOz=Sc0nOCC()3|uzBQY@5)~?w@rU#W8?DRD=7j{!yk(S{KmNQ=B&1j zTj%e8iLJj`Arb^ATaMnKwUo~bQ+$I~;~w@S813~=j;?b~yc(lBx_^?iqvB)bsU0~erb|0U znEC3O6;7gvN<35C%d$v}N!`*8!zGM8ae)^Z-um>8r;XH*gf5%BD`P!p1RZgWfAMl^ zWcP~$-v>|gE(xXbF!HhUEITq$zph4Q5Ne#tvBOynelt16H}mRuHoY0#sT|xA`9vRC zcv+jw8iC3gSHxwA9%eaLl+RtyY@K%1;^=<8= zGn8SEjmtFM{Pgz<&2}c!g5mPo?KMct>%gtr9Kl^UwdBChA&!e&xcZ5NMW94Sm^t7& zpDP~(2Gfx+qQDVC>*fsBk$(YPy22n@;^unZ#o#5yn>+~Cj|vJ|qWZtzt^+af2o zkPrNh8$|+VDL>#{ouV9?y%(y#-}9)23}SmxJvWMxyplUw)%%eG19s1Mr3}&JIpe3} zL_}Rwe_D)ZIs-?5%*1QHFJCg@GPa(v3Dd&NrRYe=b8d z|F}-r5rAyWLyXEj@bw>HBgX00^Mm{vfEJFX=26~4ln6?BRr3tQYV!ul3+EV8Cmod3 zl;|$<*Vl1sZ+tCnCBVYm2601V#iJJMObiUVz@`2=!0xkGSM>8YH=DIC`n0}N81zkr zPQ<$_2-W6I3gITrA_h4RY{~=Nl?T*lo z06OVEZMsQ;!%JkC7_!wacW^*RuBvxeXeY;x*()=~w}z~Cj|*J)08BLU7yeUVdOXUw zF(dYaMfV z!hiJ&r>3waXWLRfMuXFGxkAmYMnFG+95SMFm;aQGEj{yH;nJIU}W!5i4(+g*!pr%H~OYWu7F}o0t zA#JD`igaEmiSqHYENu@yBQ#69$VVPkYa`^K)nz$vuj@1o?TcOxGP!EALkuiZ5;Ilu zx0u-I#enNUw?I)kBb`H&XkS1%TO!lcZrel45%LQ5n?}x5dfobZ1e0%H^ZA2pdjanl zcZxErt}$M)N!)5JG&tMvAo`<(hJO2wzs_n6#9Zup%IF+0!&wFV6`Z-TzTqS*PsnjLX9x!XH zt*kR%keKN288}It&4td%eR`hKv4FP<4AulLHv}%nv9Z1^oEpbRN3>oTgkFMsx<~kG z^ZBYCGHR2&FPHrgk%22gw?JoV#xLN|fRB7chFGo&z9ntv)3K!O6gZcmhna_2)2i=8RwhOR%B87e^+me1uEZcMXj4ZF z?wzWFRSB6jSqwx=LzO_+0};d?3;rd!gP?|#arW#^?qYSp#}QD-y7&;|clwZlJDI;} z{_FK@0GN74rrlqVQhrR5Q2SK&NP(R?z8oG&uRgb@yz04@_X*X`dO{ZZndV3 zyrn<@8w3BfBk|wB4m=7e@Uf2Q*K;fFW!Z>lk`vImCh+ALB_$y zGY7V{crBn z`w|eA3NNQ`uFch3;Z6kRF&a}}<$-&gTL4()f9wC{X8g~)0sv`HFy=39+~<}d2tMoo zw+;CJxBY+jv2_5TGBC1HOV_AuH<5Ufo7cvFywIeu5Pk%xL;wTbt^dn&q6F5dyARAb zLuV5Bl~*r0xwxtPUfCuX2{yWXMt#D;r6>R(Q&FFjvohT{c@fPCfSijC ^vy5x32 zPdtc95s0eC0Mz6&N(op7rHe;#jLT}NA{n}^T37-8^KfPJEz38{WoBL7Rx4Ji>7&v} zJ$;BM$X7DH6eHvnjRB@j_e#nx_r ztPe`J*9B7L&zci>NMmuS>40v3kqMU!7RdZ!#|zj>4B{a4g34>hSIXO4dqqsv_9MeG zTYzQ=EBzE6rVxKE(u*l;v=Ag_Z0T+AAQXY-)jxgVkH&rNeVZO_k0=3uq*8*`q4IZv zaB6R@cxt5c{v*Rv`W6~8Twi(WQjBphC_My4>Z>s#^=oO{V=xJS9=VB_t?hPGNBaZ{ zgyh5%Dh@$kHdZuWqsut;FbRgiPCWDRiS9->^-9HH~m^V)XtOX6Wo z`4gJBPHitB4vAWRnQ}}siHft^3K4WbDzhraeIFei$9m1thu!#rO|wgfEuwlh@P(+m z9!&)S0aAw7f4&=@=FwS*H3nD@1PvlM>CcnW;6_3STy~;mva~zeo^a_8qW(dYo-xF! z33~TZ5Xi}3JJ$0*c$(h%d>`*&agSa}>SG;+#rI1uL4J*D zLnG}nC83b1pq&#(*+`+LK$;&1z_c^~Jq)ae@8!7}74M?B>gU^#6&0 zJ#}m2(yLXvS;N3nLzb9E8FH%(XX zkD4Gb8)3&c%BLk^HI=c7B3RW&!vW2_ zlUhl=Ah)~W?Qn(D#BPS^+G%wSzIxY>v{`~ zDkvb}WPXQz%QY})G3a(2HYuGtTw?(oicTqN0XpkX7-d8o<(gCzC_sBfFwiOlG>BpS zvsevMIub$HND_gk%S<-5s8qk4ncRn%lxK8SK2e8HP~JOeQFZE+H0%Xp#Q&IZtHA~9 zAzZ#aQuDDnev)Y;5PaKDAZR|zd0x$4g9hBH@j9zks7c1pcA&h2$S7%ArizM^hNn72 zs{)7$HA;8~W)bjaYK&_;gKNOSMfrgUMvZUO8QYc)wep0r$Emdl z3MP$M3?aKsLbG`eom~_Hf|h6kf{;s>s2H@B-%>i_V(IMW+ zfC|^%ZC+enui}Al9Nil{ymz54utT$lx3YR`KdTEEan zW`~6x34c%`{g7NX$mk`%_0&qI_dV!^c-%9bCj5(56OZ92CC7i+B@iSJC=akCCyVQ0 z_LB)OSmSyxuH$S+RdGMu%NgySS97k>R{FH9VF3@?i;YYo`ALZ#@plM1p?SzI4qa#s z_eVyC%ZSxrAUW2+s9bityx`y`Et;L0&HqpyV;RQMtRbDsdsyX$_!No$I8EAvg)^Hx zNp=5*|Jo@3Ngnp4x%OcDfT=qoe62(tvgs0-u9xw!y&5V7*X__ARD|p+2U2`)`~YSS zYL3t2qa$HYO@{@-Wjm~7hTgtu>aY(+TVX%ko%@irVQ({C2*rHl<7%IYISx9pVBHxf zLT&!SQU9d2Ljc8bC)~EY>+#U5VkV1^pbpUNlvrBY!m}*vl6%@rbh~Wd zI1%?5os(x(uH!!qTEILV`SKouS!-kYMY%olDu;PUm+NO4UwP-M3VeJCYA+~Id+9*O zsUQd<80QSBd9w~72r>~DFF+PE7jLRqBS;`4Df$>DVNh7-kv{M1=D<0RYb`m0+mvlM zw>wqsF&%t4?G~ycK?D2lJ|N>)z+?<=r2QV_lp4Cc@T4AsN2UEC*VMVm*6PTcd|)KK zCz;OTc-Z-R8__K4JrrJYk`gr6PmED-#!muq?DZnt3Ob?Lg{PQD5B#`O&Y4NGzc|)}Zh6*^2J)|<*qX6_8(S7y2 z@%yi4!AR-INHY1#VHpL5@4bFSPR`F3CteZ*K&&if;P!$(=nq-59qp;Q^V} z3wI>Aj>e&rN5RF*p)xQBW~7J$UrdT|0Uk9bKL3&!Kzs%uxt`NoEUsS88JUJRGBRXf zic}stJ{@~X!2&fRAeBx_kre-{2u?BoG?11NjK7W;6tz}(F^zn5%v(@etyq6ak^5J` zsQ2jW)z_LyJCWU{GTNq%hfF*8NtMyu|IP*U7~7hnO0S++V*<(~XKsvJ5OTsBfkvi> z5JiM=$<#R3aXl+|dB%7#&^%nk=(FY5lKC^T6<~Soi%zuD|CkT4Lz&y~J6hLGhBto( zzk^0(7N;~1%FZlX^A+y`dwg76GJDeNKXiuwZVw8OQIr&~W5r3q4kK1dac`e)(zvTx zEb=l^u#+~tYp=op&PjQ(x?VC& zedCaQ9PTAvxAlc7xr6>iZT1%oFREnQ>I}2!UpULxFtGkf zqd!-yHv7b*&I@g65WC>P%3MS7yVC|WkUW2%?W?GoaB3|ro3fq~d|GzYP!JE>ioz$X zE|X9c#YTg|@)&|g!`ET7>ct}kYF|kOH-XH1R9)R#sYAh~#@^R5c^9~BS}~vFA#a}o zF4%g!6Zc!ID1KIs&6#RAz@c1mps?5ws|!I&O*DP-XlHXy+UT7C5VV91>9v3JHyg}( zJai{MQYIT(C^EHkC4g~H~rX2d1!7N`9%dHZ1sAKoQFJ? z{|Gyn#FkE%c07UR(w!Bj{naZlH71ws=v#WdIB`Aj_ma|){s2O~h!G#*U(37`Z$lZL z6w)@8X-4#)tY`qMeZF(HsYFfQaSZ|a{1a~W=>9wgN^470#cn-k6W=YpX|s0@e_P9& zR|d|b6k8t!vU|2~(Fg$iWx)&FCWJXx_gqnrJ+TAmNih0p>K=do__UKwOfyjygL1Y5 zh5d%u!WN)(RIl^r(x*GP8%MBT6-$R*#5cwD~sF!l3HZWa*B1cH;@PL-_5} z>Py9|S69@AaK^##JPPSlq|ZI{`xD) zeQZT6zKd1MJ8X=g+D3Sb zh8pLMbOG*RO1wkW0V<>>UdU?b;l8gi27lY#{dYRG{~w?z0Rp{d<`OY#{rQ`QzKfJh z-NmGpQJ_+GC18h3nbcgxf(-l?3q!5)w}WJ8{7%n4>FMdQ^W)7TXU{k0E#G^V5EN;? z@pO$YLN>x(Kf|;@E&JH|g;4K@J4Z8{d}e-QZf{`?pg1?(=8omoYq= zC(1i9lGR4WHQo}J{o{?IH&zNy_mz1o40NbYF&ehSzBKH|XODYq@qApZ>^*1UWjY&@ ztzorJ#U>uKn3a)Eg!&;NUO7td>Q}_9O=J_=gSgsf3tuxczvlZrEV5Jf&0oh45nK;-aVVF$z6%t*LJjIqQW2?{D&)#1pOVd&Q(IWA*bTJWH#{V@~h; zlEqphMi!37)o+0Ti*RAlZ{&X_l>$9WoN`yyFPy6T#zNTs)X8Gxl^q@oAvz8j& zeoid+-c?3$X&9A5i=nnZ3SLvi4A6ZHulB){<)RSl7m)T%S3qU2INEN_L!dL`gAV}s zhKrxsGC{A(dDs*hD=MmJg(L(jIIFm5#{mKqCso#k*~@cTztp(w7$4X9KO%C5iyo)v zuj9kq0t5wR*H?6)DL(o4Hyn#;X+*wQ_Bi$yPOu<2@mr;s$}th@8SCR9-$bJC>eXl{ zS_U&3aM1{a+e8I4WgsMXe(WRB5O=K>Pnd=~q9;?qk7@!0*~|Dua;l^+I3h}Sk|6)x ze~$(okQZi;#^lY*b5OysjNB1`DI0ZZ znR3~0RKkY?tWX#b*%i=h^M6qDcZMLca_6MUHpOimU27jnuwZ6M2u{)ze1-+!Z4l(L z$toL|T%P}Gejp7)hBHSO)vc*S5uIAfLs&Oor@D(wQC6y$0x*Rlq~ z7N#?$y(aBdZkjh;KjaYsCnowWe|zA5pg3jOeAY|)txXnwcICKBd6f^3i?qUdKG&pv zKo9I*L4imI0vG;C{I=zmYv|jUKe!?ygL1cqtei0v3rV`$s4P%+1@?Ac@xy*@gaOA_ zhfZ9lDbiA(H%$`Ybo%_{(cTM>FtPJ+C;%w+VsE#o43St^T~aRI zoQ;<6_YOFRqTmVRZG@s&#Yp;9`nPt|y4sDYAMAGR5=tB6Zab|{H7!(Hbu+cMh_$cLvuMgRO0C>|C7V+KK>nO6|@3pPj44dZ(xZwANAt`I7ijx<5N)EV$12L6?)bU z`Pa%a?cao6nz!DR2nxn;;Vlv!h|? zU2wvkhH4~q%K1&Fz%`RmzV%8#Y*;N9qyV5-TRasxnwcch`1WL~QvdAbl>TGcA!k-Q z5GXN>(&e?!FgG!=CM1K_eQkvzzHIt)#PL;EyZQ^titvgx-h_OBp%}U=#iOBllZk&| z6a1eCceQB83_Rn@^dEUv#)`Tzl`0uKM8Bg&p*JgA-jhFWjRzy|v zDm{$NSULLp`)4=-Ri*N5xtFET<+Z;l>@%cB_^X z7!~^2EHeL1++R^g2TswLHBLs_tKsrMot>Zk!%WUG1c=JKl2bTZ4(h+#KZq0IldjiM zWwA>`a92@L#3_HE9OAelBEce}1S&r4^eB-ekW&$suxfNS%$Bvm@BedvN#a$LvQ7jR zpUzwrRWd7@Kd%$4d+D!;^)^4AqSn;ViR=w}Cu~sYT(TdgQ%COlWqdlmJkczmMPwHf z4PQfzMt3yikL@+yO8*I>>6|}8MH?^^{oM+x18C?${(pYn`f*+S%c^~t?B?2|7v z-&!tsc*SD&z8UAf|MHY2UQm{j@lr^G2u{oM4{P_VW6<930CK2fr9z7vi9%Vq#M@o?F$g9MWXK-`hvO% zzQ0i4Sbxyq7~!GQ6hkt5mWNRPJQ-u_%k)9GTDngfuCk>OaQ@B;ZU)OvwtI_>cf0el*+3Wf^2rlhKWk9jAC9YM!aC zowbt847yz1*|?GZ>32*$4k!Lthbio^1nUOemZg{nL#OTshtBCc!lF+P zZ`AFq7k232T+mkFl#v83b8L+vpx(qJZts9cZ1Z`j z^A1RCuG6R2q37>f`{?n}gDMI9)jSV*orzJXhZLN|5tz{P0pvgukQspPeE%8;8%Oo) z+EvpTs?$6PRmRtFY)@!xDii2P%*?P&iRW^13vO z_wuv$EIre1_8m^Y3PHRpGSM&Kde1(Iwi*5fGb;t`Kp^8Zp5tLPwAaWo!vJc#2${xE zvm5}Nw9VpHQVj569VNxhw00W+P&}(Z2h`LA5l{`LCn6>iV8v!(Wfd?Y_vP(*6gje) zp^*>p6&{ie(e5En0DclZFT^P_;q6uZ`%w?KioQ0MlBhnt z@WDC0a=cVW#XbkB+LMF(gE%*9#gFM0SJ^oFWa)xJ7vaOqf0)#6!v%F%rinJ|;6)dJ zw1oOtY4Z*gtFMr6Xss5@EB2a8tsIwZwJiL18^VtjICKm&W%q(L&hywk?oYwL`dIo^ zKYi)Kzy^h(Me=FSK50?=k0dFFED5QZ|939nUGo1$-gm||^>ph7Q4vv55s)IN`~~SK zy@QBIuc0?VO6Wy;R}lpXiqbns3nfVJp@?*l5+c3#DjhWfe zVKRHq-fL#fTI*TQnt?&cP9cn!EuK-@Yx1MTNf&I~9NOR>mgpI`xa!?;tk6&7G21 z?n`LMI6mj?`W0RNpzC3m&NJ6{dzGJ4iJCzR#D3%dQ$uff=4`IpQsEUu}*ht=2d4ifU(PyNRwQob+Yg&v`Oy z9&di`BGuRU@OPv~WE@;yUA6MASsA9(9KQG4Q@%v0A$)S-m_tHn!rMWvUl%WN&54X z&^GWqml@+&pYnAnb$x9RQ25}kb$(br%Kwxy#m7LL1Ttd>+-r)HQ}29(`RB{752@~d z%U-r~*2>Y+axa&DVARPj+STKUEGvMm`t->Gu5cw?&98asdHN?O zZ3_NH(R~*V0nK z$w!rxdOc0W-h>VtTh$npxO^g4=+QL5!?yY9xPqC5fq_0%)R zx0x0AsBX`!Ckwpb!Qpzc=ldDs?%fjzEqpIpW~O7DdB6Q!W8OEN2k`Z!?|K%+aZC;F zH9l9ICO>B~DqNqDa0BwY5ablXzFAqC+}|~nS&_7;xsqksIUSREruw1lPMMsQXt7xe zsoy0eZO*TFRO0!INv0(5@n-mp`;?zF3_F&h;170+e6mQ;|A&*>nftRzJol|PX4?g0 zx0<#|AbFlZ(YYVTTqb)HisGke3{Zrd6UV*H`I+W){q}sHG_PywCi0egdy8a=j8#qJ?!%&%sa3A_q=Qdxm9_C$ ztu9K+%ZK0Co#&YpY)cd#!1>OJ+~|825(d9^#&Z4A2Xe^;DSpAbzkS99s0us6k(jT* zJK*7=f%E+O?t*eK*CT;!fRw$9W+1%mk1G8$k*M95R*URy4lErV9%_Lura6luJa>-7 zc{m$f%0IN+Z@*vr(2UMNU`#x=^j#-MM2A)a3*XKT@2h`!$(T5qx{VZJ(3O@{YJDef z4Vq}qsIR}IooTlOLu51e9(G!vXMywF7wVzA7=L}{mm&#S?uEB)XJ`>VgzQ@oJ@$eX zapEgGTDW~0%gOk4Q>0-13iTi%Pq%~1g{yEcIe_Nzowz>9RiBaA*%sBe%g+ynq8!+- z$D3vEG_i2XEIRpK>V?kYuh{5+s&lPU;wQhdM$hZqPf86TV>{1y1(G~;?lmQd;KPIG zE7|=hzilu=AaOvaf9tq>O4}1LTzvhaQtI0`Yw@F!+;3K#t*!~l)|K*%m+rs?r*2dr zdr&eQf1(vM7tQ**-OjpD(hmVW?Q+?*v|`OfZN$XrXxE7wQt1d&)J4mAn+I)EVqjXP?V#)7Q}b9^>W8OWMt!Qa^9kr+m?VIQ+5nZadvs zULk2*GyGSRv2UExIzy+Wmr6DYu239wLQNV1-H`nhFrxza2>y$`ozN+Njw_&(4d8~$ zgl4k5Nl}nLd|)HWAeQ(H&T7kt(M~aXu$xjf%Jd?#bmdh(dWwZ`_gi?xq4D(YHYj4w zR&>>Ebi-_E^!>(hs4gxW4PD#75iE^l7DBj5G9G1I2G;5;tx3M+EQn_lPuG#C-iKDvjB9>FDKtWxBMG(EX(9jt%VV2{odpOefU`AA@I%+DPB! zyojJ@S~%`(ZEbb!@L64573pzAX{GoZ2Ap7zyZZabMLC8(_UwBZ3u|-s?1x|78?jwn z<}7!eeb!8GeXgk)2FV!6AY;8a%r6_fdUZ#^EI3d4^!(F{qT9%h9!w*TutW#p-3O9( zw|KJW{ ztLEqZ8JV{?7ZJ!{vI7la~>?dy{+Be2UyqW;L! zLqT!DqXy=K`P>M`h(Y5Fi_fk%$j(9ddH7D7G6^%cW!>qPE?s&}u)TX*%?!+h8CGB^ zsw>iEq?cz!a6Ec3n#mbnmAJ!p?lkFZozvZVM<#9=h`GqO6Y9-|wdv!D5DF=lD?HL;+m zX8fyB^nCJ?%detHC&Zw|q`-{{qmu3S*rD)21P#;iv*(M+!|L2B+R*Y#4oebny%&KQ zWq)QjgqLhFKcChlZ5&agD5xF>fUQvcHyJ!u222l(esq}_yX@5f(_8}f_0_o0&94hn z*qVJXpTmzs9~5h~248$~@SNwlZn|Fjxdu+NsWiRh+@8!HmMvkbiyX{%o;Ob|rlHl@ z5o4LsEyBC+vHlVbP+eXAFL$-5hW61$?xTrdFoNenHb+O+TJz_Pqipf5y~DyWZJ7M2 z((>|o--}>yn`$0j+rRQDEiP@X7D|V*gPlHJq_HW;+dzhg*jpgAl0Ip3;yj+fOIP|& z$GuxD$tAlmirgwt=WK@^p+@bnZm7ar8{oA1*_1mhs)SxgEsk~60Z~KFFkWAQ27$| z6En8#hw@O(e%_yS!Ncd5N8a|+taAc8J6~l&WKyZ!0#a72Xx7t>C1_u zWxNd~zvjjml&0>fE-{C{iihMfzFbEdzPznV4yq9#ecc3BT{7k?r4pGwx-(`}@=^I@(I%;W3|+ zvkSjE#KfLKt!8Ft?}4Y8c?vLwn6C!OCR;cK&M4``Ml zw6-==pKWrZXj~Xo)4fhCH+xzSJ{ti#r2_~)>CvTdtu3ZGlZp=%J`YY7 zrVIoU?#t^)uIg3P!O=pnLm&;eAHhSl7hxl_^d6SvK_m>VY>r|^(?**^#?C%e+ z(xxExe7!W5*ZVa!^f&)+Sf#5kPzwi93V>1pLOG^B9Xt?Q&7V%M?F3GLKcXZ1ZO*#WWZf4RJtz zp1;0DwIp%tmp*inn^zty&z>uCRAT))1Uglzp@qjB6{hUW*v614P$Rg@}+hINN zJ?(0DJr>1Fl9(oc_VmhWnc3j%*NBr!xpjkqRS#Wjsk{sX6Wj_=GQmQq3+J-yr-V84 zpY*=a?*lJfR|hyK5r-ayYd@TQnFJ^umI`Gqz2Z82Q%{-#Js)BUrI7#lj^wJ94HKN} zpWWWO2KiiiN;n0$vEsZo!^)gE&oB9^)ZAYy2*D&^8-wkEYi5@qIuuy0c!3&rXTp%O zl5y?QjmuCF%Vgec*VNq*u{5j5^(fK%)EBYg{{sh2x&LC{>6dcedEU;WTlQvw%KXNb zLcu|wP<`7&oj2%{`|13FY#4|xUwr=Koj<#=dlPt9sU$i7C&LeRB}mxjhIxuSU?>W7 z&9dk&#SZbDi47vd!NvIg{QXx0j?1D?-cZ~cuTKR(n41A;GgxVF_i#{DuL{!!Z<^_G z%IzL1SkvYMKhFdbvlHzved&w?oQ2z66Iv{60|CWY14BgR;9V*z)fPJs`aY1e%7E5? zLud&4ADBY;#OKlQUm4hhTDy*HvCS53APORo7uyE8OsKQn z1x47Z-uqNVXP%K0vJE+p`W(q8PtEeucZh}#5(6qaAno^rIn}?UVvWsExD=&u<%PAL zk+)28NS!byo35cDZO~2z3f@VE`fF^g-PT)XB9iwtCYoDsd2_wUW?8a0)<%%l5u8}@ zZ!7>@;ZjdD^wA@{rzYFi0?foQ+n?)pC=Xdgz;8|D#9bsUpO1YmJYr7Edtk;zJbYS% z7EIG7&4pze+c84#U?B#jsrlkN_#wEK2qEKh3JX=F18bkBT}jFuZdiRXIq=`-chfx~ z?#awQgfo33T_=EqK!S<2FikIxavFSXROj+d4&b$LH9-w zx?h#rl9Y&5Sk)B7i{H^z9k+3s*^MceL3K3#0pCMnz%Zq&>GKDr{G1?!96s0}zGpB3 z%xH6a820f#=xSvdh9VZ%H2oodYs7Cxc=jcY@Wf~I;6%we8-EqyI48AXAXe z17v3|&OqZlQYSZ!fqT?JlOz}koSdB4y~7K_9wSzybsTqpmz_VT0Q)yd0?C(z4*w8nklHc99S&Wr39pVlc!enl zKp)Nt5#=*MoSaoht!&cvXDn$Ch57Z0@A7eJ1mN9MiW%KL)Yp#+fmVPgpSekL|Qt6B4v~|aK((iX| z+70Gp5bsk5gfXM*R<)tQRctF!KaX_?|}o`z6bCD zrOBk?u_fjOq-m6C*4B|353@8gaejM(*sW=idWm)|E0=Y>U|?_Ia`A2d4@409zD4VK zKiK5=Yye6@nr6XdsV*-ns9dce*|zQ9_JfWarjDu{)k|uqs@oF>Go6R{Owm9;1U!Cl zsp3!x9Ng?s$IDl)4j27tcy(ez+uhnK2O7$lfKhK+Byz7H$?30t3lS%3a(vR+cCn4F}rShC6}K^lIY3Gm=J~d;M-=mi!MW+VA4rYKq0t4uMz6bi^O^**UA;lH5xakzf4jqu9;+{TOyNl- z5D#g#%egn#M3^itiN8*i=0i}D94yBEnCHpS$1UXHgFu7x0Rvw~ zCC<)tU*Z+YA@Gu!)T+~hy@Tby1#|~M7iT=jJRP&J?UUK>fAbN$%nKx{!#mhp?{;_L zwXS<{#5wX8>6oEe6O0_u=4aBJQ2b6Dbq!E+57Buj&1SYv)xe~_=F{C8!91Oed>yIOAfsk={OHHpn!1b~}% zC+54fHt*zPC5AYMzZdHOGG5*rVhnM7fdE~b`g@j6({~_650w_=_`$>h@*xYmG8@NU zk-t@t#$C>sjqGirNQk&+u-F1vMTzf+$V-2&`*6}IM=ORn=Yku+gL=pq7$ivDky0Hp zOXY?rVE((+5Q+no+|O00Y-;xr;s`Zs?d=x% z7B4@y%3b=~RP8)yDkmLK`P_MuU$#xYt@L%+wS85QSkJiR9HZ2&2WHaD2HdrZUy1h} zzSct8WG&_ER@&%&z-iFcdY0C%N%%e9ZiFLjTbsCI{#qe(X>jba>dWkx>oJX46^CI5 z2*rWO%zfmCqx=&(RpP{IN(D23;diN9l4&jMtJdMSeAa}|o>A&PGR>cxiSi)&Il)QT z*)M4ahtgAJg(z7okJU*P)v9ANN|g8~a{r~)-q*Z0!aUuDfh zHmv_AeV)+R-O_1zHN-GhYnNmf4f|+|xY~#M{sr6*GJJ>Frvb|43w}FR6U+xq)yp3) zkDziT@<#3${+;=gz#FOlOot4}ib_i?930d_e|wfI@GB@s`dj=@&A)T|2n@Ecs4?(CC^9iItsS-W)-E-6|DC+xSt%Y%US-7}xvToYV(k*m zf{{!o`^?0_yf=p6K=( z6%ZzOb9UV+bcPJeopaiArP6nRPJ2Mt;oK=8pAWePW`A{X^xVw=A^$`v=93t^Rs+4% zjG(|*QU169;&}c`K}x@+u+#gr8=VFiytZrgpft@MPiy!uZQ(W)a}S-W_p5i~cY^|i z66Zmvt{a|BNhJOQghU|2;tPOGX^FiyuPCMaOXUc6^;_Ifb0bR7K1X{B!6dFK|g))pK2bD76 zcT7Xu-u|b0X+#ni31$f?P5Dr8lLn^E;-u7C)6~?|b1n}5lViO0aiUOS(^!vR_43e> zmF*#ed%4qW4hz1gRSwMq91*#8%8^mMa^ivyNq*QuI!rgNw>@0KX{w`o>ZhTht82Xl z^{^XgOdyVQIE?WDv0(`~MaY~b?&=1Dgm;>Q*(`O8_-$e9WzrH7vmmocoVB!x%McMP zfj|ntO5(&*Y$b`apVJT&%MKxanJz}0vl}0k)iABh(^*ZQAVmF=@wg`;ld?Rk@8wsu zA04CQ-lY?*fmj2%*ZicggW&^8Gw!Iw+r+pJIEO66rbAuInwGDvvR^~nlp5F+(D!Oz zd}BJHjiyssvUr2n0CTj2uJPW`+A*LgQJIBLv|JDs-nlE2)A;HTA*3O?u*l`~S8GV!T`+ih_1QZ4a&CX6`H)!>M3LYl z6ADE5KmB1F|6YuRPe6vv3o&`aK~?RqdkJw7eoL0epIGTw2qLO`f3HQCI1=X*@Q40g zG5mk}M~4ydc51(m6{K{3;(Mf>y&AG2`sxcO)2e9I3^KF%bj% zjoym|(#{Hk;X2dH&RPIroMMzv)3tSP?yQ|FR>s#i*i9J*5b)~Mf$*^0+$Wi@sp)8o zS#yQ8>LSKF+WQqCYhVi<&}Y**qgXcEAzmXw{baRjYirA!iapFaFxzK7ewd#>!2eCw zsVK2~c^7Z=?WC7~+b_YYzYQuOyr?>IRxaMffUOK8K}`DV0FJ7Q;_W6+FE$@1j;DD@ z-ZRax$-d`ZVjG9}P4=)!wXszH-SNJt)*SMGF-9N| z{kx^I$1kcVCtJ)nRDTEkb}YNp-_L!#K2M-wdP?B_=pqOJZ4ALFV5p|(Trf`4hB(6d zckgfQ?A&!UJ59qVD=UK(_u!4RFR}5()v@3E))N%N!mbWnWw~bjXGcR{k&!_V9MQ|&Tx zC%yhc<}}QrQVW>iQAfLnHz8Py|B%@G(-&crQsskpJ^IZWNX}h@ z8OFPH;8wAZ35N&QC|T^j-{$tVE#-p)y=ekWlMcje*S6G*1r^k z;^~O=V~Gx84mf8+4Y$`wV^XC1BApEqd_h7& zLI-Kk^Ww-Vn=wz~H%juPADdWZzq@?!W>)r)nc3@8LhQ`^x^F`-Rn^8&motXZHApAx zQcWT--Kv4J5v>Su2;*PcB&6v+Sk$i7rFHh)a{6;LK5}rvevsH z8s_~uPnQ>)!(Yc+5A-pSS#M;(5Z)!iEwTrRBc|RZ8{N>Z=Xm?_mzsdaP&}T6o+jvS zwVMSlxAnP37r&e)ixlVw!gceYstx9=MRZq^ z?}gmJJRb3(8~R6yArL?5BWY>&t~8pGk`mAKWcb^3;Mj}=N2`v;zW0n$O=85hh}m=l zA?tDUwYPWAUuydBA5-<_=*&(iEQdwn`S>2@($5uEGLFmFuF`28S)bK(;P%#k$akR= zHZMo1GOZQnNFghrFQUOle0XIPJr3gcjGX^IA@t2mNPx6$`Zc2inOSoaT9L_v()3bDj5P zbT1?LlU65Em-7G*%T4b?(h<1~8kqaP&lPo5a>H{iRp-IB(#f9Zt1QwOve~e=WbE*$ z&SNm#nh~?IC|W1oBP=iN-v|g?RPy|Ix{^EI^dhzjCz$hR#`A5*X;6T1S1}7G<^8oV za`gPyJ;VR%LbQB>kp9-p>&ypdp4BMl$Feb5J8h8RvVz&66x&@Nwj!kNriX`z7wn6o zctw<>TMyA}XSKNg^?L)&eaa|^nay+aXChmZ zrQ7|Q@MkJRwzDqxG@GHP$EwOvHyKs@G-NYE&S`l3wTtEnzjGcfPMIMMK1xUue6az1 z|A+{PxJ*s|qa^YOOcGql_<6Gb9Q#)*w|~l^olUn(#uX^YGwE+ICFj{Q*K1*<{VVpw znn}#Bf&<9FhbG)XBhXs|xjLKCH1Ong-o*;>oB2@|67*&|DXE+pG2c|w{-8I86~Z)_AFIR=O{xj&m)ZH9Hc2_;pM{Gn|3x1 zUQRQ%wbLNeY?{p^atHzhg2D7v0h>I%0ow`&vu9G^Xzk!SMg0ns>42Rtg}_VTITyV4 zv+{4`0@3*Q_A38R{oza^b9HZ*wTkF>OaKOEuv^TnxQF6`3Ur+Y9PAvr8v5z4TS%#| z)8AF^r!qRjo}OGHyR|^g!OO4iV{p>d-x(JdRkep@TsfA)lAQpKgq%n^fY3!(;q5QA zDlhL{zQssa2;$X~V?#ngVfi;519*Fou3csLE=3BV2ta-(r&vVs<7onj`4+z|J``qA zhm!#h?+rCQHJ3Bqiz@%*hnBs@CH@FmJZpaG~|7nh9iLIXXm$Got z-IzG-nX{Bz*BNL0Qv{|Jm;-YmGLbV?G!RF#I%Q6jN!9^{5QmQ*;_;!42#u|Q0`3^n zo;01aNnbG8<~W?w-D&5e=6*tUnqOM)zdoX@e%g4&5$90{A_WXF`4{9@TB=pbamK{_ zvJo?T(0YsJPUf>*A)R$KZuD;9lHDt}WQm3iv*ypW!X~I5xyuwKq~u+Df)`3FhkkdGuyreV zOwZ5|F&#LOzGy~3v<=0wTd60i>9jN1tk#AVyEUFj((GC9Fy~6Ma!Q=82J{gsIUiA< zSGFk}4tLhIZ|zUwI>F;7o8>n;Q`84VG0FKkz>{}8vEDc`Ox3i}htybFjGP?#9 z&pT{xXu1omp782W=O$y6j!=HN0Itm6Uycu*&TY+w<+FUtLljqe3~Uk_G+!J%v8tGa zUl<-gQvLNoZ0x970;GfnKI)+c&!TA8o+h-fT!3q|zvbHGkt{Ogcxu<}*?!-v`D^xA ziA$Cc*Hq!!buX>LG1FHFwbenxJGE&B*})?w@h&c#N9z@KQ}o`yhSfET;Y+P6??^*%S{S0FkKA<=QpOd!<0n$(>6^&gaZg zG<`SO?f;Ho2Y9K+N1rV06iP~REdf1mMPs@oGWo~43d4fQ+iY`$pMIYTJoH~31t8o~ zP`QS*2#qQF59_hU(s~mApQYms)R4#6h1?Z{9}gO&8sp-s;_S3`;kOi?@7BQ;o{QS( z^;+OZ?FrzZo`@4aiNRIaB)@NP-{$CSJTqbQ_ynY@$iz&oKQ}ByEYk1n7S$df?pZx3 z6P))(A;ZQReV4%rf_uMH@VgGns5XwJM`AV;{>v|D97yj>XO-QNeO=ep*@KSmkF)Rl zeG-cbjAa8Uz~vW>L9Fjyg<3YMk{foBe%_xMzP z`yYTD;?nCcqcqkb2Nc4hX5xjz3+2ytDgV$yFd(xrd_z%Od%tr?I|_Tn)hdCXeQ8B# z=KcWzWwb!+vcTjjejvdIWdyLOKEEk;%)rUWC=d&Kv&Uz(kC_Mx<=ONLd6G4W{1%AI zoIZHc|C#MpV|;(WB7ESq|CMx#&*(t^kAMwtDSteQDrm?a?+3uZshVGPg7}LyRAp>I zYlPn6Z_5bXSxsop*h0!;B2R}|MV<|H>E#j%#Clw>6H@c!EN8rzV*q0Sm+(WW(Q|_Toe@& zep+|Wc|Vu=(oCkU&$VHavgtG3qtsK|3No^=MDbZNN%ZE0&FR2}Kxmx~Q#`I(-3*^O zbz7oR_stu*gn(6ImJ+KLq@rxCLokr!OpqV@aFHy`w{f|kFOX^RWNC7ni~QO(tG-6n zz(!x%S=}zBPEVed0&x+MVT6zU!Smb~r^or7eKRdF38okMO>u@tl_z>d7s-NbXd3%p z1sQYt&Zg#B6Wev3#FTw~}CO4YkoqXYQ! z*C80d{Qe!!PALcu%JJLKMqeBX?)&-))a%ceU>hd=x64*xJKrrsoM>dx71JkA6D}Ng zS2gAfhkXQgx!5>G^v!`Cs)~Dh>(6Pr7hM0FlSzfDCG+WXbo0LL@z$HPg|%p}+_eGs zG$}i(uC>FH4y3?Z@%O&8I?l)a>k9pmm@sv<=z>$9F<#%5WHv_WR0wLGuG(ta=p@*L<(6;dM!H!|2UW`?1&(qQBwXlzt*Csh z&e>S-0vTCC>izr=Bf_2KN`X!xcw~Q`9c;mXI_KtEcG5G^RuR1yro!#gyWdmS4(s=N z$_w~op5?SGnY3Gu#WY5}OW@jjzWhSa@Q>$xynm7L%osAJ*y2oD`ZM|>e*166D|nrl zGwA5~A>`L+vUnxW=M8qZH%lkp?4eiXY2EaGT5bV^wPahb58qYXL^g5Q@rA^ds{Vc z7d1XLsOQ}Or7mG_%ZuLg7PSMi1^%u&Gedtka-xQt@hp zdh?Yj(o(qQf<_%Nqu{$epbod+F!p#%Q7Na)sCwk$WiiEtI~pT&rZfaDgWU#_i>*)e zP`dVaEot~reb`=~$$df|Il$@X9Plf4jr5ZJCtmIkmMUJnL_O*=&#GMbw!BiaZZ_bL z<)iyI7QkDk*JY9*vyz=O7-`P#q z{NCdObOtJdxJj|?!4Yb;TX-2E6weK9Ex!GtQ@>}-nzc9{!?CB2ysB-QthsKqSZG22 zAyBZ^lj%#b1BjxHOf}sA?mZEBo?U}@I<1h!p#jPjqhC|H%6n(UWt(<^DTWNyV`$*m zdtevA-aQG-fN>jw-geC9C(OxMr>vK|muYQ70JHjz&?`P|F!gCrXCGPPkc!) z1h4+%5$Cw%C86@5f5ne=*!w{kd021DpxRIA(+}?T|6KcB?_aPu?%UJRQ$QF{g-nnF zkeX^Ka&D!Lt$Kg1i4S@sN4X4%S_i~0ty4{}Mo#g^k!yB!^`p~4|;85jQf+Q-E}czI%#Xx&;BIJ zA7%MU=ku7f0@v3gM990pCH;nA%r#u*sVi3StVJLOr7_un! zS6k{xr<}LAqI>f%ZL+rSm@T)91WGQ6cbg%3QR_D=wtj0tCdf3I%}I0C%)z1`U3cuU z<-GgHYcnnlB+f;SUW_Ne%zXxmmWk=;;=j=NA=D3CwXi>-V7lz5`Q8|Mgtg)<@U9q5fsHlmD5RjRpX0baPY3mGrJOK5ey@ z%w%9%zsDKBoXz$8%_8xxHoP zS|)jv|NVli-mAJ76wXrw`(^AxNAR=u}m|iWdXhhFH0S)JJ%H>xJ5-sqW!&M@w|2-!xNRgtIguFUX1J zxIcjKmdSyTMn%7Ad+q}v5bJz$95e0J3U~~{(r9gxd-&6p&Q3AS@sW+cG`lg2KAF1l z!=wa2iP#nA^~7@(S|{ooPVR4RXVO0vNz=}GQ}5%XAP&!a_Zfsfp0TJTj<{YprV04s zP`@4Q>JH5JxYh5aU>Vtz|7Pt1E+_!O*66M8@nQAI{NQUMKd*fOPlGslFc##>CMG}v z%gKgq$vRc7CnE&NgUuwbpuNu@9r9-XH!Zij#5n{fo-Q9=fB3 zMtT#;kau+2iC+fPWnYXqks65jz#?_(6SiSV6zaNrhpN1sm{yWUzHED+Em|+J zgChOP;)xv9h$y}-J$KN7Iq+9rwi>3|th1IJ&SRph~bz?w}=RL!IGJRU_% zt`mGJ+;4c43#YQm@KL;Q7qB0H3Y?5#A8Y5sB=;14)MRhJcAR13!^bwveuTQPzPgMo zcd)lFAg|QUEV-w&x3DWSMl2n16R2rV9_h+f7Jnrw78e9M?k6~gr{-d-zLur zq06dgc|yRz+se_iA>>c28=Y{&k)WD!IOgb;_R#!qUCU?atNyD(lRLJQku?OI-RMPg zU(;8$TMh*GrdI6Snsvf_t5*{NH?!cM$@7!!+@=soPxg;hRX^_>JU%5#Scxcsq zZ^?$9^+;rJAa^GYrx~b;Ji}@>Uy!JCI?+Q9$xi9ef1KKh_LUrK=&~v7k@}gGT5r1T zM3+CrBO<@`LqMrK+MuwJkDzEH8VG&!2E}ic%AzM>GulXa)05=DFI6U#5FXq0IO4;6 zfN_fJDy&_C#Xk>l3ywWW^)?5c5R#^bw~*MWn!UW{=7D&P_3%2Lfuwa)w~O!4&B=>` z_d;WOeq1e}mGeT4n#oPA1Q|CsXSA$^u9;D7W6}u*LAsu!aRo241*Hp-m54WFtXBpK z;22RVJ7~_66^)$ErNuY3V$wk~%2fz4uK4ob4}MPQR&CmQuI+s`>r`TWShvI+2Na4p zFZ4kw$wIh z^>@GJ0VidQAZhATA%+t|I(w32aqE`cdPy$sVvm3WbHrc*sm|$J_c}~qk1&T zun%GY+Y5Ir?b;-4niN^)^-S#*Y<({__Qx8ozy5-F4867VNQ?T$L92(v{WZN}&Bcr! zi*YrklLaXoX{rJfpQCKCDjPBb81DW$4b~d_9*Ujmz2ET_t(TiL@7Jm1CuG2H?Sgx) z8hdE+D5?!`A!a7rH14cbxxBR+VvdfXQ8!z1pm9(j(jJ@V*+!NZw|n}JX(E{WA$EXs zGxKwV_xD52$)l|lw62yd&7j=W2Akd?c96UDwP-h#D(j?}uq;>_4)e89f=-5p*`QnA zQJqK(@%`da5%vEfbQO9>1tM@PXkW!!n0k~13i)4NC<`0sh|{l!W=K4W%zTBO9J%n; zIw^^|=aK>Er1T*V+TLzcs1RijNC9PtJ`Pri?`{!EoR6+r6ghQ?d+VU=t`3n5GlG~y zH33(^Kx3rmV_r!ef3tsW7_iW7d=Q+Er~5&wxvr_V|SXJuQEBTn05 z%ZY~w@&;yL*|Rs4VkQgmTf(^gQ451ka5#4W7vwjoi84euCh0R9rw6ldC?ZLe-v`&K zw;tJSxoD^2ZEiiWt@;~D<$+gCfok{6wICYu6FLoIx39ii=NrF_C=fW?CZy0I4TP@^ zyfxHKPJqL1T6NNVbuV@@TO=1~UmxBx?B-bYD9;;!H+Z~P;T(e>#b{PWnRQ8nr^ zzs2xn_PR}Krg-w3mhS@Z7F_jjJ6*j8u><>Q&g7I16^Sc5^gT}nOl7;4QN5Zae+Pcw zzO{hwg3Vn%qXvL?dx$}T7O*#67c&cj@42v9Cx)()TeRt< zfHaVp<|u5mk6TR_wtdOLez2yY?>m>o^C=i|$NXo3nSoTp>d36H{?vIYnN5(?oXG*Z zEEGuC`nfW6=WsGch+^jvMzj-}wqHH4&x<(h-`U&w`dXsD4)IFJ>pqISwqDvXZ@X)d z3J9+$c>&{vqfad2m5_-{<6d@gU#wyP39>X|GU?E6(A}&& zQZLc0JzX?lG{guHWTd-u@Eeo9b4C-!A+YCShZ&Ss6@Jes3;(!WI)}JJlx5*Wr3Nb7 zBYQh$ARQXeOfOhrc6+O+xl<6E%%2U8GEgcQkCce=Dt+!zQc4k|N|+J_esAQ<&`;0( zR4Dnb-o)TbOm)`)DwE>t#7R<;Mjv_JhIcg2UJSiKr`A;L#^n%y`?e{6FY4Om@RBVD z(O^r8{Mkn^*&8DbISV6ycA8=qtPE+HhDTN!+gT|qUY^;gsto5ljpmQW>fMQFQmVR& zl7-m)MD%_@C<@XuojDJxCrSiWFo7-9`fLpvSY{!0oE_|+0JKJ`^X0XjNFijcc831i zMxnnddx!KNH$QzA-tO98V%4!v77`~K=hvhHeG%E{e}0-Er)C*PFQ@x%Dt8aA&?)cF zuQ`G)8tGu|_$!k8BMvFkl?ixP6SlTzd^A_`97A3RGBBNgJG>D@nYV5yc2~fL`GvIZ zp*;De#Yb96MDDpe+u2>tin`-+`a94o!}6ez`qX=hqn-Rjkb|rE#1vXF()qGE(Xnqn zR=Mxj=HwV>6QC^EDX6L%io6dvmFeBxnG?N+X|eoa2-t%~%+dxrt5MTCd#nbVeR|Hz z%a?VsM7xFTH#V}B`U)GKwNy-9JQ&&zL?@dCVp1l>SHj(Gi_bMNDJd(bET*4~Ug*2F zSYrTRH}L7ZktBa2U)qJ=9EnsT)$ozOAm8OKfmCHj`4lGwLTZ+3=Nn6-wG7(SSV%Mj zgwD?D?1s42kF<=t^+`&mewrxKi~3{;Z6a_-{ydN59z5XGW0Ati4h5a6%K3R{CgqHH zz%B-_>BgJonJ^qaoWXA(SaSb6%r8zUb1@$N3{>9#BUC4RmIQ z9DQo!l>=50Eo_O>o?8kyr|})^^OLS!S55=5KdTm#&9dP^H|k)n6q$HJowQ=f&3G|! z%0hO}kKk)#jiHeA%EFVtl&MBSz~*%9Kaj)dF9}3<6N9l_B9~c=&*!Kj96ciuaV7r% zqZjoE8DXvVblKd^-9O5FrftB9T#jG)^VuZ&6At)&P1&#M}pt zKHAuy(~$Rq*UN!q>p7RR$EACM_;stq8rwG$A>$ileC$2xak&V;Wo>~Kg!`?ZS6~pd zLnm$b!PXn*l01MJNg#>f9I>ntf>;8n;?>tl;uKAPlFnMeX(5|Riwo2pCeo5?*|`7~X@<683ZE($>B$CEaE(8E(2H-wc{wXH=zs>h7N273=Uj zzp7EcFa^uM;3Tp5&TW@M(0A$P0mCDM2+Z~^h(7F)VfrQ`L?{NNOIne5>DQz)=i|Fl z`ppu#Au&~)?AlTFxNN4NBZjIg*RE7up}h8ZdY|wS_wz)KPx=Z(M02_xn*xZ4xx`0k z-UA>NzamLV(=rQcu=uLq`E-k=A4f*~GzHrcU&`majsp9tJ-l5$o!pCO{+VRPrB%tW zNk=kcNTW)SCjyI*joiEveE@fUXuD>fK^}Fpl9W2(yKRG0P+(@xS&6QvdBo04dcHJ1 zN8B`i;p`rX3^ZQTah{1)1hTQVUbPHscdk3%BvybTs76Seu|mR(zLcnwnk6Fia=CP# z!2{Hs{DcwAp4u-0|5W2vMJwYgECNNA!uC&Id-&!Zo!E^_=t5k$(okns6>3mh02B+t zhBPA89Mj*i32V4i3##VatlGl-CX>DuoTWS-mqCf=*F?AV)H%Ca^9u^~|IwbiH6&m- zHDy}b@`Ip;$uX;eTZrm@K@Zi1Ipc|v-aZ3Wx^_~6v_KxkC4VJF?eENEMHihrI= zRm=aZT=-#*Uc$QxuyOlnnyhbl?IkJyihmm;*FM8IlCLQqCt~~ z!wfX_W5NeBxUiycarK8YjSmLfCev<@4-5^tkM(`qe^*NK1@AB671L-?tLSAqq#2jy z9t~;gQ0pEUOyGP;t?jsTJWS?c7dLWF0~?rn8d3Oaws(207_*iBl!3=;XFUMDbf(RL z`!iiHu5K2Y%Ek%sR`nDVA-L9JzWATW7)+?xkwb4h5!!$l4b zH*-I0)swBl>{HJZ>rq7=ee=NgY1<7$X>=m9w#?eI0Dfl&oqJ!>zVPa480!?8Sr>h! z?(qx-RXVIq)QdKl4R;=CjNnhLqEpYvLbtMc6-RxQ$*mW2i;k0Tx6imR$C4UxDM8w7 z6>Lx$hhj ziliKUWc53^?`jqckJIftiT@;Za6TNYM#iq{w!dk1Vi7cVo{}1Yk}g^gJ6uq6dX{Po zJaO@_HUpV!gCw*ZBCk8iTvWjpO9;i9gb~Sh^#v6IqQ>4Tru(7?)MS}is@s{L zJxxl)pGSDlOxO^c^0_OOgsF#KnB5-h+KC$Sm^@we)si0eg4W`P+nTSUsf6(vB9r= z8{LkOe`N(~R=z;}Nuvmi@?FI`1{u87CJ{~9XU^T`wUd0J{CLwU@9W**dThIS^N&O4to)^w@% zh#_O3O*uA*NBjm^Xk*!&P?H?hz5FrrMHI=Q)gd19@=Yz+xv+x+-)_L+?3<`xar<)A zRc%Mb%7>8ePl7{Z$n3dNUun}VJ$eECUwplXKh^#JKYl1JNf9o)5<*CJW&_!q?Ck8l zX9-Ej-lK5ry~jba_c}PX>`lh8{hlvf*Y&RJ^S%B4059kDd_5oQe!o8+j|-i{^Lx?5 z`p!;cPmV9Ctc_phR{)xQm_vcycPS|RCL2C1RkVD4ND_7G4G;RFQoC~TRo{WZ1hyBa zq?@x~Rg+cl_{_cM#(TF7*dOD@aMr#`0YNW?Al4q+lKLs3&t;Put+W`O%10KjR!PW> z`_XFhPBO8Q1ATiV^VCGhmWN-_$L8Y$Z*>d4K0HqHoSIR~H-PK4(<>=CD=PNc4@FZ_ zjB|-89q8|gg%4OFGnZ_1!<{9Q^=aX0BFXY3En-Jy)JA)1Hdy7A8lut8vun(|`&E-~ z7EZ6Dyj3D8S500Y618CtmUP-hZP}iov<}w;jZu#dz3+HT%Ap$8^_njjcPKl%u5xx= zCYrjI?aFFWNwoob@aj>!OheD-bF)OtImG@*v{19(kIA)d_e@Pk8T+hDveHp@2XEcI2MgGE zEk6b^$NyBAJWDYkXo?$nU|kxS=DVuXdKlkzd1x_OJPwy5i33Z;m1cE?Wm^Hi#<}0F z>lqv16_i?Avk$taQ>jx`A6XT6w(aZ8DzjU|(N(yFEn>}T6q@R9G%9SJ8RPv__fpxD z^>S?6DsuF~Tba9LBw5r5=RHivby+L@5cL;kfP$gd3Gt-Gn39i0#WRaVZNTC@;gMne z<}7(-bC=Ep`N(*H^VIKFsL!s;M0UGA%gT2zfoxK=u+I(Iv$VJ?}3g-7dKu#l&}A63e~USdVB>Q#O^C94*%q&r#y5 zLquImre;-2e?mY&$n)$YvwrG_nN`l4e;H^{hTsTT$eF!*WaOtV*>5kY7mg!vrm2g( zJQ)~ov9zMEcfmz!uJWXUzi{IGC2gaxiV8g6o&oY*23 zF#2kgvQDvqEwHciyK|fI%;`bFwi`@F4Ww>MP(^UTqGO;GF(`R4(=0%@-huPO^))t- zA^3w-<#h?OSQZ~I*BgW5BvJVLdN5-#*W5w5grb2tuRHIC^*kmuVG{2*>t^aJ7n3}<^e7-|9 zHeVjUa?->5>%-ja;S%kj8m9#p40gua{8uwIs&Ejwdyf^>^oeeL!|?8$x#!chzfmV# zlu%`pVf-?{`?A#5RwF#DK1?C1?;7q=@pDW5zm5=GD)=>;)2^Q}*}ovUJM~Lg(6t+@ z@61vyn%wi20$D?Ho7(8E8pv|YAoCRQ?ad@?E9h^O2Zc}O_Lln!lop@8CGOmBGqY$l zrY)3VC%{(XZ>HtvK?rB!N+|aFN=uZQrX(Rj*NUxsVpm~EM!w_0AZK{Iff?4jtF~H| z2RLRb!FL$!YEH>e(V(B}lH&921~kRdfC?7oKo}^|vo^3pcech9>9ePon2?R#qvAOv zLIK?^JD;NVc3>u^*EM(5HI9aFRXXSo0Hy!8V+uk5WZ;Escl`QLM+Dx9I z4cQ0l_8j9!`wttpLF7Z~5*;gpBs!8nB6XB1J~1~Uhtg1^SIK?|3?DECoKWEi?RWT< zbaPS%>ZSGI#Fa+r1%?Vj5>@5M2%LKnj!uW1@jH68PeJ69V>ruKy$UKfYC`xxeVBSw zA2#ZRoN!>mnO@E5)(BD8WoDW9u5=%*^(NXadG^&pqT<7<%*du#&bs-4spN%5lfiUu zYsH){^SfCmWk>BJHTIF2UPcf18XCL%?%#@dG_^ zUJ6pQgP5;tPrl_`j(T~q>mMv2P3w)B-wEhZ<+|`5A4jQOx8ASpY%2j8OshUlJT8?N zF*RbUy{{ZO?QAy1Gg3OgFC1c_QX!eV5ucP+$+vOrti+1FJoVDV%TO6<(a>2KbCq-m zpg)3kkDh+RPV5-sqpIzJYqS>n`}BShg_6O9$EJb{AU))+ArkR-Y{!ZU@h`d@L_Cx! zI*?B8(tnw7I3-)}B=G)%7y9Fk9Jc+Kz>il7?29R0+mc!a7b!ordt_*S>aungC)wV$ zW)}$cxq%kI<+#fy4)BZ4?QLs^omLTsfG&@S=gX9u$4N3e)z@dfR0Mluq3T=RE3M(| z5JJ%LSBrE)0-QTgh~hOh{)+&&MJh=Tqf(~w1ND+M=S)?e1bjH#$%aJY`}o?m!NWw( zrAU>mOvHk4*(uS|3y2ds1kA2pD)-DhgPRBZQthC!HX*x)#`eh*1h-zC%{?EUNBV1= zyo1kYC<{ujPCvz_VYH_JN+21YW}OoSkR&$@@1QG!Bry2e4>}U{4{TZ#s7@#A)7ssbG05_Q=5q&Vx7|ySds{t%wLhXhBx%bh zETg~9PUG=i&;U!eB$gVjjbtLf==zkIPVpGEC$O1J*j0I2xF2W;a<;)whl0NQ#NmuP&{_Nhnu<(fF9mJCPHXOatRk?hHUHjO;h zEuQr!sGQasC%14WU2#mIfiE555-n!++GKDDmOSpB?AAKybNXvO+gi`-rMHmpgP~iR zsXP=ONGrm!d-1tE;Lu z%M%9=>u1W%6&<2gp{%-k7b>uXNO+Ej+$tQC*DFdxE8`{6Nri+Z?v z^Xnc;se|2ZT@vm;cN7#a%l|OFAwO>Ujd6jSRGs5V$ro4Gc8?`?^1;psU&eO)VuMrX z1xze1S(??X}`PRQ{{ilKMM8y&>f{lt^YH2F>6%r0#R>_syzdAx=C)InR0 z{0q}zS%Kzi^OyznhIrXxnxFqThb_B=o!Ewl)(Y7-<5cKNG06xEX&_>wTXuExHAbDjtGxTo>Ye`V0G4X(R7|ISkg|XI2lk}{Ap|Wy${|$! zb9)VDm$QAHYsSOUZdOX(8fHEhM+1=TYv!O+^p-QOPG)MfbMe_eCt?r4A+r4X1kKxw zit5qmlH%mf2`*CaB{;3=9ATFQO|M92C8Poend&$s8Z9#ZZW{Z5>7gvpHa`x19OQCW zyCxya-aurHm1=WDYjBWQufl&FtpVzBm*>95kYmI~p)6cic&``zbu9~wf?R~8&_Mq@ zq@oI%oMy@i5xK|5HKsDk%_>WYS#DXL3|xR3GZa)H(y+8KblctI1vg>7ViUEQLGn}B z14A)*+Hv*MnRB!O)5sm33?hoKHKHkp?SZtTstL!!=L5#y8+5lH#Ij{iZoBjsQK@l$ zOjy)SM#yHB-8Zf~lk7F=cULmFq0w)p<9h5m;Iycr&G+$ncXEI#}gItyPm|<_fM+0M*U^o53?()<_&OADr+wA%% zczdl=VC+;6??MpTw1OeBHTLHI+pAuRL}RC>ou>(w)br@scq)rfyMc^cB?KcUq==k_wzU<@gd-<{{+1abX2HK+uGCVxXLb|!n0j>p> z@#~hnSK-X*6HW5@wOJcLDIh#aZo6pp!zDTg8U+zAd3QY6$a-OEj*xjSo!X#xtDZ;@;@jHbd5Y|NdjbYA^%xPo;wdeliE51D`Gi$Hu!Uv|^;lh`7kT8d=Cv-5 zQjRxw&P*%PP7Uu+`miCOdMAmuD&+BQ%c$1!mP2iC*9mc0U>CQOd*`E69IbSt@Szioe=81 zMnnlZs5gx<09y%&<6gN;=D*}(IsRR_HGF)3->#W2I_2>p!ZCErPbFpi$ru=9#mCEQ z3lvJ@>+CzPDklaTD*`TEe&6h~6tS;a7-RG#CO(F7Dk<*x8G>E6;IM{El)-(Hs_Mfv z><$YhX4k1tXE#jQ92#zxWy2jNRfqQX<>Y^8i}6Xc`qG```zLygei=n_=%E7 z?pgMP<@Zm;ZnL2tEgILI=j^>-RJbF*J%#JcQf^$g&Pd>({R*{8qhn9v!;SX+jY5nP z@{}H{Wzq1Xybc_DeAowItcZYV8V@q4-~gq@`cuv)$wiyE!H7bAx&|5IB` zAO`6_e2?^C^$6K@bGkj16iXq3`-1JpA_h~L@1*S=%GDo7I-QFJ%`^h9jfzT_tv5!! z;?xv9+a54+<%j=-)4EiuTiov7+3uCz$O~?^Ni*pltcx@c&tk>c^e&5IaL{e)IR%P| zad+(R(UB!~9Pf4tWs}XbqGZDz0;+-F$GL;oaQpRuh6&ytw`OA6GLjBEJXklV=2BqU ztTF12wY#5)=vYeHAl?Ga>5kDpqvo23WWth1HS^NxR zFb444z@5j%{bAZC%$h$(yh@|f%&Jtgqqn?hqC;-Vljw-O(8E{5lee2((ON_W)eN2P zkYP0A{RObi3v2`2@a)o)?PQmD|BDXVhgYhr9m~bdwLhHw)OvFARUS*B@f+BA1(yvw zWw%N&Xi0X~B7= zWZD#w7YM)K5attwcT1lB2YlMCqDHMk=tL_n8@-S%^E6mZVh4;1wh*L@dXzF)nH^gc zC(YIM!vWMV6{JOWcTZzIRt|mDlUw7QE}YimN>*ie>Z?jD0ALp@mP1Oyl*=ti6o~a#b8NQln>p)NN5@r@@ zr;SJThI1-Dzc7n?x*cuJC2HVe6Je$_Nkp?WQ<0_cO8p+s=JuXkmayy$>tjADI3M5Z zuo=z(GbtU)w!z*^x?3!%_uzUBpBg?(&hk9-Lf5KHODs!7!4w+eo)YiWjZvdH*IR_&bGy4!|mqdQ-R18@7KcGF~%3R z5r(nPkaiBo>Nz*=u!a<}={HDxOW;@+>l>#M7y>}NGDIhQ6;f{?VU|9FLrC?T1D@xr z+qeuO@#*c4si_R^`_SEGNxfg_cd~@)GP@3;wu~-v7Zp-E4`^3X6KCu;7T+Q(Z}LCV zT)kw-Z7mV{2KDx;psRLKWtX409gT_5&yxSz{9-oL)l$Ck&+D$e$3^aW^iJfeks^+3Q^&G&Ep)UR_ zy7-zlOoJv2pV`kewkIje*4S~Jk_C{LXzSRBDP(X)Jq!Y3lGke2@K(QH#A0@RtcCt` z{0t%~n10-Q-bdOjWS@Ar<7tnrMOpaV&hzBn`n4#$kpC7q5(Z}V{NuZ|OO=%~t(7Y= zH?G-C3tOaX%u7r*I>hA#4#lS)P(B>^LEK(Ey)dbdwe7R&)9lPu>#i>rOhx*>O3_fi z2&=(Xpmu7$OBK`{o$lLeIr|1B*#?A`k&|`>OxAH%bCyDW<}c2;uu_7~CE9vkLSwO) z3dpkngWe(&FPH9Urrg)7c}^PTV9fM_jl=Y8Y7E#E5nhgpnUI=iah=Xt020}SEO+do ziKB^pJGXK6pZ2>4-F7iN?lm>!oLY1)*%lM}!Z6JzQ_te9T8-F8ItAdOVd`%WG_*^v z?v$TdpB#=`s2+5ZoIJJ>Dsj!jeH`?>f42J*_Il5WgqoD>6)>33>=C|%gv~r?sHS0n zsBB2>`|jxI&AOP|;|*oQpbJ%3aCR61{jQL3r0R`c;Um{dt8^VnCly565QUW@+Rd?4&7OLmk9vjX|*X}FCJ8PPh za1~|PyqDLI0~~!$mi?RlADyaYmdS%B#RX!M0^e85%==SIq@JWz0MM!f^(tiR^&^S$ zJ<#i+k5#k_FI4*=QMOGUSmeG!WZpMI$-(u#u#G=ga;~>EUP`XLC^qY4&I}BI#CCG7 zv$(l*xG0IVREnC+=LzKAY-QuJbC)LT-UxkCEc0U3&)dwt#HC*|=fnzOpFN&R zzG825v>dPK&d1jxk<2|erewV4`jErsk|)e~FenS3&3RsKO}sNi10+?Twc!IQt8BeW ze`dcKSXI`$C!`c)11GIlHBv*Maz?`=--Uqf(+wowko}-j8v8970xT!*Hi=d_CF=;# z;7`wE^}t5tA1r`c# zk~XFF)%OlA?LPoK2WUFHcIB#V9Njg0S7tA)u;#YSsmV}E%LF?<3NYdinh&SG?I_jG zZF$_=^YG4W-3&^?!7rT>zStyd&ZCq67_`NQ#Rg}^**0&FIz`X8Cd3Z4yoRPRKLbpfF&Mep3>xrU2-=G_Mw&1jWazcFEd5 z3Y@|uNd`9#5KE}w98w|aodVg`&5x7lH&6+Y-VOAgns32W6lqbb_>qvSu=J;)9(rtbWc22JkM3g09<_Kul(9)(~aO^s!CJC4_$KINA$rGb}(OG=y>#n0yqf5lLa z%UP)`J|EphmPbu$znsHNXmZX4pqz&8Fh3&-B@*%r5U6>yVI?4n(F<+mUEq~*% zwu#QAYhl7Yk58btlL8&(#+QP>IheKDq$Skr5ncpIO+gJUQ?yy>{cc8ZnS5JFst7;XQrva`^MP zr_?zR2>&F*rL4KB*zUf3f7zO}30>^|PL&q>0&t3k=o^(k;*`$ck^Slsn{^we;ap*S zvFXi%`z#gt-6^z3Yd7fwHj_S{My)x?1?^k8{rvfHLb|nOHxegBZ_ccDbh)$JK)WPZ5zlp@w zp4^IXFFqP>C?e&Qd4?V0YbJF#c&U`})$3Lh+;z2k8e=DZ1>|#IRDi26lijumY5uQf z2RwJU`ESir5hhAXwLvIbw3~UE4={MK7sYQW&rsIO+u~D;=eH__exb3Qc z;i#{0x5IbRXyo1GP7lsPmXrE>@GMw4bCJ-&xQh7h+Rx6yQ0M8VD0d8ZPHDHwGt+N! zrfT_dBB|G$b9fdMT?wosTE$k<|5@@ct-MN?t`*?YV==BPtyW=o5?ze9*r@7#1S zF!)d4&A3tb2mW%V7)aUQ?-Tug$Y&w!9Tj|+pJUlK{|B0H`RK=e#6jadL&fvEKEeu^ z*~*i~R$w)lu5N9r?4L+V7|R~H8XVtw=S3ns_mpv%>$D&{pszEM|FkRL(j$>h4Zq*C zpX>Ed_BftyiBsA+GZC?; zS$?bDOO~;zs)rGnf9I zI#O0U?J9%Hg%PQb0awl);z|_b_l<6qFLVOg;qg#m;N!b5@OW+M)Z!(D{AWkGmiDH0>yCF5`xRC+qCMn_z6 zTKE=t3dH7;tK^AM4n<{wzm)lHD(GR&rOb@tfVj3&+8Ys7*Lzu6me{=}-)V1N{_tp^ zX>?c=UAyj}gvgJyr`Df{RDPwGm57=%Oia1U_k`}=eL*m!6vA$`78={HIUoj=<8%4I zd@8hF>8U%fML12j)=c(diT}UwvRG-rBi%JDu#aO(_zB;pvFS|uW;cjS_KZcnU5PLO z%5mpb;>sK~7s#pYx5w$nvfJ;;8DiyWM4g&+zVl8H1M3r`6SB{J^M)kFPjL8k?hq~`B8>&(KPc1TVq z+yF7joTk5U&?Y=m@~QsL3sO@}&WBI<2X(W|gJ<_9v zsLjZTh`pHIv3PN;(L22P$@`Ao{0W_Bh`({#GSig)i~u{2ZX4tsdIj(RPoB1xuA%+B zNTd0$IEA6<`E|!P1twk2VV2iC;Q!nIth;r0O|1bDo6orhzZbjvyu83fZ5OdBQRFKI zJ)kP;H_amx6m3)7_xA~o9J#HkNpo#|U*tM^wjJI}y?b-5n+Yf8R&tuAW$b{KM+pK~OQy z++*MIVLBK8f#1M@`jv_FfK;kQ3Y-y*sHU3%AJ6L=8QPl^1MFK{69|MVZxJW;b^-9l z(x{h1dyKXYK~e@j(A+%Fky*#Q98mE0gi#K*NjAj3DKyH| zb2?{NH$-B)7%9umyE%Cal{%ZMqrQ|B z;oXId9z22oBpkb2)rvqh9&pW!jfAgdvaHx$VetbUwOXQrt7``A7DWvi9ZM`JG01#vP_sCG)h|Mew`r2=@9F@Hvs99;^i=&^q z`TjCj=DPH!wMEDG9V78^#8jk4HwiLba@SIr(l{ZvFGB$6Qlr^f?~!5#l|OTIsw6V- z=RS{vG`~+)G4nB;%@wcmYN1|MMOZm6g6ly-$^;&ebPrxX^=~QxU2#uXqC2GD#FGJY zjTG8YUdj`&4Y;Ik74gi=6v+Uss~|l1z0`!{9k=lhn?9Zp z2boi_5Ad^6%bJ1hW7HZxfktuyZ4QHj4aKE}!Jh>Mb{#HW;p-NVgwA4XX8B@3LWRPS zT#?@I5s}$x?VB4Wz27FRd2VESLqlF&W=B@XI3maRGH|2d+2h?sE(x?Gh21A%ge zo{@_#IXR@@qGw2q8}d0U$YChHaP*uZqTE0=7ai38U9N0K1jIS0;*Yy%7#i9LfuAsd zLyS5gIoo~Qsb2il9`6>J%wH5g&RUAoj%uf0VUD~>ATh?9#*i9ry&&?w*y0x2yndoM z_M_<_?)1F>=u%*Nmx5v ze+H2;ky!Lh9*k4vx%xYT(Co4$hd~G?f0h`tE{3;ofKxY@lvqVHT;bcu2Ytyjk2}%U07VY@@HERL!$T8H@SlH2% z1Zgw8kD-F)ZIP-sd)O{4WkHd3**QKf0_Q6fi-!KN`DO3R`bUQ@xG}^6Z$dmto`FAi z=sU?fidSS_O$FjG^c(>pYMJE?&EqlPcD%7h=}Cd#v1ItW`$96IR@j?^g!+%5cAHY` zVbUdF_SLj-u79wAL0ji3zQF)73oa@sH7$soYbxH0~D1N0Fb*Me9;fV1io$re>Kss$04{k2_NlDIsg!zB! zOoQ`FUjjyOF@5bw9@<2&i+8tqa5M6BD|1*YWnHSQU2yu_{$I5#{8i0kl2M~kqhgOo zKo~{&W5H!($MBNW%_+<)xKnan6udO~`uJzEJbQ55hxxsyY{?5)W^7X*iwMZ3Xu(Uw zmH$np+wF1XV+&X?vu9B9yEoQ+a^VajF;0vQfkL&V=zpxg=Z=7(r=>)_u= zuTmHa$vy7V5bde#AATmsRKGrrp*~Uqd>FhuX)(EB;Gr7wm@|uVnrzKcxSb0+1ZzCJ zem3i6%pA#KS{@@vCF`^sy}RmpfT<3)*Qjf`RvS`w&!0_1Ch+mHmer|oSi6rJ95&m< z-zL3v54=GbGZl~4ADYqc&yOji_iWEN$;68uV=#T%gqhFGzNLCLQu5)1O3go1k-lxAW<6ys$#L#6#4f!zh|60AIv%BmD1U}ra3%E&9XSVx+awzRl zt>f6~PTXtMOHi7UjAZ$bumynBT~OoiA|Q#Nb|W~xLC7b?<>+9He`e{l2&)_n5i2j- zcifoJWK2NCh8(<-a{rQMI@?#I2AWHxKHk6q18Y6UxhL9}PPiG^=6p*ELm`IrQ1%NI zJj~RsZ)>7N!87(AuP0+M@s5~YEIe4g>^ymu<+{`^)*^=W%UmiQCLj&iOg&rRj_O_4Qs+qWE5Q z=n)0Kviu}aqUh0fE(@8q(mz<`c4+v_j%f;JruHL7A_( zun?(>_@|+T&@{^we9&Qaz>)CBREYuqmt6k?4<+Q8C#{kk)x)W@+U;lyMB{rX7Ys7! z#|$mKRn{dH@pdcFmsX`kz5HC8_gi_KYySVRa;Vt*71~U}?XnX_)6&|8i_t*2JVTR{A$&xCnSZ5%C-Pc4SP~-B3~;zVv-vJ#q#*8uWerK=djM)A zl}2Claxi2>mbG)XpKGH3cfZ@|NISu%4&XtbEUkXi#gsP2d&IA8F(N6%F=AmjW;t!N z?9m7~1sw2EN~V)57SZ+kHTHbv?z-}p&$iOi8wo`T*s1W&YOE^!HVNNMGp;QTkWwZh`sHx{J32a;5)n z@%dI5>_1x8Getp8Kx&@l0WyVj8i;gX2C9^;BZ~j^;GCy$p2?uSot}z+qN+^U-*6AE z#~%@@wMa+0~Ns+$sQcYoOLz2_p*%=R~g-~-wZ<9=p?=8mb_T$4V&trqHYP% zq!=(oMzP1BhauPu)KlXzQ6nWC{A)LF6RzgHRZd8zSV#a~)YHoVv}jssoqXNNc13vu z(Qyptho}>^XU3Bd?N8t-21Zu|K8K3dot>Sy((@w!eQzLe|F6pUM5$=cQW*XARMj{G zuTI9Ry|P6@GTJMG;oyJMCN1P|T>uEsq!^oM&#=otokN8s%og0 z>I&(=gUb>Xo6&%^7L`h-BRRqHM8*!*@YNK3uMgCIt=?{gHT-(MFa-uMJpGc!pLSks zLY(GnzW5bhPZ;kxK{mJ zxVDrml=Rjybj!W@TnmEXnrK%C^6wr{eOd?OTdglB#;pwx$aE; zvv@w3^8Tu56+C5tzxL?eg_D7^pG5;J@A`px7*90YC@+ETvM{Yz534o-9K7v39ja_L z%JT;BMW8>l3A9z0Wd9k~Q$!JuaMqj>x{SxczLKQF6VTXc0iX;-2Dsv7AN*g1CTI<~ zoZGly4loS$7Z6f@ZQH+jO}hIks7yxTey3J$hU_1%fif>&^LHe}D){cwggbv(wUJsn)hwLj)3iq!4gjT?CSw(Xnkf4#jYMY;i;8wjSBHsuQ z!>XeO-;yaEG{!Y?1YSEcHPJC1dQ4h^`SHq)o}{eqUSzx!a57x(|6Yk|%zv#L;k;6# zwrBz?rvOt1yzM-6Ohu%nyN!54`^=DBN&4z36qQ?;B}OCwYm;$e;8+_3Tv7=SxBPz! zRTvYf?kq1Le5A_-{;pF&j;Cd9Ada&>A-h}e1-?R{IjCnQj*kr)QTgKbKOAvCudQtG zo$Jyur~1GIFitDzRx`%#2aObPy1jmC*C@b*(4%h)Pn1e8{DFMXpxFr^IAWzZ$D<6yFzrCb3|30;pGvL#4PlS8odatLCOj+E{V~A7 z@Eq8q3=l_i%wk%@%8l{StbgI^ov?r% z-dcb=IF-J;a14g!@`$##^+&0Qvm2r|3PV3kdDegENfJrBv2s)i8wPdyYn%UIQmQn7 z2LIXb$9=e03B2+@GhY0;CGt7`VM6J^KocPlVot3iu?de;hA z%r3YApglTlLWZL+MKkKxoB(1_{HXUzo0KlCMT>_=h0y&6nuC=>tAkj6$D(o!?$~o9 zGS;iWXnQCZ@nm=j2v(DRkHKX|zyJ6+JFv(}Mr4Dm&+PfQVif~g~CtwV<*n+y$fxy^@#WtWxU4Lepxm-2_TF404fG8M4 zZ+i|>qS4WX>?}DsrYUyGZ*bO-Lg#RU=W?-5zj;7hrA$@m8geKJNL6BXj=xT9YyMR& z3R`9a^RYcD?e|@VZWG;h3yX>&U?Q#9S&bm+Lxh542n*&J{)exJk~#~4#8E?tB!vlr ze}2luKT%X>xP7)R*?H)2Uboaw4vnI4rvdB&$`;zTEgm^a*DATtg}mVSW7JR*+q6)# z%R4!YI!a)fVOgC?%J-yA2zT9|=@_F!fKM## z0{2GEqWC}eR+uEm{tp%ah=~GaVl0?A2p8#L5O^~?M}SBS30`3)U79$!jZ?Wy2B3!J zL@Zc-50Hi-_-y><^G4`@>23dYcAmPxx@3%K$Irl!_1$_+z%x5(X=(?XX$pX@XC(S( zk?o978h9ly2%}tbs5S08W~YBZAECkJz7hVr!&{)dl%xv%m_jMsYLZbd`s1P;zYL9b zh!qW3@m>NJpS7J4fCWn*ooAZNb}&5qn(~PZ)%dOE(mA!qxvuT9hJeYeffFVx`vp@H zL<9fMDtwMIIB)uFBOlNx0N_ayM{TF!S{KL$na};%JLWK4@qELn&9h7$wZ6f zO>c}xCWN0^tJ){znWmtH(y}J!47tm|@vSm4GDvT9Xl6pv`~P4RusCSdaRe7==B$g5 zJQi6Sj^)o!Dt}?GO+t60`X|{g7y33>|GpFoW?InNDAScCy|ka6c8r{aD1Jz&M+kcb zP!`WrD6LCz58KGj@iQ{hr|Mi5S^P4wkY%?03jDb?n6TmO0u&>ntzq#%2qGyLT?sm; zyHYV>{=iqI6As<**ogV@?rRX9d{MnV6;Y{4E&yrMhd+R|b7TE`ZcAVk!xtqB(Xc=1 zG_HyH(ezXnm@-&xlXTe7?6S9Ky!T2@wTm^n0&Al1s+I-w?$U@xjQ-2*58sKHtZUp zaw3AvDi_-UEHf~SU52$ry-;?hMjZ|S^8ayVUF0l!0kEV9@G6#9|7_O}wr8B#5d<<9 zUZ!V&Pbe^V2@#e$6$RQ(eOi#N6uO-tE_Z&SWD|wSv;*TP57YckJC+D;ci~KSaGvSx z$o9yasc=I&wid9@++D&D&`HBG;dQZZ_Qubi! z$QKM4yUxVNPG^Tb=B~^O=Z0EUTwFYqy|AuM57sP!!&KLViWf#k#)O1(Kh9kyiGMP4 z*kyc^fF35;Sh_0O%MeQR26P_%H%XbQhmOJ*kZu#2ofP6??M^i3U|iqA*V4+FpRAn} z${4U4krCl(1z-|!UW9o_{I8h2xy7)jKUU>5(1VHL=?|87OeslGaNSFB^DlS!JEFN? z0iAv%w!7SwmHn-lFtW`hbk^=x8AG_54DdI9FlI1IeDE4$eW^jU{Z9s(7utUn<*H3W zz?(D6#KKuHJu8+X5MQQxM@`a z+A$rehWD%NV7c}5I}q$|2sMvxJjU=O1vUB9>5mzNiwg&34JZ4`B)G*1g~2tEiK!4< zCP1lH;sll+>n#wxIUZsF!#pPSXwRdvfX4J_!f_E9hUQ77xzN zoT6sSwZ!x$HO$hX)?9R;g$}zI%#&Qb&-D2I_-OaS9SIN z<4#gpu(z!iMMA|WU{$Wd z#-0L^J^!(ND*uuD|4jUUSR%|6Y;6>b@6lh3qOt0&2?ZB$3aCLd5b`f&Z~{r`48O=6<@bSGKO+O&J;`lL(mMBn2-$G1&|I5Wi3*Wcm%+5H3d2z zP06t$I3+`lS72P1z**W2LR?$I!2Q>*znG-~+`DWo(id{M)Q0Sgs9@=$SVE0m377C| z)pA^$?W#6`Ac~0U_r!Yuve4yfaY~OgyYkxZuwSuEa@l(c#IhAgn&!D|ka>L~de?t1 z%@7O(`JE#ZBnQLyU>@ba(UsUq%U^!VQ6xi8M40HC~f>!VUpBPpr z3k+0{`v0=YPA5C)UHiS~0Uqhzr{4mMPA1bRFa z2g?tEzwo4A-l@wf2uQ9M7(K_|gip$Jsot>z)W^NivED9bv(s9C{&tHL)b$af zqF}pQO?1fu50(U#$5UMA<#}_~V)l4zbg40!rStYHKHKjH#uP+%VRqelYIy8G%eOWR zBnIaVUjHVXFa>CJa@^GfP&r5AxPFDPG==#m1I{mD03{ z{PQv!7IzcaNj}t+88Z;{Grr_kDPo0q1aV?^{3QiIXE#)9+Bb}=r3|~W|Y4Ad^j~=?ZwcNmfWr8Kcuz}D2DN77?sdi*GwvnYy*|7424P`SV zPf>)imTKv0ini^WkXa;9MP8W4{*|^T`TQr4d9TH+fV`ic=WA%w2{Q#Ihqtb`Sl3qyq9d#4Jl9YhKdSWlJyVjrlt&Bn$U-~h(P~7)g+d+WIZU7sR&KLk;4y0Aj#mHPcPs?B^y z;tNkM!fvvCxD30Q(nJ`p_bPNpA1M;A`{M54i_l|hxj^-BzE`lg%CfSJNJWncx*T#% z+19Ds;w7Uiv4yGS`9SCN)E5Mfr%dRr@R zl5O?8FJH7HW_}=#Ub_hse0dl84Y7I*o!d^8;<4)F1z$fW;&gvLJ3Jr=NqB>&wU5#L z4;FAIj*M&UVRvnX@IYsU%CwXrTX=lOl`|4 zmp1p=zJ9LGDw-o2(sI^WsDjb+ObpVsw=w9qcNxq2Jtd_hHkv0u?k1pkA_+2~Q_=PT z&-nP4=B9f6CN|#S;4t}&3xRX)5V}F`nt7#!ISH!-jNqi(MJtYXsBUHjeEd{3`n6^^ zr%^(pL2THa*;}|UL6(Xj#_rxXY%3%&8e-&RpY{`R5-P%_SNKK8=Ag& zq&KO*ZEfh&XCB&5TnFqouc?mm$sKkr$ek_?eZ`CCkbB(!Fw0?RctapxCqY&BHlXFH zz-RgEFExb_ZOOh!MM}HCTi$U|{87=dRQODa#=6wEcjhl^wPnRRi28y(UBq!Z8RpwZ z79)o{9bT9)8Imm+tp4Tt3`4X3P8d1&1bpBd29f^#_Oa0>Q6=ZEuy3>dc_g!FQvPrs zxRR`>R=LR#X$jWfmm3?p+&qX|@J>sO@#jwcpq23v|} zla!p+LnTymfhNLNdM2e@A8I43c=XqNk6uvUa&!3fJo_ra(ba;BOjRqhGIpn*i+>y-Nd`(2jvs&*Qu86=jkMgFYEU*_3mi)j%uBDot^1kyj78O) zm`B~Se{RdJow9U2f|DV$A+5*hE-iGcNmW%XGSt}?qF}I1X<+A`8@6lJA*)bQ%C(lOg&9a`dbTwiK_&g2G+H}1SNV&P6`)Ouw z&ht;fVTzo8SMhb-cALZ8dvhXzWz_!9d~QgDe9h3+-^UU|JdAsXiymwfCv|w4{GE}$ zfTp3M_|o_|DWSyVeegx@L0{zhjuts>xFz;! z4^?Y5_LN42s9A*F?lAT$RKwHetqM?>U_qYI9dNO_h8Lr$dsT>=sCNT$!5d_)d~#lE z-M;y=$FPTS=hHUSyXY_$Y>+&^eeu;VXrD zx!nZ1+%Ge!m9E*?=pDD*#VpA_m=brILolZ#w=B|yL%Vt_AdIVh=F zzC3T9cg9}z-Hg!tn2lADqS`J_PnGdqPwC3kEhqOi4sPpeNrjNVDA)Wx_KjEdD#-wz zGfoKDcwL2dkr7Yx-a+@uQVOS?Igm6sde1p$Ja`me#;{EHlPoh@-63O) zLo-AA&4aA(Goi|y8zGzm`*I3+x2$w_b)E@5_g@KY6S4^eFSBq3Jf?>^H(oh}+^{Dv z&&1lm@7bl)hW0!ft-H$VY%g!dak&31$==^y=o?wT6@FcegYxP{O!@Z2N(p%subuev;om^T&MN_j#Xl zp6B~~zt1`6o!Qvbw+_#wi2mGl(L26Z8uYxJyRrsOE0A@N$Ec^w9#vYH#FIHdLo~<+(1EVRC%x24nqyDa!>d z$`HiUtoI*QY@2vCJY7I@K!%D9y-vRwno^VZxY+z=MG%<93}_b5Ux!L|;i1V`nhvTO zmDP=-C@fD|Hj#y+juO!lN^(Z{r1*I114vSccdt>5cTyL4jewO4WR~WoV?QvLP-9aw zUC~-{z}h?OpVb%DUZhQ%dA`|Ynl?P-^65To0K?GgQHBrLb7qxg-rnZ#H~mF3)0Z;( zPGK9_`(2MqG_VM>1*Api&rRhh`(B7w)x_?Dw$Lbp1 zvCp$oSSausDXZ};z5qt?6EuqElrJAjvxL(ZjlB#zPa_HP-9j~K@q(!99EonEMWTP{ zA1jhh)YUM37Wl(xPuixUiGnY#U5eQ?}eiFT8XB7$sV}?*<3E( zzp0%HA8<{gLELTYS1!_-$ylR_EJLq?2@pV0VZ|yu^iCDV*?i}$s^wE+AQ6*oBeJWA zXRxKKMaEOTboT66QA#oXTB*})abEv@jx%B=&JjPuIL$yh!{CRc{*<)&vhRSi^T((q z3MD(@tz(EmO+dT>WzH6f#C(cDjekXq6<%@;(~uZ78GPMy>*BU34dR5TLKMFK_q6h{ zNC)c+WzIpOLV|E~7|_07prEWFXpJaZp13E+j zJ^f?-U^qL`WD-;8>N)MpZMpC#Vytg)IecC!frp4jXvY;?Ko6OVag&|3;40W0`#cLX z{(NfdcxB;V;5gtVP7s957k_KRdHaR@E!L3ld62r}mt88UuiB>bWEzB$ef&n3o7UI$ zAvDv-+VBg7iQ-938rb<>ISvqrgFwN->UF+ru*A~`TS|p>yNm4L0L;N1iRt&JrVStI z;hgVptI%Vfcvo|2|2PnQ7O0#^_a2v7P%KgF`Bt8}?_QPjARIM1CMY(;qdcA;UD1ql z4LFVIQw-O)ji6~-WIKs$) zzrDR9=0K8zUZ4-1er}nTINCG27Ov0o*gSs}Jsy9i@h*4+O*0uH>en}x>rEB$zVwp8 z-(1Q>r?MmLQ+$V&TJ(+ManqCeDn3q=hn0r_nrGRh7j6+Nt`Une4L}m(0El2)uK=2R z?YHu}-v#6}O!g8GF73SHcq1OMJ#mH}6`Rw^7c($qLnconZ&|yJQ=ObFj2A|#*Y7ZO zsFESm%A>L~zNWgkWiW96{?$FWG~^^+Al3xQwG9T(ILscGZ}i@B7kv=pUGws=QW= zXQh_yX$V(`>l0>l*TE5x5RKcII!6EQ8(Hy+8_DFX_1QT9Hxp>9akbrYKJ0v!3 z-1!7Louv#yiCo2%5QBa!nG?`3uZ^YfWI87|^J~ zKDa78RFINgsch)LmR0!$6gO&^>{@8>0jEz_kGq%yrD^<1?Gs`yQ)}h8ym!+Vbkv0} zXp7mz)_aNlR^q>QC<9rTISxHqscWYMVcm-Uua{86aht$kd(P--oxif(S17qcQQaH5 zLv^fj&3qSFc~T2xYa7WuK-@%tw9(|`Jy*7K!6(Lz*sG)Oka~SBi&ngu-gZzGxVv{B zltW3N-ze+UzvL{)ZJk~+9^US7VZjHXu%Ar)`lWU0w&waRNdIbE9GSRk z8M3&@&XxyOkYSwhryz4hA^g?W>m4tto+FMG9Wto>|T4|@#zk;6kXvXGwyTi_>b-!6vs7}Ng$+J%4?17LTj z)GO3}Cw3P)J9@J}SlL~lL~jE(gM#zgF)|GcYm~P=Lr(DlwxN;BDU%T#MOM;wp}*MWM~?5!B^PZ`47QN@ZULJg@oQU&UBCj@U;d@m(7 z6kzT;lx(a=20A5LqU1gd49F{&O&wy_m6Z?Inn}=5CL(}G%TRvZNztq5c~LCMka&#a z3;s{I&eTxp*m0~D=SMGKj1CPzlSjp!ufF-{N^C<@+Bze7h4WkgIc9BF!}u09o4d>_ zRo(dze+1juVTpDx@$~jx=~d4KOc~uuzf|LiUKfJw$lo?~tA?QT z)H7#xGod4|CA_^{d6>)>79AO212}lIn(D;{2ItEFpA3eZahEzU#{vOo4@?^|_?x9Z RK@)rf)6+K6Dn?uj{uiXq!D#>h literal 0 HcmV?d00001 diff --git a/logo/dampack-hex-icon.svg b/logo/dampack-hex-icon.svg new file mode 100644 index 0000000..e11e782 --- /dev/null +++ b/logo/dampack-hex-icon.svg @@ -0,0 +1,12 @@ + +dampack + + + + + + + + + + diff --git a/logo/dampack-hex-small.svg b/logo/dampack-hex-small.svg new file mode 100644 index 0000000..0686aaf --- /dev/null +++ b/logo/dampack-hex-small.svg @@ -0,0 +1,25 @@ + +dampack + + + + + + + + + + + + + + + + + + +WTP → +dampack + + + diff --git a/logo/dampack-hex.svg b/logo/dampack-hex.svg new file mode 100644 index 0000000..d4334ca --- /dev/null +++ b/logo/dampack-hex.svg @@ -0,0 +1,39 @@ + +dampack + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +WTP → +dampack + + + diff --git a/logo/make_hex.R b/logo/make_hex.R new file mode 100644 index 0000000..6bd560c --- /dev/null +++ b/logo/make_hex.R @@ -0,0 +1,130 @@ +# Builds the dampack hex sticker from the package's own output. +# +# The sticker shows the cost-effectiveness acceptability curves (CEACs) with the +# acceptability frontier (CEAF) above the expected loss curves (ELCs) of the three +# strategies in `cancer_psa_data`, on one willingness-to-pay (WTP) axis. Both frontiers +# switch from Radio to Chemo at the same threshold, marked by the dotted line +# (Alarid-Escudero et al. 2019, Value in Health 22(5):611-618, doi:10.1016/j.jval.2019.02.008). +# +# Run from the package root: Rscript logo/make_hex.R +# It writes three SVGs to logo/: +# dampack-hex.svg the sticker (CEAC/CEAF and ELCs) +# dampack-hex-small.svg ELCs only, for small sizes such as a navbar +# dampack-hex-icon.svg ELCs only and no wordmark, for favicons +# logo/README.md explains how the PNGs in man/figures/ and pkgdown/favicon/ are made from them. + +library(dampack) + +# ---- data ------------------------------------------------------------------------------- +data("cancer_psa_data") +l_psa <- make_psa_obj(cost = cancer_psa_data$cost, + effectiveness = cancer_psa_data$effectiveness, + parameters = cancer_psa_data$parameters, + strategies = cancer_psa_data$strategies) +v_wtp <- seq(0, 100000, by = 2500) +df_ceac <- as.data.frame(ceac(l_psa, wtp = v_wtp)) +df_elc <- as.data.frame(calc_exp_loss(l_psa, wtp = v_wtp)) + +# ---- colours and geometry --------------------------------------------------------------- +# viridis option D, the palette of col = "access" in dampack's plots +l_col <- list(bg = "#eff5f2", ink = "#440154", rim = "#31688e", + strat = c("#3b528b", "#21908c", "#5dc863"), mark = "#5dc863") +v_dash <- c("none", "3 2", "1 2") # one line type per strategy, as in col = "bw" +hex_points <- "86.6,0 173.2,50 173.2,150 86.6,200 0,150 0,50" # pointy-top hex, 173.2 x 200 +v_log_loss <- range(log10(df_elc$Expected_Loss)) + +f1 <- function(x) formatC(x, format = "f", digits = 1) +wtp_x <- function(w, x0 = 34, x1 = 139) x0 + w / max(v_wtp) * (x1 - x0) +y_prob <- function(p, top, bottom) bottom - p * (bottom - top) +y_loss <- function(v, top, bottom) { + bottom - (log10(v) - v_log_loss[1]) / diff(v_log_loss) * (bottom - top) +} + +path_d <- function(x, y) paste0("M", paste(f1(x), f1(y), collapse = " L")) + +# strategy curves, frontier line, and frontier squares every 12,500 for one panel +panel <- function(df, value, yfun, lwd_front, sq, x0 = 34, x1 = 139, lwd_strat = 1.2) { + v_strat <- sort(unique(df$Strategy)) + v_curves <- vapply(seq_along(v_strat), function(i) { + d <- df[df$Strategy == v_strat[i], ] + d <- d[order(d$WTP), ] + dash <- if (v_dash[i] == "none") "" else sprintf(' stroke-dasharray="%s"', v_dash[i]) + sprintf('', + path_d(wtp_x(d$WTP, x0, x1), yfun(d[[value]])), l_col$strat[i], lwd_strat, dash) + }, character(1)) + d_f <- df[df$On_Frontier, ] + d_f <- d_f[order(d_f$WTP), ] + front <- sprintf('', + path_d(wtp_x(d_f$WTP, x0, x1), yfun(d_f[[value]])), l_col$ink, lwd_front) + d_sq <- d_f[d_f$WTP %% 12500 == 0, ] + squares <- sprintf('', + f1(wtp_x(d_sq$WTP, x0, x1) - sq / 2), f1(yfun(d_sq[[value]]) - sq / 2), sq, sq, + l_col$mark, l_col$ink) + c(v_curves, front, squares) +} + +grid <- function(v_y) { + sprintf('', + paste(sprintf("M34,%s H139", v_y), collapse = " "), l_col$ink) +} + +wordmark <- function(y) { + sprintf('dampack', + y, l_col$ink) +} +label <- function(x, y, txt, anchor = "end") { + sprintf('%s', + x, y, anchor, l_col$ink, txt) +} + +hex_svg <- function(body, font = TRUE) { + c('', + 'dampack', + if (font) "", + sprintf('', hex_points), + '', + sprintf('', hex_points, l_col$bg), + body, + sprintf('', hex_points, l_col$rim), + "", "") +} + +# ---- the sticker: CEAC/CEAF on top, ELCs below ------------------------------------------ +top <- c(41, 75) # CEAC panel, probability 1 at the top +bot <- c(88, 122) # ELC panel, log expected loss +d_front <- df_ceac[df_ceac$On_Frontier, ] +d_front <- d_front[order(d_front$WTP), ] +wtp_switch <- d_front$WTP[which(d_front$Strategy != d_front$Strategy[1])[1]] +d_front_elc <- df_elc[df_elc$On_Frontier, ] +d_front_elc <- d_front_elc[order(d_front_elc$WTP), ] +stopifnot(wtp_switch == d_front_elc$WTP[which(d_front_elc$Strategy != d_front_elc$Strategy[1])[1]]) + +v_sticker <- hex_svg(c( + grid(c(top, bot)), + sprintf('', + f1(wtp_x(wtp_switch)), top[1] - 3, bot[2] + 3, l_col$ink), + panel(df_ceac, "Proportion", function(p) y_prob(p, top[1], top[2]), 3.2, 4.5), + panel(df_elc, "Expected_Loss", function(v) y_loss(v, bot[1], bot[2]), 3.2, 4.5), + label(139, 131, "WTP →"), + wordmark(156) +)) + +# ---- small version: ELCs only ----------------------------------------------------------- +v_small <- hex_svg(c( + grid(c(46, 83, 120)), + panel(df_elc, "Expected_Loss", function(v) y_loss(v, 46, 120), 3.6, 5.5), + label(139, 129, "WTP →"), + wordmark(153) +)) + +# ---- icon: ELCs only, no wordmark, no text (so it needs no font) ------------------------ +# the icon keeps only the strategy curves and the frontier, drawn wider and thicker +v_icon <- hex_svg(panel(df_elc, "Expected_Loss", function(v) y_loss(v, 48, 146), 12, 0, + x0 = 24, x1 = 149, lwd_strat = 6), font = FALSE) +v_icon <- v_icon[!grepl("^ +render() { + python3 -I - "$1" "$2" "$3" "$5" "$TMP/page.html" <<'EOF' +import sys +svg, w, h, bg, out = sys.argv[1:6] +s = open(svg).read().replace('width="173.2" height="200"', + f'width="{w}" height="{h}" preserveAspectRatio="xMidYMid meet"') +open(out, "w").write("" + s + "") +EOF + "$CHROME" --headless=new --disable-gpu --hide-scrollbars --force-device-scale-factor=1 \ + --default-background-color=00000000 --virtual-time-budget=8000 \ + --window-size="$2,$3" --screenshot="$4" "file://$TMP/page.html" > /dev/null 2>&1 +} + +# the sticker: README, pkgdown home page, and print (600 dpi at 2 in tall) +render logo/dampack-hex.svg 240 278 man/figures/logo.png transparent +render logo/dampack-hex.svg 1039 1200 logo/dampack-hex-600dpi.png transparent + +# favicons for the pkgdown site, from the icon (no text, so no font needed) +cp logo/dampack-hex-icon.svg pkgdown/favicon/favicon.svg +render logo/dampack-hex-icon.svg 96 96 pkgdown/favicon/favicon-96x96.png transparent +render logo/dampack-hex-icon.svg 192 192 pkgdown/favicon/web-app-manifest-192x192.png transparent +render logo/dampack-hex-icon.svg 512 512 pkgdown/favicon/web-app-manifest-512x512.png transparent +render logo/dampack-hex-icon.svg 180 180 pkgdown/favicon/apple-touch-icon.png "#ffffff" +for s in 16 32 48; do render logo/dampack-hex-icon.svg $s $s "$TMP/icon-$s.png" transparent; done + +# favicon.ico holding the 16, 32 and 48 px PNGs +python3 -I - "$TMP" pkgdown/favicon/favicon.ico <<'EOF' +import struct, sys +tmp, out = sys.argv[1], sys.argv[2] +v_png = [open(f"{tmp}/icon-{s}.png", "rb").read() for s in (16, 32, 48)] +header = struct.pack(" pkgdown/favicon/site.webmanifest <<'EOF' +{ + "name": "dampack", + "short_name": "dampack", + "icons": [ + {"src": "/web-app-manifest-192x192.png", "sizes": "192x192", "type": "image/png", "purpose": "maskable"}, + {"src": "/web-app-manifest-512x512.png", "sizes": "512x512", "type": "image/png", "purpose": "maskable"} + ], + "theme_color": "#ffffff", + "background_color": "#ffffff", + "display": "standalone" +} +EOF +echo "Rendered man/figures/logo.png, logo/dampack-hex-600dpi.png and pkgdown/favicon/" diff --git a/man/figures/logo.png b/man/figures/logo.png new file mode 100644 index 0000000000000000000000000000000000000000..cd79329ecc4e25573f2d431f879d5ca4f35880d9 GIT binary patch literal 16772 zcmY+sV|1ilu&_O`ZQGvMwr$%sXEL!TwrzE6+fF8#*fzhO=bZDd_5JEzy=tp=^}efW z*L6oIDM%u~;=+FW_6%(^TVO*Xh3cGa>{48 zz0b7eliAPzN#dB2pV7Q3?P}v}v2N?}o%Wu7;g5sBrkpS(s)Z{s^bEWa!N;bIz`>>* z>1|1Q@okgg7%JL-EPf;TqkyH-n4~&Q?@a*$pZF7I^_SJ%A?gG(QkWsy{H;ZDQi?Tx zV-s^&PwC$2_5J%wVW3YXIDBFd%&J)0``BrE2*tQz<6TtoL-x_GQFUEoQfK!^qqc1? zGM`*8sAaz--CMX=fqg`}x4>)X%*+%^TW9Bc1pk-+2`To^5E!kz`|PK}u-vG<(AC8( z5)2G0bc=mM!PSWuFEzjEMCHmqQy`*F%phs;K1n_4MI({$M-HgzTvcy_)WL8)DKS1@!vf6DmermrBhdNdqF^a59m7*@+`$h zMb7klKGSpvGG1f`!DG3w^tWGOU){TEP1?LMKX`|1PlO6{*==raSoG?WGWf|3Rb%va zLT#|Zf|X7(x?$}oX7|eKsS&6I9je?9ZWm3jN1Z_m`^wB;d%W;-7Zpyx5yti`hLZ|O zC@9Rd>u9L-=x^zTB}xDjgkN0?fd0X$Nx!a~?~?AeZM=_}HZ3S7IwH4*z)+{)m6g48 zHeu$)E$i)d?_*YQk?7Tl<2#6~XSSpZX3tH1YNLKmNU_&z4Fv1sPh&RV++ha3x;nE9 z)jBCS*jwOo@1*y!7bM+A7W4IQnC*)TrE>cPXu~VF+9tYVE#YvD9WK>HeuF*r`*9T>_3lCT$!GtQ}avJB~QKfmOLUxn)EgKCeo=ti$7RoW_kl? zY{f~PKJ%l}r7JGXzVEV!vv`+tD3)NPGUkJq>t*Qu6}JKpGR~23Q;L^+w?$xoGj8h7 z+~ms)-#1suqSbwb`u#T2D_e2+Z#uaHzxfOnYh19gQd$s#lzpb5sMHW<8b=i!=X5jr z9jpW=4n>8EgJ0)gD{yq$fT+eYdlnYnuN%0i?ZcH{tc19fMqo<0Q;Jvv{p&mG^QZ0W zx`h;DqlpR2&!j7~cYQLXHP`NqU)+3TU;Y-lb^&QYnK<1XJu)JlSvS)Q!6>-Bm7ey@ z1-q;_1Ba^N(g#Q0MgZ}FUhqx1W7|7dKx4lhsy5+9DM57l2iBFzQn&0VP0}4t?iyw$ z7Ud2lrQ0Lv(d@UQ>etox91_#b6Eff4aTRzHVkYIkae*}^U^rxVKY+4GlshFRe?_87 zCxZoLCm5fFN|Yj>JI#%bzIAfcGj-~plZcf~oUCMH0WBkEjtLQt4v+c$qs7ltN6r23U=o}^dtfVbZN-2Cp z*~nl6G+4yfjdz~cj=x)_Q0$pgGmCkEA@NMbJq;1c+PidnWlp(X_Sx-2%ivV1`@TBD zPKyNWW~B{GCYMkc0p<-8{j?xu?So$9QUQ*h@_}}jJw4HlcR2(b-S;;ua7%r9;L^IR zmp~8g0xl7^PgM_s0Dz{Y;%cVC6C=RT@SF{V9W8Kj!E2Zgg-bA6hM7b#c}9=V2715J zjk&CTcB%U^1%sn{XDixA4aqF6E^EjaEJ3%ja0v1Y`x^-TYq^fOLeL%=GxD;ANmnT? zHhj=It78{c!n7l?`hI*$lR>Xv1{4#u*X5g`C_ECiZD7b~3exQWE)=>aGn!&!=KF>T zf}BdP+e=}WR%cx8)XelcDyu}8aYMxVyIjl#a6ePlFuCPqdKV8@_*ZR11SW{eIX#!t zm=jxoy9Mzb;)uVbSoOuI0wBJC>_lc4pfAn&l1&8#;K|G_7q*WQwBrvBPg65E%FM7= zz6N#3nyO`jQG*CF=E-U+2QbVBVG+p<0jRY=7HFYz7=QEo*#|6}(H?Vur?`i2d`V{u zB#P=atR4kWbAhO48ix1WQ~6im&+1Ju#hf8f@zOEN5!vV#Eny<*efQre2m+t93xNNy zLI8*h|EDLLDk8#VNciD|Ypa@d6N_#_z!0n)vIsj&A!c1ShoH?xVe)$p&M>f%e3`oQ zM*_4R&-4sH3uqxMyp997A8Zu>Ww;Eut{A(9uxgEl*)%Jxd<&?DYG7LNk zod2z-l=#T~RgF81=atg!IWr+Bu7RJ%S+Ea@ifO#%~uznlQ9~SV1|r59$E?`>vsE-JI7s%%*lm zgC=4*344&BI*=FK#Vavq5X#cA*x}zBcC~ySyH^SN#kANbXe3HasVG?6Y?vg%0L{OL zrE9~v1kB=Y0RayTk7*Im>USlM-{}^&@-k(tYX=I#nrZ0TKioS!lU*c7zvcrFw~Ioe zFw!{)(lbT^Y9=#y{gAVOs_}(IE^a`xpiK-kw26U3rq?DjcMq~#9kHz0aU8jMN|Dx> zL(o4S1D%YF5lwhY3z3N=PifvJYNh_<86kKT^8uB)@dbyF0}_)GI0d&A@2pnBk+b*V zbxd!#f3s{E^>auiMqwK@3~nVH9Mm*$q-J)CzX!+N5FAntMyYfXD3H5iJLiBhN;GceZBPl;d+1)A+cI zXUP$I^ATFnDb{IN?O}A^EFnp}R$5_OYb!7Qm^MvaMxOSis(UGXq$=$uPi0OV-1#Yz zYCdr#a?98i#%4B**0GZPy~9;hLf5*}vN)=2#gkt}R~D!@FBIO`oByz4q6Gi?1$Zpc zidI`uK8MX6oXTpJ{P-kM4K0Ihov*xoz+EZQzRQzk4oxe*9h|Y_bgn zag%C7fVhr4sbaoQ9b9MTCXLOUeJZ!RIa{D zH+n-1g${%qif$y!Wiv|J^CR|Z$GWgoU9whnu%9-JMk4XbrE80KPic0`gheA^E#odm zegQ|1cOn$==mZP{1jcnl&y0Dr=Sky_b^eM0w#$6THdapO_KY$@f zi1UZuLryIupbe9dCGJtU6h8(RJ+Na%)P1~ladp1_y1Ww#2E%Y7K`NM)Vw^LTg0pmC%K=a)Fs%*`rZ5>gOVilcM+b8zKftbn zGYD5q8?_Iu({44p7FjdwIR%sNaD*1*ZS zi@Sx$e2sP0J8kz|Z?!4@MbYYcyIFAFHsQ5yc9jzP(Z_7zMWMz?0owj-{2VPwTya!h z5>He2naJQ(eDU=+(^`<^(uYq6R+GpQ^M}Q8(L;u-?AY^ny@gtBTzyUqWc@X9&UK93 z`%5kW@h@YwCLNGej1-bLgilj_T=>Eb=y{)>=La?^grwSxDKOU4LUmZL`>cHIYqKk! zS}d$lXX_{RtaUfU6K8FmZC|eoX@@i3dMZM1fFR$`P4`cH&Oj`!#$W#OBEuhYgWGcR z-h6IOzN=hXoL<@Yt?-iP90Ma(T-IUuLx$w29c z{&Lx|IOYsL_N;%Q>L7szfOw$XE0t5`+?+2 z_Hf^Uq&p!J!6styxXb9Mo~LuSUe~Au+??j_s2}W3x9r_f%20}{Xnr8Vd(z4zOqT$^ z;`%md3O~EP{Se*rffom5S`c2|Ua!aI!pGvU>+0>2L(!&erE6?!A-5P{iBr)#cD*6p zi#OzKU3-lqpmS42mZm4rn8tyyqxd1MFqKSWQ@7|LtyAj9Xkts-5S2LhcQM1*5b=0! zJq?>jFZ2=h(J#H&)ZlG(OAlM<3flrsSIMd6_Q@J|%|j#5!?B(6kG2?(YNQ&jP&nRN zw&xr{!9USGhmI0qAog?qVwP?M%P{u75TZnC<0|_l2RA&a0LH&zZkE1I5AG5|REBI~ zsNUD;7FfcY^57ds?AeC5fojw|6ayVyf|S)3EGH$HqUB|QTWx93T4vQWDy$lrUD8IN z%ZcMP>@kQ;Ic?DsDRap8`0e)3cD$UAmpp9f{zOOvy``@%Bjw+kn$OW`caU=mnCKV= zXiM>k{x27xsrXrT_dI(9m2gt0y!qI+K)~7K1lZ(7GL^BK(<%m@34hq-tHk zdk~+RM0}$Gi_L%pbRqKlif?M;_9=EUX3aC$Iw-dSvg+Xh3LMwgMGuv(fDoRDO7C2en) zGVoedt&nZU&JU>dip1)xWd?rgWK7hO$FuZ-Z-Sfg3ty4?P^mfT55wk2f1`1$@_Kvc zXRO=3uE!Yc%P{RcZyVxD_ku-Z$=K*g2_)c+y%PSq|0J)&jC(#7GfOh-mpzrvkbIb7 zC);y&P1Q-*!qi(gP}ca-ASdsfBv_X9mV0p}M4C)2o)Edw!0SbChEK&6(hO=TGi_2e zg3YD1YWh?QRex!Smmna~(dDpzE{Sr4289QQckg;S(1odsAJ#uF;#RMT6)KxjO3c7b zAew_WdXbt$H>LsV8wK^Fe!736bQoL@=_%?8O*@06S0`KP(rsEhDLHjlS@WSuA;pk# zy%Q_wQ{kiaLyoj)w167kgAm-6_4}nOB%(ea=-q?~`Igmco;`}yG{hhqG19M2RdKC{ zz`(zMinUN@Rj6R&#G;3-L(F=<_pfhig-`wEo=3>u@DvgX7u*EXbN>Ft#bVZbFXiZY>vm2qa+qXpe4gAf}^NlX1T@8WLOdRC^? zRDUgLSZc6u>E&8`p(q$+l!@HbVm~o8Vo2XU3e)i0N}KJ;>CB=Q@mmSCij%4v*j>uh z!K10Ub1Pj(6w75HiJIH%53CvgD(S=qxF4eFVJLPOzc>J=i$-UO9KF)J;q{lMtskb^ zR8nN6{nRZen*v>9zh!>UttfX<7qvEFYggDk>%J_tc?4LH)O@&t{e>cP`j9FL-78n| zVx+StIVqyqSu}nrqdvqb^LHW$d(o&PEamo?@~cNK_8dRe8(mKKV6`{8GJ)gO z_~o02_xbVxj(SBYd|vV5Edw`hY-=8sJSX@QiH$fVJ+1i8tHn6mdkTyb0-^#GJqe>% zvw|>oHjqg?1a#WO5}OsJ-rDOem0u^WxdKy5x?S$x+MGsZdy{TYvoP0H&<&vKS>GVj zutV%2gjWW!IZ;n#H&w~%3h(Z(w&x?D@n=&bqB(^9ouR1~vf!?SEgNZz8b*1n)tN0a z{DsmXrxQY5dcfW7StY+ps7=n*M_kMYm96D|G&Qt0u}Ona*da^Fog^jFow2vhW2dKm zvL%?a1qbZW#Mhzn*Qf8?*DtwBl)3`FAH`3p##I~GZgr}fG00JaN-64!Klsjg<@fk6 zq*kihu8|)80u%D13}@(Lk#tpkLUHdQBAgC=W#i%nZbnm+X_UaAU!k zRZvDuUQia7HU6S*T>O!q#c9UOT01vX3{r364iL!o$bp7|6LE?YYPukpJm@UJ;kYuh zB&pB)mfm1Oj)iH|tJHiwA7!?RjZuDcCn-nJ?;kTUF<{8);Pg<0&ro~~XdEhfJT-#~ zF^epdD*l;T?Ys|ex)>)7^Q?F2{Ve;@=1`s7hK@-dPtH{95Hai<9YP{KH9C@g48IvM zB_yYZnN)EqWf$MW`g(sI7S+g5HHOUGw}KP<=-yG{7Q339#p@*#SS?uC6-C!#UFhrJ zNshix)?A0&pn@SnScNg2H0DHC>7#dQ>ZnPvLuXyCg%Xhb_D7mloD&IY5(aMX(J}E! z%%UPni}$xFiGl*On^}2)4yrKRvxPBwCjbS? z)1dD=YJ3HtY$@eZd{yp4t@Cf-&i=he{<9q17{?sLNnIw z9~X(vF)+a1kkB8^?#@xmyEK=YoSKWmziUuc+#)|ba%B&)8$w-`MKGOXqX1{v+0z~w zLl@?R_fVcWlYj8WwRHK%yFyR&k3A~eIm&baou`;%6u@EX064GV+>u8l_fBc{>A>5d$esCe5 zU~W$RW9>M2aJ(NBu+O?0K6JdZ3u;^c>WH75Tz=CQOfa^+-OnZ&_4BiOIDxUgw8*Do zLfS6vGOMZ;9Hm9Y=~Ton*@#B1Y%esdKoZNam78-4`!{7XE2gL*hOd2}=J(!}xQ(Wp zz>in7zKI0I;T0EBml@oZV?`ZjovF_eQ$a z=bo>klz^kvVWOLDN{=+I6#j0@Q%k7&1t${YRmAV(eixS96@5dKwJ#41#0E z_~>9q`J(eH@o4ERngZ3;?c#Vq{;3JhLsHgA&eSOM2AIs!lRMPB8Zsk3HQ+&UfONUr z-#l~E+hAD4(EX&JFS6(bOG?S2(hK3xRN{2L6-bqG3{2 zvz3V`>2_A1>B)-^g()=)s+zLE3wDW9HD}}@pLy~nsfgUJ&Y>isQeKlu0q~e9?}#D*m6km`x)_~WdiCY)U2o^{(XO!Jp;A&!%$D5tEY$cm4TK~={oB40HVJv8ElZ0}#2!;O7ZY2_;G;jxmgsv$Sg#DpP( zasu(D#=xSEPgVXVu|U?Gf-TSTG)!y-68QKRJdMW)NJRBZlQzNplm>)O~ zeLDa51>A1Gj)l2U;9pN1kQR$jz>EiX!%vez{Wz%Ld<0e1m$r6(RG1Uo8k|nAcRB<$6J1z2+Uyb$w;g+m0luQV$nUCny5tIK zx*@AvAvZS_tUg_@1N;V=MQtgfn;TP52O7Q3EThx9!OJBx3@q1>rA>Rg)(^Cs8?MZx zU#pvXHR{>xH0Wp>)rv#Uga+bIo`y>MCWlzdEb1H4c~cKK@R4 zYbt5{h4d|h@FjgCr=O%}p7s9LLZs}3RsoA4(H0Y=^l!9ZTAR@x?rp>^O-ULss|iiEQ`xE=9%zh{Ei`;wGNNH!%x(xr9e0_mEA;MB!eGwYEC z^uzk6RP-m$Tb?npusgWRz6>~FNl~l_g#G}NnN>~Lm<6Z0-PQKcIAjyVW@$k(C2jN?XeOLMx4&@mSes0VM6Bj0D7eBpSXwO=$oqIaz*HWj^hI}Z*jG@77%ND-wsmpx4sqo1%VZ`$~a z-3uooNx2mq`H-Y_;lj_S*kWnOWd1HJC&HwW_&siRgzYse5VkMn}7Lth-BuySsndu3tPn-PEqf@%j__e!6QN9X*`vKxhhLjY3Br?@JxH;fCD4ti9ri_bOX-}nAK71>0JxYhDIj_27)IS z7pZ$rsJND^AnA(@QS}-TwE%h;+Hze@s*Uf1Nd|-8${^mwECZ`#U(tg7b1Q8Q94ICG zBu`bF1y{X^^#-R`-`5oI^#YnsEggnHVrH(bUWujbi|WXBOYkmsU3T{X1+**%Z%2@m zgBI6(<=hvAotly7SY~O0yRw(Uv&d{-!$*9;4c&yzbIoaKXmsXfO`P6?j-;4Tc zWoWLXW}JNS=~0~4TRG?B2fPGiUAv4T3!U2{cEqujx+H1JC^t2L6Cu12MZ)dtP0%o@ z9Dh5@t2vM}50$9}m(LQqiDwWvkWWi-`{{4U@EQEP1@EAu2)^H9!< znHB2EKR!Ne}iR zIyL{aqu{DLhyL|VzweOn7?u}{ca>+YcXAV6GEfj==tL!n(uz^WL}Tgyasj*4sc??U zVZdjXLe3*$jR!JM`P|9#6eK6R$wEiP?;vUaX#I?dYT4_JBpV&0{Yh(R z;BIGIeWnIsb^KMJvo)V$r+0cn;)#O_>&y8@`?u|-qnBcubWJCO1VNBv-QdY7jAq zzpP=hs-;cZJA>l^SVB>Un~y1|)lJyD$4>}1Y^mt%#CK_f9fHN!H&d;UQ4Uq17KhbJ z-Vn*B9`cXc7;Tmd@pVI@9^EvXuZAwljT4tOmL+rp#B$V8kv4hM{e3KF9zfEUbZ}#Y zO|g*(Ju!i5?{rnakL#@2_Vx2@^*Zy2(UyAN`l&#OxH&vKNd8l+i5&sa=>V{0}HHqv;4rNBB8wFCqd68bykwzYyv}=`>hDB%{i~hg7XN`LkNkOx7-LiLe zuA#=x;o3&8^wf&A?g{Qbe776kpY#pvso*im{OTf;XK&8)LmG0|)Z5=Gqh+HJ$8(OTDc0!1ibS?9I&A%M5Q`nIl^~sj<9#1fjVNn+m81iu=oyeTPk&TZ(I>)|eMZk=BzpGaGPhmh z7?+xPv2#_EEDd@C>h%wg0a2(HsKN)JC#-ml4MpYr;5Dhd3vKD_-b*f<9OI`Ruq5yh zPf4dz`AA`r;zY7bPr<-|!K@#}+uKG&e)zk>mG;6e`QG_d-Tz|{KkJX1LD%@`W;CKj z+l?JWx`EnIvNM6B{wu=yX&i^~k7G$t=Z6>+Yn`=duDwVh)@UeUxYe0)MbjP+3y4#w zY8VVdd60PxkBph53jcB`o#bj^BK)4_&{2R+#R2z$6p^9U5uYp_45a)?Sbc4bkdUR} z^$!XU%S9+%nHS_uKJTa%Z?;jk(_)fM6EI(X%~*9nuiPNKJ+LC;r)tqgXF#8Z*>CH- z*OI`WQk*YiNR)ZoMFv*%$q`%WK{!CadpS7mnyV`93Cg7A&yo9Q6VUjNRvo;swVYZ1 zIpB9e8IzhlPgv6qU1bkmHx3k&6<99$1+6ETgpL2>>ErFpR_s07eqy30)|YWe&EkXz zCOc6m*^rXDHX!rApVIypB>_tu@-TabSW}la^a&sPjpV-Gt-HWfUuQFbQQLhOj24_F zK(E))@f|Tl&NmXP7TL4&b^+earwA+qPgMpOhWMlf+18u<;i=iELO8APemTNI;TU&{ zSPn^B%{J!JDGnlwmDnBsafVwvS997N=up5cX^&mY&(G%k2U{*%5C*jzJoMMUXl0HT z0^gqk@hycQ)I5*p0-ixx$3Wu;>aeE_uad{-t0QZw$f2O!+RL8#6Q;DGmGl83RsFTJ2yXCy{N+7W@^ad+2Pf z6%r()9l6=W2MpSnxj_~f?|!e*<1*LYq z@FLc7=Ras&!SG)XH>~|cWfK425E3OR=mb3;aA`KbrFO3E5>7`@9+Mj2M_6cx3rhmL z6OVpHP73n$>p`OW6sUYi1H6iK6Lj88x3BCFor2~6kANPQ4k~Tv-ciN^-|c`5`g?_S zcj?-9gEx1|Ux>0cBV|Z^Fk0jwlU?kZK^fd?SE^+FSwl=abCwT^vkuX9MIO zmUX-q0blffqiV{pAMk0Qx1`*f{X?~dPGKO!(aF{7jfZTu4ssH`ZtriZgPP3jW$7e` zGJHNU?oh$XczrNwhvQ|`)ybZ1_>^#Xz$-XWCs4+=5|MQy3>~=tiC|&82G2>3zq6^G z)xX-HvtXA*=7DeK8{#>bGGp6l8CZqRsZxC~YW|j9g=dju92K*d_UTo zmK*=s*y3QRfp48{4Jx$G)A7===xd*0A*~5RK+@Mf3tu=)HjBWbxjSnwW~nEL8_|vC zd&2kh86&18MSuP3*w~~oAKr}`b>R5X=dbv&ySRlpPZnf576OaLJW%F*8G~9dgTtdS2a^~9<5hT&0q{v9XlzO&NYP$q5LTcl zL$P1)Ylv<_rj9;H{@JyHjx7|cPN@C%rLc{g&ys-t0N+V=z10OXh@jD%r!V{Cg`Qn{ zP)vB`*XGV7e6kzCdISu172K2&G|)-^f1Um}el=ydyY-BE)|Lm1C)2P~NN?A=ms7b~ z&Z`rt-@14IxW>3S_rC0IH9EeIH++Qhm{faqG%&Vh6eVR4Ib0coGBv<~eEQbHb2Sk@ zGYGs**SrXexrEfU#CyNVm6r;a+qJe_Eh=(id4GuRqxbkNm-fz>~}j z4G_Qdc}@+=H24XO#acG!y`nL!d6`Lyo~dyfXq+6aKb@(3Y&Fu+Z% zitW&_H1ukI5SwAynzc01N-PTQYWU)rw(H8f(McuA*BeGu9oUD^Bh@t73d1`kXD(Dl zb>rWj)Y3b!UQsL8wGqw5OFmgh-91ImlSO@lX2GjjtTR9K5EJ7GUH(Go(eY7 z*J77)WYw-SP15g9*Zw-ReziDN)T(>(zqCED+#M7PfTF#f`F!zQim8cDwbF|d38{NH zGi)VyWN-7{#cUz?x?}tE_J+`O0!(Agnw;o%uZPHgZD}82fAF#4%C!`pDdf+X8mDy~ zb2I7|G448&@7_ghJNJlY+h)fd0!#z54u3z+ykBaAdm@RM(u_0o@1{?jzU#)O-B8P+ zBwmSJJqH*DuR*1D18`-$2Od&@x4QHh-D;Iy&GmO9AFIzVu|6~9G=DM{(H-TFrvfgV zK4?_SuWh1QDDO$`)}F6g!xA_+z7#vWnw_0JxdOZbXvSqGnh1N`4;c^R-Wy`QnuP;f zLQK1bEkrPJC$`~QmR>sMOs>{{Z>ED%GVDwi+0{^3L;JYiSo^p5H$tI|kTw$*ZVB6; zYc(M$mtN%wybAnDJR4v66_7I>=P}syl6a=~b{VqjQ!A-^5}LQN{W9>$Q!9xdGkD0< z9HqH(B^4q`l++wSmAjWXXL!QSmVfm5yz9tA1pScDpiciWxgZ#bTM{YlnXl+rv7s{W zdgOJi}Ele>hWS4#$6$$;i|{WNxX~F5Kb%{0w02@rL|!(b?&p?JU!0I^mQ@KEo(azks30zc6-b`=x4+n5)7flF8X;l33q30B`ps&E3fwQ>Tq|WtU-#%?Evm#d%0&(% z5tA7qW%2MM#*ws0lK81B2CS8>i1s$F2Uq~|U=W45UZOvf!S``B!g74lt^Ri51Lf)+ zYAvgzYQ~~@b=Jw)m0#{f)3@)c4x0wRSL;m5bNIx)%Kuq|gnRuu1MsZSk*h^Un$-V9 zZAf%&V{xKTV`FO-vxQSfb$(A7ZaxkUf*cYfRbFZ?%TpcQetb3r=JE>lJ$MF@Wh)%Z z;dq&B^`ZiDLf$^rDX5`Z9-#>~b9`IpNJQmLYEP+lfVCZeccIwx`O_qGPLNb3N`tJ4 zMUNLTnK5LGy=-r(UnCu*de!94TvHYV<(6zQf42t;xWc(!4p-6`*(0Iahfh|L)v=8% zU(NnsE}%>!B3-Iyls*Pvb9RZCdlMcSM02w5x{QAP+xAP;Tiboe?e$G9vW4@=dU;4p7<|9S1e9P5?fD%^BX@$+27sw+o$(z>{TI$yPj;4e|3 z^Gz8!R$ki$6s=#Iw%e=(aqjg0VrbjLxP9atV!oTC0JG%|?>=eK@-NXX-hqtL3o)z5 zN3VN(@;Y&I+@BGSaX;d@-S-3nHvCb_*B~BpR(1@N42=5L2aa)fL;h87mF2t$s2*Ks zplLlGUNM`S^zU7tv^qXS;^X1P36O0@!a0|>X?1?klbWb_XFJYKZ-)67Fo$}cHmd4cCu-yxMpDovi4c&9=+y85qD=lHfr3=f6HD3_&{hO zw$0f<#Y98!@N>94A`34;F#;=_UWK?kvSW5X2rKlXVsbmV%%6KwTU?VOTLw!GJ*zLg6s#ERzC0YWR}5Znyy0`!6-}+l%umj2Th(QTB}9Knw-;1;6#1j0 z^rM;ige&wc?eA|NT`y!|%-(_JmFI2kOJBwAb(b#aEHdfA_2qk7M`kX zq-YeK<2Fid>2R~|lJ|>O2o$xKj-`hT zDm#2^Pv6IMbQ>BPUIQg%E8F$8cw&m9Fs|NV#Pd6CdKj#4MmoRi9lbaN)NI1i$)pz>H@JhVW~`M)z-7-zqS4=)ea?(1L$bGkvH4!cEv^oupBP{@ zr0js$=~9RR^Rjz*reY}!Yd*9QSkT$ox3RjTRIN0nba9Qsv_!f@zI0fFyIZ`7Aj_zy z)o%U?_ccbN*4}*M{>OIO6!F$ijkM&;5iMZD9prlm>M&^asX?Z><44WhF6FlT0I&*d z#g~9*Ti4g%VKa6@XC%^1IY2fJ;RMY3?a7yY(8c;n{ruyFDCX`stGZg%-Ev|`vN2N* z%W}|GJ4Jd5rn$B_hLEgdvF%U$mmT?n!)XoaMfoO=CsLFpL+zcxY1Vw0vtxAC!c-|p z+j`}U4r)bAaR-x=PJV_pbOnEfaD{kbN`@gxl=_rmdAO(N=fvp-idJEw75Ji2Nba z7tyQW9A4M3XP6eP#QqrX>dlA2weh~JDBEyJQdiG=aA8mGu15CUlEymW&wFF(V2f9w zj&nB3n};L{+}c`iM}D8g8G}3FCDijSC#cFtLlFa#p5yM^OjDP76?Tui5!O@p8aA#N zHz4zK4BwR}s-}4tKyUo$z64ez}zLzNFmW zlzs4+d*C!Y8`94sKKuxgq2!xskbv1~=7_&vj)h9Au;NqYo=8C;9d5j>oKnq7>*Z#{ z#5xSv348g7HrCy8_dIpqT<(g$mnA|N)`xcHXA4?mw?867UrehTJ;~l?+9JNeQ3sb6 z-8nr)t>c&7GOobUod?+*+#e9b1%a{zQ19O&?;B{xN=U!A#EE#{?msd%-_f2KF z`oxG<2Um6FT67LDn@lc4K)!{LgLor)#NxSW!V%k|Gf^h=HqiBWN^-Q0~r zw%{oOlYidM5Ne9&hp(}S`U2wuidcPQec874pQFKho@e-GdnVojXAYuK+OgS+i2~YQ znx`?;H278sW)BkA=cnd1R*`3Y{ic4(~V!sUj#dD9bC`E5`^U5!w_#{fg6g zB+*&v?7Vkxl1YXWzZ|Bj%8ks#GX5K-d;-e#SBX$8%bn~|BL*j}nl zP4xpCnCEYWXjN;AgokV2ax&Jr2ej6c#~dCN?F$*?1>wBl6AITj-8;?Nq#@&m)ry%8{gEA=OugaR z9$6AV!$_+;>S59sBeMprln7Zv%dN?l9q!&fz#Q`~&oAu3`{VC}?0wrBbmL@&c74_} z^?mGJad;e9K>vN3KS~$oeL04AO^1fyqUg$AmwC^Uv@Z;ppa+lRZ;S-eSD#6S#4a{x zXASG~{hfWDvxaWdPa8K=p^2l_EZ#Fl&@et%0bPH=4c?5Ym~K?;m16##c6?P{H1ZZm zZ9w#Em#+8xhaxcj)f#P}g#0kQ#4g8cDnDqfYvLZj{wA*AUzPj1shjo zbvqNudau6I-nalJ&?4w59be22o!P*qFuC(&yC0>0>#zHC_?bD{uNgh3GcslN91?YY zoEZ$39{uP;iZI`*3#I}86^-DdIPZIx*c|2^k^1NE{!c0H>Yeyqc&bUfAK8D%9}toe z*VDOF!7+PK|V)8r(OrrM0Oi-O-Xek+%%nj1L8I=)0#%~Of--MM6iX83(`Rv%JXLXA;e ziQB-Sv3qrCv`aQ z=^|scv&%pX&{6Ib$`Q+#8CxD9X5hVHk8fK11`b^{Gq*JAV?}q|L~!m9{W~~Ajw|aw z%1l9c`fOLlWh6EEE)z48r--g3?p)z3$OuTR*xbGC&8kn!k3a#d{ckdz}pe#fI09@1;q( z=ngXF%MwAixx4#6(#}6yB(%5eOyPgo(Shl-q|I(1zlpAEbUIt(Wuc{c=?hu-8&&7) zR0^DfAa1AI5gIu6M;C#dv~13c{@s=Z<09w+r^){X32&Mi<~7pRF~h=|qwN!n_3&Zh zhHv>pn_qBXLDU0GfZ7lE2PV!PE-;wfv7-N^+giRmFkdG5oXM7%=1cYKP4sx*Y$PY(Vckb*MDBaq*+ce;p1$8 z0-}9pp(nRb_0aDRGwOdh1a%IG8Fab-{)kd=g6{UBHQX?vF!LK&!-R4J&P8>1a>|Py z$jf>D0I|7itV36<&8F73N|Sil3xN|?6%!^ZIl}PQQPS(DdzY(Z=)T(*0{6PEM3`h& zD4U8)q57*k6}SG*hi{K%M;rGRxyd=K2@qD4tMKD)X7E6GeAdl1Pfvt&->qO+*TE4W zAt=02%G}Nf{=4cu%w-=C+Ft~&2cw^2ZAr5>y#lbZw2#0k9ANfFZe)P{@gGkX-ZC5h zPmUZyPukQj>WYsYdP}6Ums#MK8BlkW3`VgMtH71qj-17@ZWPoDfvNUf^2IeK^w$H` zy?}KM21GE~dcAG!Y^IUB-AhUnXiAU978BL0c?RGIh^bVC&w%*U@E!^f(4x%D&9`C) zt@)*qh#Vodao@lYdllmBpJqQ74@6LbPjv$c2b7-9zrz;5k*gJYlbJUPoKZ1yBZxS1 zQ9FSq2m}LEfzn=i96blVjJwB6*ak;N*hc}1*e|D=$KgNns~POwthVF!T?ieJH0CCR<`l2nuGXq|8UKxzlaCLVYx3^9QcW2-M-5 zj(oo|?VtV?fRyT%0A~R&r+|f8F_u z@9+?4bpmYt0PN*uJVyn>*;Biy#lxqC$42O$)c!L+RUp>{OSvO4i@mYCFPxvoT~a!> zff=<=p~v0>L_C2(PQ`!Szat;p>NWwPXdnQZ1+|8rU!Xr=7XTRo(+XpC0T?=CDStMe zu(r6f1p#8Ok^c*5&s%CA_D+N71u*8XV=i&@!0F{hB2|waT2cdrupb)aMi78!hIPs0 z31|a@O6E3^nM}9>!cBg|xbQ#XyL#w5zfZd_1K#16Lj9nh8zdxpsCZQW~_j{f#90z`h_$Ni*`T_AeY$(q)MIX z4xae|7s#C!q#F&}NRwOr)*>+Rofpd6=nooL#uPxxJ|L#udlz&Wfx5IQ_PT9BePbVcn8$c1Qly-x4F(^&7cXpz-h^?{JE#m0;4^6iYSbC?e z#1|VZ5d`k(FdSOlH!I-T7x{eiBY#tXp|9<52LpZddE*^cHYrh8HyMniADo`}zL^2i zoBIbY?Pck~Eq69>Y-qH)S*u_6^(1LiimgCSKWTUs(-zocKk$ZX2WUQ3zt-O6fJJa< ze}U9@=U8b5I1u~9lL$6akD+pT2nKMv-&IW z(+Gj0lmEY$7>s>^ef#$H1&0zU4t$pa=l=)& Cdkc;L literal 0 HcmV?d00001 diff --git a/pkgdown/favicon/apple-touch-icon.png b/pkgdown/favicon/apple-touch-icon.png new file mode 100644 index 0000000000000000000000000000000000000000..d2a2f618acb8ec9b8aeedf78c8da4c49b5d06ba7 GIT binary patch literal 8386 zcmYkCWmHsO*!BkyP#S3%KuIYX$)UTur5kA&YUoB}=}piYvUP z_4C#Tnz%%>rNzvQ@7~qC`Jnq5>5cIlZRua5+3T#Wf>6}sUDFNb&dZk2=BI6kw@{mqp4 zq|SyL;V@7SzTZAaPsR3aikbd+DEc%s5^;W=#}J0f;)39QtGe!jl>lGim=h>(ne@#nD7EnF(3tvjsxBn)raeWGb z0Nb%~t-}n4-upHpL&tuwqAwz^lujLprRm0;N=n3!5WzE$1|A{P>39~X&#A#BfPPPP z&<4MW66g+(Jrov15l1!sR!xl+@VVu|)>BoJVPD8hl&X!3iFiUi z;2s>S>JFC-5TC!W#ZfdSl)P`d8R0-@B#C%45D4~scu*=LqG)o_`XI7fxg+UrpFxD> zydk{O74e{y?2JmbAICkJ-DazkQ=R*M%K6d&2$lV{mZ?~_>6!yymA{WFm_Ekh5{_4g z=yWPIsc9|*!r1Lt)(p7>mXeV#NdyQ{#{&I4-~%gcE~O*$aV=K9;(AqlQ-s{-jK-{N zpmwgs^t+AH)cbVkf*ho4@lP((B2w)-P2k!2FaJac!YRO#C?!QGRn>*JTzY=y!B#I8 zx(0^E2@(SKQo&z8E#?X|cy!bJvIDGiAK?6T|HNnKUr%9VG6dL10enR@G0 zu|+alIN#7&sB_lm!acPBIh64AhVTvnALpE5dvn=E5UcGf4XeMv$1g(nce2Mh$N);_ z^zJG)j)x%kruEwv2>|57{k2agqx30f&HSJc^>0Pg zQ{2bN+@dyxvC$K)4^1FJZ2|Oj59nAJQrh9e#2Q5kGIt2w!(mMSrx67==6LtAE$MR~ z$bvI60M_=vclPNs6QC>R9}6PMJK^{hZ+UmZOp`wwOdc}Y=EYzXbU+mqmO{$m=dUZX z9lLrvFaO4;HDFA0QsD)8SsVm~d`({%IGA>ri-^~CG7-1_W-JYmsomjv&xmz#{Q0wc zksL%g-sf*YM3LNhmS_Oo!&r_6B*4fUAVf;9iq@LN3Q%TlTW0fvVW9(DsH2SKuakS| zZ9REJ@#y%HHlVUt`Xe$EW#av=x_}%cNEh=h(O?C(bxmh0&4`UHfm-A@DNL5uw`?P1 zxBsTpednp79GrN>EOe$} zkj`tpNk^s--;xyAsSUU|UYYFF*ZByTYWjzy2)U+#Ipc?JJo%{Oa>zq`%|-+YffhcX zPwLKw_oPA_f?u01B$UKIWd^OUU3#{z6ADLgOi&ryIp%dPDr_bvSm^2DQ}|a2uZaTe zAD#6eT89)eF28eefA+UVl2^%0H?QC#gsXPC{0d^ofPJDt5Z%L1D_@&0q3e7*5m+k8 zhnD9)dlNbEG$w#@(Z7`FJ@_2-@&AY~V80~AQRi~3*yh|B7zATN1H4Q>q^n#m2!?qL z-_NmJp1kN6g2RmEYuO3NA%Ufl_ODL#xDkfw zBu$gA55jl3r@pWUbCvNzmKp z5v@j60J`>Ol!P=2$T=333JBW#)>VBasAUyvMFl#3fyhSi*4se9+^Gn0*^1J8UUv4- zUSn6X({HYl1lB)kCp_K3!cCeQM+*7VB|*Qi1|+|}>;=wL{LA8R-DdpaOHUIL zXB{V3J%p)9A6k}3ij&;6WPo=X8RcGhm0SFWz6Q}*igm19W3X(vjVKi$LuGnpf&TcI zj)d%QOnNIlJ$-GiZ$QzP$~K{P!e;CA*l*o#*W*;i9%MDX6QPpXMt08e<6jZabFq00 zLyAmGg_yka*n#17f}6)ZGs)Vj@o|RTPE{)A#-Xxei=E^x!Hk_%T=|E3JWqp1XPeEB zaO6J;vUrq9PggtRas<+m(Iz_YkB^O^F7As{($S+4lnqo33EW4+>3Tl{53c}H`IwoR zvW9DeAy`UNhDk~xN&U4DOV^`c*{{@%`YwQ$HI-?&E{DC_wY{h3;nSzZiGEG*41dlb zh1z`?r8w_{Q&Gh6V3_b%zaOPwdN)LP*HEqYnsduM=#b6CR5XA*tXQj!>z&c^cNxa# zmS)^6dOj^9hBU~p9v z>-lC*s`b;J5Gd`!dZP<$SnPe=AE$SV&SpT6lVofRuo^sYMouRwtZ2F^bYhX|b$M#_ zU5$>(&dOL>A#*%7cKyl1tY0>`7jLxW+hAGJEc5xf4(=9Mi1<%p>lUJ%GtG&uLY@7A zeJwTn?TWNr3H4$CwH=C%$O$)RFTR8x1(51>tGVMU!o$}HPqVYj?|)sen-zxW1!pP# z@SC;!q-HS6v3bB+hu;eAx;{TIG%ftL8#BrDL4F!C??=EzpnB((M=tfXI6yYX?b>B2 z_K4@FWDbENq&U=s;^v4+t7CX<$l8{JI!^R37O|Wzx7gs_X_-e#oot@eLve_GoRfTH$xp%5A4)sIzD{>l^vjU%Ljf$)Nee5YY&5D7 z3OTa=JKg*`M|r1UYmVvJK}}r)l0wfO*qWU^H)? zG^txxXly8K7{*J>vZpr8>^FWN{GRVMN96nN5X047$xnjKuPIZ4qpg1=9YBPSHQG0~ zJSO*xDcCg1y)^`MY;|vjY9|+%q)ERnFgR3f@35AgeAC$&!z7yW6u)66%Sq;&xdS`Q zv=hfSC>;$%wU2lPNS*6J9RlBXXPDp|;7t9Fyu^{OQPU@4BZSbglTktmT&T*c!%fBN z2(PHJNJfLVE#N5FRONJH=3PkWrsyT&nuY_$MDwe>sjim}MICx=2e*T7>=dZeXQJ@i zc9jXhKZB`*)?ncyy0P+HK(pIrHsJU;sl?IZxGTAS0QQ60hgW1IpsnloI@(=myRyyf zd-f%d_c~VX*{RsmF&_?bdt!Ge+sdjC;^7%9SxC%c?PL580Jxm3i#B$r4lXkSg8KFX zM*@U&gw%$TgF}sJ;{xkPq~RQhEVfxWN$k+A428`=o7#-9r5Om+S+?s67Ur1C6%_Y}>$9gHd0HFh&y8MuX-iLcQf@8^jg z*t30NOw4Y`}NKiaqM!m#?x7HmpO=R`aS+| z3v(ZvM|8k#XzFM*Br5pVV*M3PrBnQdaDaqs$sZFjNK*>v;0|-eyq&8*LX30J8qMZF zmb6t|HAgeJJg==RB-eR%&0`(zj%ArMy6v=Q&hMGg+o7#SaKd>eVpx+|Lap^vpSC3M z<1og<6tBoWlP|m3_a#ypg<|t>0;$vQ>>R21!tTWjnz0Wk9MR;$oy)=tOcs@(iiqP# ztYS@Ij?t_A{flUdaTnA6WIRBmen~c`XS3joQE~>U;SOG28hw}G z>CPb7kbS;-)(jf(_BFUu$tT?WiZamg$1_rJ;MqGs zyx4OyrWHy8VT&95xpT!5nXYJ;c&9D-J`X=^gFRSnKf@Eh}XMF z(FNTJnbb%xeA7Vptm0A@DW5{{Y*lpO8-pYs`pASd`m)?@xgqR^xDFt~Ii}CGgLr7N zX>>J@ec*3F91UL|y|s!gpir-b=TY5A!OwM6P(vF9{dA$reo&kZ#2;eLb=T>~?T4H0 zE%AlNk_EJoZ+&cPrBP+UMu8iWkW$*g`>{V9xast}PmP~ju#L#6Quj#L$1Wjt7e-Zx zDH9!vni!FM%sOue`?ugdY>F|Jrib`SDB~EGu_JB!Gp&=!D?eh~M*OqZDSDi)-5jqW zCdgo25Hs`M4xvkVhZS1gd#v0lCO&>UPSF3?0w_P#ob1f?ghnRHSo&W?5VOX8MOVp{ zN^}Q>#fM+X1=qJHdVMPk?_@tD%b?@s^|W@Z&BD#ag#4cHuz%>9caw3tr@nPGT?$6w zReA$O+AoBu`8cjP%xTvHyi!tB^bj+&1LTXsu$0l{g``S(8m@c&U|fB%v7;0m2KozV zly8X8^)h}K8JbL1fw-wV*Clo>vdP-VR~LLw6SBE{;J(*2!Bdq$gMJd$!Q zYM_6*4-e`?e+`*|&8)guPg zoA9c}`rFG%=L^_@gSm#FMi;4n!S zTIP>-xm-`erF*06ia5w+94DSoWM}RkT+>D+@M!AelP@rAY9sWPWn5GADRTqKw`qor z2~BoZnV+S|4FRb+VCWV2d|MQ4U1d4$+|0uyM`pblis|z82?6-23Q?cfzCqF8+UPKE zjwYq7_Ud4iK8&`eM|I*U?|y{z`GdjtwT`)}VHBMEE%^$OSia(-o9QNUsWkJeeJzwW zd9eFZ>pgW4ie^k%5z4^CXGv5N)zUW%j1kC`D*`^sl-n9@`#QdOT2nL2LceF^Tes9p z5EjrtyjvzUBB55DT0t4JER~b4I^sQfGR#G-`|zYbNIos?h;>canIoPjILy57T_d)A9M^LE(SGaxMH9C7 zw(;UhV@kZayxur9C^|Pi0TX>Iar& zftZv2cqg(aME46WbMjS~VV2lQBf5d*D?xW1k~tAC*<~~^qOBK4aLt#` zLOH=|F=WHDu5;`|4W~JU=);8d2AZ?H8?c6VaWoi+pG#9GKW@f9@Qvk&&dm{Gh9w-S%kp;8Rodz~ge+ z5Fa58r`SqZjVqhe-qtVgs_C zk$PBqBPdR=E|L!DZWC>#-PrF=k{VH0c|13hU2;fI9xr!h6Jr6p$Wzu?y%GLf36JC1 zwuwTE`@#%hKEyZuwqq(6W1MMoW~ySn03V7TU@ z!Y#M!{+iBVGjgroK3VIL?2BUeRX+10a0mPEnr0-w_{_A5Jna) z?>6_K=P_|+(tem+vYyILktk3# z?68kMOg24a5<5x{h~-#^OxG8CtD%aj1o9@MO1<65ss~7g?->ADuF7|z`YKDdcZ1<) zX2nUHI2Zdym zmn^P`4ip`cUR5@?qiHS4AERJqaAnR6;7v`tX~p`(Tx3}r_Eoz8@GoIqXs9h7?NLW*-6E{B z6VVYAl++jN$t=dGQbTanH}u9a>t7pJ-+N13I`0zixSD-GLuBx0Q8GCA84~z2WV*Zd z@$5Is{EPU+c&t%G(VAbkXp_CEh_>_6;=2mbvq~?f#r^232tYNp)ObO+kt;bsf@nvGzB&PWv&c84It&z-@8EyH5)D@Rb6WFJBiy(=xIYhE@xbxT|gMIa0} ze6w7C*RLgGnf8FBP+>1)xE%+un9<9be zTV8*=Ah$-RKySH|lJOgGKrZK8w+f_3&td~PVQevCP@R{74KtfmhOH#}1u`v9I@tOL zLj?C~-8y~WZOUKQ#)icUxB=H-C*67EvNpQcS>qoZ*7M;MSX`7pzM-e*{H|1%0mk?? zKdS3=2gpG~-qv;Uvd{X=Jj&E{^9DT(OEeQ*cyM|%@r9=l#3OyVMF5WXo+JR95`5Sz z(FnNN-#cUpdGuet6GKAMFp$!Ke8Blu@W5(Lv&wjU;CHP3B0!Hc`PbKq0iqORuP&dn zbX?d9UQ_UsasOwQ_>Dd?5USVgS8}mIJ{WN~K5e?}uhj6WXBgXE&p&>%3SjonNB@ad zimf-Y(GaeDHoQ94$sBIJs2zmMMhf}Rq}KC%Z-!Px&_3H0-T&))1RolT`acc}?h#tK11qp3Gk@L+Bis3^of4IU*Gq!@ ze`G#&Hpo-P>5I}AfhooQUgK00t~^DTaLDJ!$54iUC3|`jlpj+tb1G6?C@szOeB~*) zAOB`B4V~I&H7vTC%WK*@fV3|+r!9B;U`-93kxT_5m3DVct0u}%ju3Ur z+{($QcX)G$)n$f2wDn-?$>z(l=IvrIFs~vTui@QW!wWE59NVaGv!~KKNmh(>yyIrO zGmfu`lE*Z;43G~WEV%uAnLUhFSLPXYbB5ama+I`rwJvsu7=QV1>vlXv=b_eKNy4ZZmqW2s1g`Pc^s*0i^j z+m$+Ry)P~GUJ`RTm`XgvW4IG&_n90B8;~3=ME#M5TYP;`V1P7eH)xNaA;Ws?`+P)2 zOjZsl$b zs}JGN(TPx|0oW3u>#*ug`mR*hC?l~_moL0Xj%VS5Q3Vn*++nM;9&Xkx6=~m>YgCK~ zKnpuvis;f{5A3>Kzl{vsWEYKu;(fQRfz_ks zRM*i&&S}e2{rCA?FOiS?bKlW+!XNr{kmQVA)0b{CMwv^kMf=HZ z$DkDvWc9-p1}_!`$QGfi=R@|Dv*=ifX)Yi~jak5(mX)ClgFq|g@Wnz4nep<+1kqq2 zixB`)(qas<$Bd=~hf0Y-})e>rP|s`UT>RHaV3H6>|3f0)gLnHKA3qEJNp*1hE! zZq%I>mhmC7-LI4BTgX*|1+CbA9ZpIhBQl8;nIZo6 zmv4(_{YEc77(RvuK$wb{H5&6X2@wF>-z~s2Kb|3^vxxA|{7Yh<3|PpR-rdRc_85BZ zsC>~17u4{;dgx3NnMWcj>+U(~Sg8@=W3L8#y;V&akpN+6ARq1FiDD6^4xq}^s;)H% zbBbolY%M)STQjud(<1su&feFatp644_v^{Xiylsu7Z?IBUFnTVIH>07#_P97Apf_d zwoVpekF}=9ye|8KT=@+Xu)irT`<^3CWtohjX8oj{ZQMlOUD${@OspG(e)8hf;BOxw z8bt8*b3U-{aMWA7m5@jh+rX7U>IlL^uR@`OQq3Ud?xB)xH(_Tp<#?aLxDrUW>8bBK zQ#GK`FGTC_7YQ%xY+QgGbKTUFGLZS3st+s{;` zz3a{q{hb@-d}TUV<1lJw51$IEKeI2+jS>@u2Aa5CaCa7AydZd*+Y^R9h-4!xT(|GC z9k?T^>!HhygL;~9{uRgwmW>W96V!bn!1R~bkGl*%)1bHGWk?T%gYmm)ZeFZzq>JO2 z#A*b`0rd8C(*<;2B_`1)^;qP!z_WDOm+JwB|Lz4mI&C%Sq;hxsn}Q+G9d*@CAO_ID z((fOD=>Gt8)$V2f>$VcyRfP=jw5};jhOl5EgK%q3dX9?FPw?f&SGDnjPK1NQH{ZZA xp&JNDD@2U~{|5m>BWnNv literal 0 HcmV?d00001 diff --git a/pkgdown/favicon/favicon-96x96.png b/pkgdown/favicon/favicon-96x96.png new file mode 100644 index 0000000000000000000000000000000000000000..a3134ca964937fafa5915a7fb786334d0f9559b8 GIT binary patch literal 4035 zcmV;!4?OURP)|v8dSw&RbaA~!EwY6@wDjjR5b*Xi1t(~^tjLz5ArC(=pp3Zby ztz&KL($PBOhFZ{Cb;b=81q5+L_8mz`AdrLv$eeR2Dwv3cOGrq5K6&KabGd>4dveaX z@41}69n6&fS@mU%kk3n`Un`n4vrleVmIE`y24+V1;tj$fA>MlmD`eAJT6N4ZTaFEa zeMQMv3##Ye-nL2sGs1iXShtQd*EVQXjjZvbdS#OxJv3=(2&f2WuNE?=I6Ly2Jw<{w z`&)`Okn>w%2AG8aGuQ8!*4(7sRavWWrR7ku8&=L4hnQe*raq^z8lN9XL4IW|Qx>}k z<#v4T%SYBOOooYL(gIBW-PSNyd+z>%N?8moy9({GWa>yv8y&9q{G^MwuCdS~Er-wmc&wJ$eHmWuF z_Qy=@J9CXG^X+UA=;OLAzeN33#@1y@m>4D|fX?QH^*dIO7W|2 zd@P(Y94}0W2Ak|BUDD1ISJ0?xV#=YuqO?+J-K@0FeqD_|q|XEx_}1tA;odHrORD6H zn||u&XN--+l4s-KEU-s!P~WJ=wqqGMa^WUZcHnV!{+`a8GVj*?CU4`)CiDS)BEal5 z+ZK_#*Ui#8B@^i(p<=u}dn|&*A{c|hN*U>v&f{8PTjO_kaa5DL>NgTUTD%=b#;5{J zfB)-I91iDTX^oue*u`XE{nd<7m=F_$zJT7qZTbEZs%qPAo&!8x%at1KRAbs2MiC%t z$zR=~2Dlw4sH~l+Bl8ajJ{Jon$Ks{QLtzT+KADBxr?Q~dG~cr|t{2w6a^stKYthT} zt^n)S+0C^L`KU@-zlzMyIrPw!5reR7`Y4Em4lpGo^-6q|bO|Ri@|ZHY-)Qx5ciynS zrC<&DJ(|%=^qK%O)_s$pQ8#~cSKjDE%aMVe#6B2{aDPvjGPlcW@aeu3Kxri=&^~P7!r)-Ge^N&Bt(CpTBg9}!)Z8|Q^b_nw7q*IKxdmc$VY^C=8nH-6Zbc= z^DFV$!BiAg$(S;=y*V7a**&o}x+MU$4}9EtN$obVi<3PTKRW_5;v--Q97(;2?Zo!3 zYrJo3R7x}n-LbuYmH^Zy4h^*5_@I3-_sPL{m3TK@*~I;g2I490IG%}v=dLs5?%Lju zDZs-v@#qMDy!rfC5HAImMx{alv9N7?E3TGfQ{n|=+$m?uKiBr|Qh?4jF%9^N+3^?^ z9so$m>MAO)PNW<5jXEMY|kJV0(5JVm!TYsiFeKb#-`q)JXj1(>eGDwY|y8oQ(}#7IiyaQK5PeUaVensuXfB?Z}L+{J!>Ia)OE{d62(wRJJjNWSO9A==bz~q-Ei1)}W5>YbaBz;a{7dB(h!cx(rKTJ%jylBp zxFJiCiAmu?Odm858rxd9I&$E_b4OLvPYX9Z4^*nF;B3d<9lm**G>?g*sR?^;-avW1 z0xw2KK}LiqkjpVVK#0_eV$6;4!^B8C(oQ*;6(fd3BZZ&6AIjC`BrqI^?a48%1Rh)u zc<|l8^80@RbhtD3b~ed-imA0CGYZ+o08Ey6FwK5%taka`{F#CkDztdULw zQJ^d&x0gBuI0m9o+lWDefyhnFN8+8MD8F5aiaS*(xn5@Q{TgZ;ab`~{hCCIGC+5ea zyO_NI3UZTFB(KAtQqN+|lkxC(vB$yUe`9874Ab=}v`YLs?kPCg3UIqY16{5IZYuAf z_+}{z&y?Vwd52I&0^VItf0tsp0vylJ#o^m|SUkobPOcIhyHkiEfp%oB?nJtl2;?ST zL(Z`hq~~9OTqcLCs1CA9IoL#1uR>2SO94)o6OXF#Fx-&=Xh_VAp|kh+G{IHOL3Sf$m@fRf8Jm4`0N8{*?k5=|Ut|q#f!j(0dFR z8~}l8gTze67cvz=u21;VBE+M zL=5uA`LqJ^%7?c3{yKF(tmSw(D@UOrJ4)*o*U2<^lYXgba!I9 z5ZYacL%k6Y+19$9+?-%)ur>y`y9zLid~~0k6p8C(UR_)!BY~IWPN78a`8qzClM7X_ zkqgM`1r5FylFzTARq^-_Dbqj)M)@H>Lr~3P)sbN$)cFK{0^{j~@tkUlHj( zd@*v`Fqk{636R&ANBZz8L<%EN(Q_14)t z^6`I(?^sOmDMl5bFuNE#R_`%b|K6TAnVTm|!l(jldvCYFx}~4Jf}xWwM>|H6;?9+lbJ`Su2K)wF+ z`_lB*&0lAob{c{;0rvj>u-@YlV+SE-Vic_ChZOVo!?Qn+PI=akA?|iL1njQ(T zXZ=CF$GLniUR=|EA17p!E(M^^bn1gHwisGweDAwA;pXECOGk&U_u1(3;oQS1H}IFe z=TO(6?r@wwL^XHy%!kgO{?-)4Ju?KB42>0y+4mQ2b$PC>p$es6-1Ue6&vsU4 zUankJnx5RK@BFT}EPMa!QRJ+o2QynNpE(+1qX*nOhjvGj(pvb3i(#pxUN6SqkEF4; zWm!k8oAw!k=BnA4n2O7{+xC@Uebc@N(YI~ooF}50Z7ZD|_^>35ylrJi9@@B7z{jGg zBQP^Q5|%)pZQR-!>(<46v~_U@^lsiA#0=Ed&AYpX)$QwthKRpv&wfX@ucz7X7EB&$ zw%gaMnzYz`D%-@jukQrS|LDx+9=(^q!PyZ@r;V~Ea4>+;QkaDKd*GdU*N*E&u90;G*+GHoXE|`O^9ESPXnDb z^#y0L3$Z2X5~}-qBDc<1QyJdJqLO)1o0w+YZnudoHks#x7%QDE&9MEgHJz z0+v&nA4)~qt?t>x#=^t|XoK3`0|Onlwc8q$gePP0%PGUU%+^FxhwS_@lgXe>>8?#| zEKE{>cBrl4YR}!@ZhOrNP^3(n2JJ`f_eD2Ff)-FtjiDMQ5Ffg`v$j1-b8c)UsLq~3{&t5HJ zPGQp~HWub1z)w(HBRbga+xD!jly>{zYC-jUwrpZ!VRizv)1J4OeogMGXB*Q#V3Yp> p00960!F9D+00006Nkl5icsVoP@pAzM_W zrR(FIeXj4paCPNp_x1h<06FM>4q!dLvjzZw3~*(|=K*=?2A1ju zp47K&O|H~b$d&{yPMcXvV_SqkqaP;xj*anmIvk$iw+8rH@>n!8P zUuLRW4kt6FkB5)8_?RCzh<{s%e=7Mp=TD|as2u%Sa#(@4OF8zeldZyKtzo|-NZLO0{ zkYh$ike=sY{EVD0a*FS>{pMJSD4sXXcLLU>)=fkIWU?}{G8n`aR!ehf$ycR z+1Y-)3MEKZC2@b49GSqjwbN`9>nx-GERPTUT8pg^sRfHFOp1xrTW`k1BuX;Yey*Z( zwT_w@8Xk)H{5R3*{JwT#X9a5hDHW)}nn{Q;X$9xutg<6Yv zg=DOSub#XlIV?iM@;6_`2J0u=12%|a(RQD|yf4%v#Yd-drReH?u) zeCIiRdAPYP{FztdADBPi)JCHC!4q&vjo z>RTXsZM$Ln_*QQGcooaa8la8{!??hry#87;cGX(WA>F`2Qrf*sVZp`-=VYH zo;hPzHJ>2 zj~lQWd2I0Rs!V)fkY$AAgE8w!C~Ci>k1$bDP8>zi#K49Eiq+%DN)Y!}U5#hwa4UPc z;oVK!I{u2&d>sW4tz#&FwV3e#Nc%b|&_yB<1(fH7Lh7Jz$D-j5eG6|HzEv86K=u); zCl51kdTIIhJc*VtH*`jt8)d@%mM@W2#;>tqoF8vQ;_jGp;=jp{_P6A#ya`z=2Q(}K z%gvH&$uk?QA!qQ3rMY;;Nzk5OveKde#te^0YW*PUEm~&wXR84Y?dG%>n#@jwsId3; z;H)@8+3j0&80GGoWO4WRf#Uc+X%Hz{pXhL9kLj?Ax-zIJjaF_%2j3*E>$06(Bbzo= zhIeK5uR^Y$2mYQis<*B9U|;-HQAsQNNPuoNh1G1(_*#ziIC4%3-?Eur7r#!iOg7?M zpkect7>qlmo|dK#(p{gQdVz*LsETfATNvYt)1#>~x4vgWHOOC^)6EiP98{W^ zJm8Y9MWj-df}0D3(!y_?I6y_=rmms+bT?A6E?tU7&vFS5J}y+CZJ9W94&vvu3m$Y5 zNh>2!HSON(a5$G!RT-n$-|rHSowO@jsBtt>kQVOxmR3)Y`%v0vPgBdaG=9yGv=hFQ zaTTNNDq_F5FObODanTL5KVd#LtzTTwExuM0eWpVL z{Q5&Q4gJ^?dPYU+vRo=2qbsAusU{{&Is4pEo?_Vy;tJ6vS6s)vAH#LLg8MHDZO3F$ z+%x4r6nZz^Ou$frQoC3F7lrP_5Fqq#3K;`0V(+g%|3jgf93nj`W4gh+Xa&+C&MYP` zP^1;lrix10W0oe~Z=abfnqBlw$*QJ{d`k^|@(^Xrp9g1b`Ar5X$iPlbP25TXjUfD# z4r)bUG<&oeo|nRuHj}La#|CKdm4hR?bTD&IJE;+1@GQE9R{SjLBT?bt+6blw8X@SA8 zF5^%!IKuNZi{Ea`xGz*p<~xruT6Ft*&gf671$e@wN-DrJR-N~5EW+KkEBFL*-X{G) zp4#}(Q{*9TNQ&05gPy#uRPBb?M&`iGDU29&CE_PB;b~s&QjRcM{8b+l2R(nC6m)9-Bf!bO(5pd2J`4HhbCR zT-;-ym&H;Q*Uvu|b+oU7P0;cV>~%r6Js%8bNF1NDcPRmROyfgtChput%mD0{7M3Z~5flG&^> zj&CC0&jS^WXk04O*X3Q6RCH(p?q<8Eu#^2P5e>K&%iW83aq$#_ZqnaTD)Uw6JGT>cBT|r} zu%Z@x82nNJAs{@*Sp`yi5+%D%^}En}MrMI(?+ZY9H@I7mfyQdExwks*#rZbT15}Cc z5J$o61E)`4Hl0>U136hojT7Rs_Ge!SmEuy_4s<}RN?AS6PZuZUh$p071TjS@w23+s zLox$q)@*22W3?$Yo$cfGM@wTX`2kYCv*_O0dHP$?5bKvoU?=x1#aYD>>IxW$8Ynaj zX>F^GZ$se2b%>wI1%Ig7d9a!MsWQb%8ZsuCMq3R0VaIqbT=3>2*XYx5bj1gT!TG$& zEUuALwLiPu5Q)$NRo$0pq>*#(7>lGkKf0AAzVL%Ca7@ht*iwsdoPOb>jM>HWB}eDC z7taGLGy_^6q?HQpOWjdO>F`5nx@bN8wpF^i4$|7*wJ!v4^yU50=!wz{FXC@7b962H zdVtF(JoAoUVX{=&%qJ4vd3)e8;Z{!PqDA|b@`}7jwQlS*^ebT@UzxR(RXy^fZTGL; zgPKg+R;yIx+Sp2Lt;NTN!PwXRj12^5-CnwMG*IVMHrTHO^mW)DUI!|^=2$Ioxdhc9G>G?& zXy1})K`1$&+}k0B)OnV+jdsso4t}!tq}e6}l=0E{rr)hTscQ);{oZeD?=i|{_zI~x zp+neB@yzUMmf228J%Hf1**8gL{EyW)c9kBdMOk+ zq-c}U@zfk2m%U>%9#KFJg5bjRB9Un!`9EEAP8bf8EJKB&o#(3G<^0f@g|)})w{Vc8 zDf@E;Ew~({bM?eWcl(%< +dampack + + + + + + + + + + diff --git a/pkgdown/favicon/site.webmanifest b/pkgdown/favicon/site.webmanifest new file mode 100644 index 0000000..105ff31 --- /dev/null +++ b/pkgdown/favicon/site.webmanifest @@ -0,0 +1,11 @@ +{ + "name": "dampack", + "short_name": "dampack", + "icons": [ + {"src": "/web-app-manifest-192x192.png", "sizes": "192x192", "type": "image/png", "purpose": "maskable"}, + {"src": "/web-app-manifest-512x512.png", "sizes": "512x512", "type": "image/png", "purpose": "maskable"} + ], + "theme_color": "#ffffff", + "background_color": "#ffffff", + "display": "standalone" +} diff --git a/pkgdown/favicon/web-app-manifest-192x192.png b/pkgdown/favicon/web-app-manifest-192x192.png new file mode 100644 index 0000000000000000000000000000000000000000..f320e0c9969ff46ce5412f351a787dccad130503 GIT binary patch literal 9825 zcmX|nWmH>D*llo9+`UlT3c;;tpio?jyA&u8q&UUh-K|)0DDLhq#e)@hcfWbR``w$h zBJ24vXV2aW>5KiC^fy{iSc`LjckS0>r6M{i!KR{G+6>T^`WFkJyeXk@Q{XAmv8d%6%$&X6ETm-al_oB%hU6~@(TcK9GX`KftKn2VJEta<%m;%A2`RN z8~@Bn%|>5iP$~euKb@z+@wi8Gfw!T3b%m{3bFOcEsd{E`W+!S@U+OI3_6O*ZoF0Gx zdZ}P9bdojGWhT@4#QmM}ADv~5>$Y(F+qfYGV!o0w{62V~vxQ?IU~*Ysd5IW{A{706 zth_WuQMlDs8%XR0k0(Y6GCM%t+gC1{L@Oix73=I$VuQDR60BY;?XX~dmnr40UpwLw zed~gHb9eu#Q?iZPbgRY%bo~y`*WjQb?Pi{cox48tZF&*9ky9EJkw8NzB1r5wfd2gF z&h=M|AA4t~e*Ruh1#_c7N}#JMHqJkw0%N%it?@L;3S+s(GClLumBjaU%Yh&)gHmL$MJm9F)}vo= zkk;_7Gi8pqbLvA<$@9WkS853xC%fogstYU8QQwK| zS!oO*8Q7FDmoO!8Yt=0=@qCf{KE9@C>ZLAZ`;P=yuUcKQb9Aq1@wnnYB0c4>4>OhZ zL-j4T>pi5rYA=~63)pXuz8|1cTSzWtm-4##jfY$p(lahijAS^3ReZ26AC;Ekjm0i zp;)9+0n^P3?y#%xV|M6*w49uxP2muR$`Z)d(Z=7yu0}&oqw&owqh~C6fM@`eXvM7! zEX5P0atQsDh#kEyPl+k21SjuzI6&xGcLqOUMpy#DJ3UHUXH8HLc|HBB2RiLSoIe z8#)V)FEa194uotkC(v&6-1oEarK+uvBfZ1Ka_izYJ4l0d1QDQ~J$9Y5f-mDTapR@R z>=M+}dTETIynIuHGUKps5)I|eJZT|Qn|ZC-re-CMAn@%eJg#o%dxEcLnyEAYnu9$r z<1Hne!mDo4+EAUl%-OD(JkK}Y_;tCyA1H>GQS+;aC0RjXV6jgIM!9TbM9XsjICTXunM|*kM#EW0tk3xQimBs28GDxaM-Q-KT5$rQux! zHlU(r_qgk+5{9*@z|z(+yaO_l{8>Sb%$-zkJNw;XPQyF75_5}r_#ECHI2;4RRo!nV zd&(J?Bii}cMDvZj-{fe@QlXVfG@YvT`A}7xgXE0r#Y#giY2BbD7~4d&6!9mAHY15hSPcKZ$IrO_*!FXHgR!)*Zn&|FdI(yu zaxtd?D3q2*krEFqB_agvLJUFb%!e7xbK8!5b3*i%JgxNAHqp(Cl?n#`tqZW%YPFSO z_%Jl_W7KH#j#i2XCLIrT(%ty53#0{psvx~4g2tQY12Cot9ngDTklhi+9&PYh?ui|S z%J4qB1j;P03aA@N*IP@2W+=Fu%aRi+kej%`CYj%?5IHW*X1cC@(_pPuX>OtVA0Ys{ zkgRrf?R?%H0xoIlF{OOjEp!BN$U@!F+0C8WN?-F#HsTC5;?S5pmG-`Dl_sDDHtkl=|3QwT0dstuL@zI9=Z?yytTYG%s zSI4DA&wE@+{*s1|A{sjs8r*dZJ2I6dF(*S2~!ZD6UkhT9)isWg<|!Ox+YC)2WEihOOwiL z@3m7VJImH-XFBIQ1TW8vl$oMwrj)I54pZ+7p)ZB88I)5d28UBACs0g53CKvynD=ej zVQ})>EMMbOi4_dFe)%VFTNE=tbo+0wU0ooY4+Y$CX0e0SkK9OI~cF+M`PbHNijjd{$!7^2k5{$)|V;2)JF%GLet9DqLRe zto#HcoCfn95QIm&+YKcclq^E|SsAfC7YO4=uEq5&Fd9k5ROPf7x`QV;HT8n~9J3;0I^#Vom!69XPx!)RVOrj7q7CPFpge?9I4vzyufl znn}#~L9v_hjN~EUUVOWytq6l;LkOSdO{|fhdlcU>$^x9d;eBc2p8D?!RUU&Lz!k{G z|Lt~r=ST>8w!IZY#N2kXMRQt-@}gDAkIY0zi&*)}*L>6yOqluoQ-m}P^~Ku0HY%qU^Sv^4?eX=^xvgO`!#Tkocrwl zP^@Zcb;5(H=-lFyJQ{aY)n}A=?!vq=_hL2F@L@~QSR(1qub?0L;aUr?X4_cck|y*Q znZd%d2(ze6kU8?$zrKG)M`EAv@lS=)OQj0(Q}QF7ZfkT27$+;JMMpoBs@u6*LTtdd zOM0iywswLBt1r}%Ks;X{xSfC5c1==B4lO%6X0Tx2Qr?4#blH~i`BFOEE-%0^G^BUS zzUGpdT>r2MWGSW04K>p?^Rs3io=(~Dh{!8<2LZqR(6OTdhHVQH2SKEW8AEA|F>t^? zIT(Dk;8u$ZWNMTyG6GFFJ@&Tss3=--cLYL|4K-qdnQ_v;Fr<_USgCKVpPlpAyuN6# zJ}xJK2;~9v!3AY_jYninVT?DEn%Aj=m~rGH1YN0B%H%eSVY}h{D^xU~bp|wc=@?oE z50IH{lG~-9WS{_=L%}U$=1`t`u|5cMj(oF&Cb3RH0x zo-dlpV7`+GZRL`cI@!0Mpr`SsdXGjHD8^t{C(w{1N9|iNh9PBk?d^I@v#9;SzcB`#4rbre@3!3-C3nx;lEnk8SjB`ggq+zjb9i4F-ZZD=@s4m*yO- zjs(V@PwWgii%h}J0ApPBvcbs6XQ_}<#r>t2uLykpPv{c zX1y_#f|SJgI&mK$NP2DW#W!6^=L6%>6=_N4{_L3NRkq;dl|dHue*ixApTx!j4?C_+ zL3ENlK+?_7AqzHSPL6-*~?3$yKj?wp#8qo5hJbMtxKGWEBREiyBM%$U{3@l=1%afW7&tsui3vAYO`@BLk zfoMdlL#yhUl;&qEfE9A(sB+}_Zj>flIuAtJ%o}8x(zqq-j%IV;Rq!f4RE8t5xdtaK zO=2VHR=flaZT$00fwU*r#Q(z0dHUe;k~V=Z{+#;yA8usY#*X*vD`Q=WuQ)QdUVs-G z$tL{?aM2<~p1Fu^g)Ycy zeEKWG$P&+hslZCk-=XzGvW#_yMzG2;1c~ys?0;JWU&Xqtrvl1Em&;YY#hzzr? zcu|3Rm0lY-(9PkBlNgWQSERs?l{$%M2>ZAn9QTO=la`KYqF!1-3e&*r%+7M60_^Cp znstk`QieL;75gQi%IQVI?SX88xf?OZqByE+Vjh!k`0txVQKw|Dx(ISN-tQo(Dm%8% zd*gyrHG}S!fjr_@&mXGAQ7MQ6InX{+vx*=V`VIqg>hgfgcNRut=a#fsyfIs`R&qyi z?i=)%N<<%6givad6~9`nRRsBvBx6m+4LYCayuPA-NVuJX>&8a@*)u*kzMt=DmzoKf z&>I2^XiqJrr~B2N*_n(DS)wbT2DN4!sHZpCrxuX&{gQs{W~m}1ml@$KHU#*aA~azt z8Fs%Q{G%-Cl*I_{D?prrvULaj{IM`jd*}6UALDe6eT0Z(nCnGV>`S3&m6)VZovNeVU9YLg`2G?-_72Jim&BoLTByy zy|NY;Jq;kN4lit*a=(a03Oj;F^u4SkpI0oAmms;Yd&JMl@uV0(GfsMJOHn&ZNJ3uA zV1_EW6HZNSTkj zOPHyma}vWgiOKj5}Y=4RF2g}#U@_}sT0H%MN8$HfHNPDKpJ!F*HPPa!6tJbjS+p!68_x+H# zLgI0e=P3CO^GQAdVQr6Lo`y`L2@@EOc%8hd*a!0@=o{5y?9JT zqB-q5dhdN|v`<;94Ll{<8EmJIiUAvlf4mGoQr6Ooxnz#hSf{xo>@J+awm5I*Gax*8Ky~N<i0R*W0 z`g(xE+L&p2$R(LhKf;7vzMl34+?s~M+c*@PvcLCn=kuY&KI_`f z9k$;Q*joMOGcUJC$>B&$jSK#=OP8w4|1Jyzv6q`e7_QEYX8L1X>zaBsk(wt%&(a!+ zJH$QICG2x;If)Cei+#1LczYeD1d3hVkd6v9TR%XYwUXN~s6Z}!x<}<457qeS?}WP0 zP$hbB`%kjXuqs>4;O_bQB$g*l9U(v#=FMV*8y2Z|TjzwR@o;ZywH3AJ($b)Lz_5C7 z{JFy^BH< zMiEwm@nkct$!+VI=h1Nyfs1WMLYjG7-foAJju+3Jzl^ii$3RSQQ;!LCsw9&G*)07G z$4rGfG$nos6+Le>2Q5EMa=xyGMOV6@x>Xh@F?rs=9#XHBk>6E#fmvvj^Rn`r#Y_ESNxo`1}2^N7vc==(UIQD_GtF-e(y%BprCSV9uaJh*ASfRf}W&Q76JsYNe zAjO$y1Pz&zn5JdJ?#qlFl}mx`o&iT>%okuzn3FkagL3$?T+<~qQPzQn+y*}hp-=zK z%e=f;%|RwWNU2MhiLnm9@lvpdj{HVA|K-?MTN2-3<)#~}bMW-;7OV-|mFH_E3;*yA zF@hgVR1^Mv>4;fvE~IPrqQJj!s;@%^HXt-xf4TEX*|Lsj@)UZ1n_Wzwi}i6neLq_U zF&{OJe(fqtR?*v$(D2!q%qhrr@2QtvV3-iFW#&Cf5nyLv-|(xbfMnwC;qM7(Tpy_}$kAdt>4gD?KN&so+w zi|S;grRnE)4@bEy+~Z%kgwaR4OL_a>9J@oX=MFW3?d~H#lwavmY0&s8biN=rAGAP> z$+>N*3?6phx~e{vO8|(l^aeIC26NGR3o!>qCR6;4)3n+Ehb*KQ8&|)?iA<#X|J^NX z`WiRGs;?q%;L|Z?NEiuWm}p@od~Z#)C7kEWAjME;`3o|sg&&>)J%{M}Qqztmu~$t0 zJE}Vqk=s8)@cLTqj&y;3Hii{EMse`ijbYWdX-YY0qoVGZ4*CD}ZL|siBD(cR&N~=W z$pfrtyxEAU*%VJ$cPf|*9p*%`j3&slt_xII~(L5$K7b7DN%!+O>f~B zZ|4ncH{{6RDbELIe6rDyd&`l;db76-Ur`|sH_{PJwUxx^wS-;4L>++*Q(5}ukw9?R z(mUVE+ud)LXV-f7Yn~ut*)dhX?E2fyFBb{j-DhJ#&7`z$Ws_uh{_E*G9Xs11mUgAK zTYlompBzGPQ`0_gDA1>8qK?5x+sL>6V?_J&-h4|t+l50eC%dzurC}qGceFOpH~JN2 zo+7A$_2r*7h`1EQ0#{HWns>ARw)ocp^>k9v(ztlQ22l($@wD)M{*Y^SPxQ^1_u^ae zpbc!md{gX3xw^-(v2{gbbH2}iQiA|c$rJu_KbL$_g`bMgH~b*tc4I>L#l~Bq3hB28 zyD>9w$^mwhOL)r?JGunJ6HZaMY47qD1Tc-f0bu?0ipmT(P2-pQF8fuNk){6^G_kMC z{%5?FuA%yF5{$tLBKdM00JjR*P5*Ve^#A+6A9rO5xS3ZGetK2E-DIf00wCO_aA zn{*0pKD(}iC?kKma1*0LV8_An{Z->s@`D|ymki7&FALBvaW;x*=(qb2!fj8iAbpX%WSgeylw1zPB8$_4$@f zIq+ia1UI$WaLsB6A>sUvofjhM-`zwR$>!TBTpcYmuCJ>V#1||ADg|`7dlCJZLCg7= zkRs~u%P0yWzKfVf6qA0jTkuAKI9%9wuV^lnvV%p56=1yZqw1&L4U*X83xVxm1$B3x zDZ)-vL$+sQ&~=ajqLC>=#rL^OezguX%Fg9icw%%}SH5fikJnu{o+pG=j5q-j^;T~q z^E;?y*yzXs@!HBrUO7Z?<#NI?DIy}FJ+k97Y1}b?nB?X~qQsBIgpmDiIiL&kK0n{X zT2AmeQ$<~yKF8fqVT6&KRJH5pkKScyscvLlX^ zd)DEMuiLFl==t4BoaWyPy@7#m&?JmAaSJ#`lDESFVk{9mSMhJAZ7sYsn-Uj3F`}Vj zxn#tVhI!=;B(QcgaIaQ&wjm5=pW5J#?vneserG!kcp~(TKd|%S_^n30thzL{nOiyD z#&O|h-ec5Mm7ax8P#5<5{_u%q(V0^Xk^x{BPiUt(J9x2sm2FI#{gulHXJ4*uDm2J; znN|9@JGY&pBGB#P=+O7KWH>c*1q(em9%0YUc*ga>tfs65m6L>f=T)Nd&efHm?ek+A zxran2{Qd{qi_SJ=NjEQbt!e(>a+Mza)XZIIEEX{mo`_ua+FWTtb6V1y&(9;sNp~mz z>ol7i_tt7>7})w|AK<|;y8o|LZKa}Mn%KgN#qil=scz%^#vZQMqxid3FU)V~8ZJ9r z?b_(be)x!cFlKtjJ}OB$T?DVp2^lYR|F6Y=Bz5Ace)oOkIxGaQ%1961b-YSX=8$|@ zQEzY}Vc|oZbEyp&xmN?F<7E@Un$X&Zv?ZBX5|Fr0a73Qqx%#GPe$VbC^Nsnf^_uDY zInkK-7d}Stjsq(V*01;S(iJlGSYyL}3lzs&w;L0cJtfK z(*XcTum9x&i1h%ng|7#-X1cn|?Z4|4tTcz|=?0C4w~f#7`TdN&xf`79XF2~k`3fqvf=I!iej&BaF@@wvy6!O-cVwF0KQdO=)0Xi>?Kwnig^PXI55i{W{yV3A@p>C^e5IB z>C2jklvVyfJN{YPg(&;POkT17D#ov77pF)1AwS*)>f`PsIxES8=0@uDo&(vde4wTEkZN?`N!nK)haBq^|BI=Kr5WP-*6;vUlmr8;n19 z8v%gOl@8M~?){x!33)8|TLpw%9^G2ns-JFx&f+QyixeUs`<;-cz@2+*M6@tJ_ff9l zNZnBt#K;VY?ARj(I?kVp1jLiZjXcR_T+aEA>^X`UuEOIDY$N4J;w|44`x&oDrY%)I zAq!T&xUDPkyVjmTMuK~65OV^dm2@x%5avl+MIcNk9?Z}>A4;kgM?Mt=6mWLFzP@XC z%XvLUwlcJL`jPpEJ);KpALD;^n<#9zovhg`7?@qy3?~|=^0OkNuL?MEpk!bgz-t$& zxWB#6xFid%f$@BWFNT%ASZV6tUj1X9u`lSsH&;Nrzh0~LsBYQ@2BRG5KNwjQZtFMZ zWyC4?QHDx#!HpSQ#D}@OJ1`^PVN8Ggg2WHoH>h6Z$)jIxPDnSdIkO4pk1B>$c-gxd9UlP%+hZKea>o&VLN@5Pv%oGY6 zum7DGuwFj_<}SBYYM-#WjhuUDo3p2}dV==YCySUPhPTWU=3MPL zfclA;3|`K^a_z5)+F@yVTDroLvQ`oxQF*x@mbj?k(?zP+@08p{`HH=ZUeEMLNjPM* zk!7}lLkcETl`I_NbV2&+aR{=}Mj&l%t$w46Ox1%7vZ_*9Bhw7Cbwk4~m%?dTcTWwosQ zBvC8RyQBxx-5)51QJ0?eb*1@B)(O-vvXJ}R8o*ajThQFf`__aEm+4_VY5gDaQi|lw|d%cug%L~K-C?k za}Dj+p_L^H?yGCHLm$i3+}BnA?}gF0OU(IcKb*ivTNM8fFDI%? zv~vHgHhKWXJ>4?TYWTzKy|`Qa7}ho1YP8Ijr;fHZ!EB*C{pqOMy0G;F7koX)O#9|Fi1& zxk4dL;sovYO&&?Il^QR;JixLlKg zcSn^+eZG9|f2exIRw!M%L9VFLP-oAwisJnS?nX3%`NR6_5gk|o=EWhvE@jIfq5QQ> zWdFsZ@=MdR{&v{6=Xk)uDuna^n%p}58Ed%`m^RO%uzc9tTcYDR2=)W{zmos~aNcp> Up6u{W6IKCyky4bblrZxDKNuPO*8l(j literal 0 HcmV?d00001 diff --git a/pkgdown/favicon/web-app-manifest-512x512.png b/pkgdown/favicon/web-app-manifest-512x512.png new file mode 100644 index 0000000000000000000000000000000000000000..3558c6159576c95ba653df3774c02c73c046e4e4 GIT binary patch literal 28908 zcmY(qcRXBC7dASG5?%BzB#170?>$kXM(-th?@S_MbV3lFAVe9x*Qg;n(T(02y$$0# z-uHg@zV|PFKhD|ftnxh1T6^zEO?4$aY$|LJ2!yAiEdLG!LIr+B1z}+TKhFK8Zb2YI zkczyFj$h`%5@yiU#l6(Ae&6O%_0_V4+^m|k42qU?!t)VPH{Jc;q2(9)rS^`cv>&GH z)oHvLqh8Cq=P?c}XB@Qo9&w#L+y})=e{K1EanB5E36wxIVm^l%W0hZ=AQ{P|AY|ivBGlOSvdbzO{N@qQ929b2aSUu%kq&z}=8gg5ikKeRQvduy#w4p5))9XS|bRrQ>GLXFbhdn0fB@X|N zUIDpI!5YmACG>sZ zy4LBeP0X~dhE9Hi-vC4?606s6qtWWEVH0~pPg=*!&B{73GfTK*cYXLX9(MQj92%=f znAmNFBE9j%R_b<(B-h{H)WMrVhlF^~4Aqar@NC;i5DcNA0#?NoJVI_Wt;wQFY79Xs zBIuHz(bm5FS@z*5B~(K538RNH@QE~8n{o9{AYPC+U-rbZ;3Lhb>T~3uN>;f~~Z|ns#H?21U(Soqn(QElT)`=gZw~!r;Q` z5#)LgY&Rw;?poWO*^ zdjrkjvTcP#s>IIWII7fSygq5gc2yvTz2}TYLNAw(EqyFDEu()U?iAbbXEdne_PHoA z>U5^kj&I2qDqTuXx34Jx`_F)dO4QrZOjjqygWb0es<2M^@>vQtQ&;EWh1N8Eo+kIr zTI6SUqk@$c@`Iu057h!6dd~4<2)cuDm~%tRem(6px%Jt0Z>{&Cj$+@>!RnU9i5MYu8oFWFCJXK5+}@8Gy-x z-jVblWKx)mMS|!hT$+OTw~ToR<9>}Sf_*!dW8(!6!Y2TydLbXz;rwP5KyJ)WAv8+y zlF@q=SDw5pH1rZjkly-$Co!cG#2|_TIJDRLMl&{TIWhRhN#`dO2;Nx)-@HGgkFZhU ztVNqv=sn+IO@Kw2s-=MOt>Z3tBerAg*KU!Dj)$ ztCU}e3s_=&_c0f~06w&geCXR{a+@aUb;KSfuIhFTsUyyL2qILq0{-heP;%DsgS>4u zeHZ=CeEkfa`X|(3_=QTA5f6TolRz$7N27zZSZe5lJyq4zCRGJ210r51tRQcQ6Q0qc ziYg4xS~QjOi2o=3m;eA_fut=LYeF|2Zb;X2DnP9U2W}Oy91skMl#UGJ0NB7F8S8}eYbMEnGIpE`A%I{)jcG{VkLy=uVvLy!>=zo%jA z?pKQ^5PbKM%QnM0?jKm(L8morRo(gJm8o54dVTmwH2~x6lNAY^v+p&S;zDa9(K z&gYYsNr?mUpz}a1Z%CDrPKx=iF)T9TA(4_5OMn{#SbeUwc5`aT2t`nUcag}yaZM^b zF%g)*!vB_Xlgg0>`Ha=uAy6VcJiA(K-l3||c7y@g$OnW4|1A&}_eu@^)9=aYSRUdz zNH37_KO0F|^D|zfr+7Um&~a>8ab_^w^iJ ze9eO>j*NIv?yct9XTCcxrba$gEJvXNbq*x+p*CNU5Pzpq`bm6OtyZvfh1}unYG(DX zCLkh){itAmH5V8)(J?)cp=|~CMnvO*|3MAjq8ZZ2*;V}+mtOsaI2i|L=Wt&#eF?9- zUPQ7~xN-rA3{;?UMrY)s7F)O~^`RF2KS;tL`qk(Yx#V1d>VcawxM5BM?tL}LLf-L| zF&3zs^nBY#1W8zI7ZP}dDxVcxvo36EG&F1BLCXPBCVtGYV*&FMn;-7I=xEYYw2lYI zbR4isjk?*@u=krdvwCT@Kq6gK6sk#(`@fJV)q4Z~{YTDivXDuEU7s-P6_ME8=^p27 zuIzzZOa1S7AO-=%IG{ZN#-fI<#q%%c+(rOBTEpzcE6tx-c2i8MZO2u0Z~Qd?X~sXg zic1{iJ4_jWg0?PQKvHKFQA$m1S6$9?!A}O7vXVzE=lvm~FD9B~Pq}tYPPv7CO}UGl zVz7*W-2{+JF6RavxEUF(F72W>(7(TQ!-E8Xa6s>JXeKkRD-90=?c(XdM%?tA^p{NU zG9a`eyEVhto96h*TkvvMY@AbnCk4+fJ|Mz%K-_eJoDqBR3RSHj7+FxvKzPAVB*t-y zeyaM-mHqFm#oklES)?EO&J~~Btc}cB@!I&kh6H5HS9OQm;MIhNtd2MlzKTHyz4|ak z3-!B-6ynlmDw@eX>#5U@7~~AP5Uq#5JR}5QEvN~tREYM|#CbLyBp%l|CB?ZtwW_<| z1rXna|Gz5h~*Vt^w&~F(4cH@6U2W1HfSOp+@zpS<25c9Hr zGt>sEotGX}n zVpBuU{$>p}u)p{TuywP*&Of^_%$H9>^A+Dfa~kLx{!&D;v*2MQvJ+Kx!$SeQxyqMW z-LIxQa#4=LHwai`daukC99N(fd<4d6Og4ympzyImFa3=<<^#<=fIjr|POB4E^i z2{1#;9{F7*3)&-AWdQ=YI?whET+`uu=$}uK$`H*~KAyzxXI0&!<+G5#zp+^XjM$I8 zbp-NV6u>~1$p30Iz;jVy!9a#aM9MvrmcJ~&+2mO^XrR@t1K>Kk{`a@7zHq>Gfw=PjD#TzlodD^ zCoR7mnA*I4>+M=zqu%_skRNYRsxl|QAal*M26f-7x*s)yu!Of&01JWIZxheH=u$<7 z{OobQs_uqXtjF#pGtC~xe$U;i2(U}f07Fx->xXwwAjQ6Sjru+Kt}+%)r6XvsivikkorHEPhD z5=Y!a21H17r&c;j>(~0*e)Zu?4c7mT3r7c|5gi6{G7m*OYEqS`gcJnJ^H*$G@*9~` zF|=gP;Sv|k)D&U|!hn?^qDgA|v2x8H!LIU^^dS;~Ws74G5aaH>;fSgOH6VndWFV>{ z+c!}AId;Pc$O%+3T*x7>a^cJiP^==6?X_h2I~G)b>}$$PU&vaoc}Ys zDz$uG%fHCFm=ZHWjsuGQe!r8UZD^RT_lsz9fm@K|YB&cX^E77nWqurBRov4HZj?v4 zWME_fvi}U(jm5q&%uX@m#mB}uIqP$-={N7&X=h;E8^Lx^TRDXK^7_o8SAuqI0EEQ0 z%z60Y&}}ibFD>uI{3uB5w5FGw*cV$2KJ? zyu2zQ-e*(Owr^G4>R-^M3EuG)&5Rnw8`;M6d`rRzgDB&{gKMbap+8Lzj=RcE(E(#T z0b@S@f1C!TI43s~eAccg0f9{E7@EY-{B!D^0mN#9Bl}(2IeeHo;gKkQ0Jq^2oR?g@_FEpDweYS2X+}bJFD3)a|8i6)fBY@~ z1`bq^#qX6TghUrb7Yp(07j;IP-%ttxC!279NFJlwU>dRiGrqHqoB*9DQ4f$(SM5u& z@j?gMoIZZ80poQ}>}4EeAl1Z$E9QsyfZQDFJMTWM>0|`%S+nkdFXE28eikW_c??!URCfzMx;P#LZdp zcw5sG5{WR5s~it=`xcXl85{iL$#4!IrNsezGR$9H{+HMEp=Ujh{m$V_xy=DnMn)x| z+F+&~+9MR2h5#M^Khp;78*PYs9D3(_2k(rB9;2(`>2}}2W9z%yDEBpNTnHpHGQFx> zroHcrv{bjNRPfzaDur0kmH{a9Ing8zKldnmxt_S1&`UOoE6y|s(E9*}@%7<5V&Z@& zfzsFiWl(ipJ)R~J0)jw{|MX9!bc(U{;cqAF4lkWTF_SfTJs3Ga&12ik#Sw^$s>fQ$jCVfo&ceEGpAe$moE1yFnh*2%DeSqpt` zPzu+D@lN7h0U%yOEH!kT~Mr3t>09Gfa=PD@uI z|G*vV;tGR^s03Qsiop%FJvhGc*h7^bPFoJC%4>8*%Hyg&(k)S<_8_!eCO`q5xwlvD z>4$#&mX^Q~&#n63>hD3oOeFESspAyTJ^nAi zV3^^z6-U`UY^i_1>&#}VENu%dfaU47mJG|vz zx!4WDjoww|BPvQxPde#fJdhR zSh^ZtPoez51N38rL``h&_T2L`(C+QK!ZyU)rly?_85hzjRm0nNco;pm2&RT+Lu7E3 zj&SmH2H~3XXTv^%m?SfL*IBz*T0tS{SC+qbx4v-VR7T!58CGlbe6ugY=TxiE^YM#-f=8TB*$x9Sx@sE?Qm)OQ0XQ z+pMQ-^zt(E=Z=I7km{uUo&A~`JsMBbwcp~q+F$16U+H3BA7>NR{k!00U=BX~l~6`y zM`OVd{FN=@j|{C*UUcyJ@I=mJ-^y+z@A&x6)BN`f=wWM{p2oZAn+eSEno$(dgN|?3 z^Y3vxXvKq(q`OR%uu$4wufFVQ!NYQFnaqC#twPk@bcc#OF3?cH%;%-IYNtmq^Ne0_XMWn&dF z1fW#$Z>2N3z%y=vlrR2qgkoj8q=23fv+1g%Cr_Cc*MDz!8S$z0LUhsN#$NogRE?0l z0{NPRNXu2kz>pEv&0PVldG7;twok?eA2n2RtL_NPU<*=JU0(Bgd9VL!?)o$rjPSnB z9O*`DiN@Bl5FF}gP-e#^@9ATmCs=v&QW3*qRY7j$=;%xL+)GO`f(gIz0R~hy2^jrK|3x9z7TR0zs&0(5t?4{W~laA}y#I zJG66DMCAJX*n%S|<)W81h&Ku9>eREL9_F9W1@U$;Q94TbH?tst>DE|AqTqd*I?;Ri z(O{W?$16sb`&W7I_~~g4(fE*P8+lT$aa+J+xpax4@=cl+k`}gyz zYASeXGQ|%ton~{J>VK!RgMc~2*Eu_F7nSL$z9L(Ns~gO|h*ju>EE*5zRTyk?;b% z0=(mJZ`FN#`j!0z;m9(?3H$bpQ5|F&WI*eivf2r?XLZ%jO00f$jvxIu2&R(BY-HuJvbCPq(%qJ{3eguu zFYzfMakl&gXE4*mrHtKu!CYJFPo>nV)`&qT*pI98p3X5%8c{7x2xsKq!9ltve7~Ze zx0N{B#GA9L1nmoGFC&A}3yN1p3kU?pt@|95dZQy))<@X;{?vp5GUP?TesqyoMuE-Q zI%$32lqH6yY+?w{1DWRE_FEcnQ!7R8Tj41{7)xPpz`QLy`N%+ zGJ4pr72JyjUd_|qg}3>r!>>|Pwqlpc{EGB-2KnG4UlinhZs@cqXkOKD(qKe*ci_1V zj0j5aMY4xmUaXx;25*p}lt*Emb`uL}bbI=aes#qcrvCW~gnEyA9U9Hm@ljJSir@T+my3ha2(6N!{iV{CN>&p>zJQoA!4LHi#oV(33!gFp9qybeG&oH@j_w zTaza|Cf+{;a2SsHLs`nx`79WaR|AZHJKT6OKg#8xPzdqhUyZbY#-B%B0!sCrf{Zh# zptY^JrwiTS8(Ks+gtC7248BNV=k$i7{a_2cy!_QNCT1_q z)TzzJOEf+oZs_PP1Z$Xj|@;>J~+rGHf{5PruZ$~TD$3leVdIy^70bp zhK9`#l6=GucQ$-3@9VHbVYN?TsvU4AFP$sibuk=li)v5V2d@%;#DQ_ITI`NwrCn&` zJz79MpLvG4L%4Z>>TA01{6xMQDSDWTLThKArUl_=lQ&%}7A9|5T`KS2=U@CBA>a5; z>-%SMu2Kx2Xk6s>^xFM^m9;WZHY6}a(uGE}wtKq=**?qCaCPC_v)6^5HNPGg8`;-D zNomrqWQ=)`*`7*8H}3o+yQ2B{?IAHr=%{p)n0d@Y$(b%Yj$mG=F+`OB@%BqTF3~{o z&{x~*dYDkP35q6zD;h55g=7VYkSOa1eRWMU&EMYV9acX0Be9ahPNpOzS&@no?WU%{ zA~a8G?c9WX57hSd#GGhpplbbRyWEWZXDkJM6r#9D+2pQU&y4i9#rCxlEQG5zZ2<<8 zI$^h4ahNpTN1`QIl`h;Hp6Jqiz^)0AclX59=bU-;*;?P`z*|ew+?q+wZ8#g-!P@9v z07Clj>EQ(t#e7MBM@AsiGbU=DrVbP}77fM5(UP`(X$Q)Vs@~>%P4lXI%6z<^$U9lF zh@0Q7>KAmZ)w6V7F`_5P!9KPr!8r}0f^1+U2N{T?y+QZ0*w^DQZKL9Lg>iA#ER0lZ z2xY)N`~(rQoQp-Q)v_4w9IMja1nKb1s=t@m4<9*aKtve`8R?I}`l zE!L-KT`WgD+%PYoNd7WGNpP*czhT7v85O&R0e~{adPzO3uN<(7X4;6 zGsAI`JXG5y?kz-JRvNr`pvduafhlE{kB!~=_9D9L&Yt{gFWmXYpMq$viZ}kc&Hlvk zgHtskyRBEDkw@f7OR4V==HZt7U&R6q3Z$g-?H}l?S8*w}E2OMd$-gHiKy88j&f%=V zB-L-`ST1+PCYt;ijp@Hwszoks%QQ>-`TMCnvIt%N9HZpL=W*WGQeGema zz5eekz#UK(SpO7-Mv8#3MD~SZ~M)U%+2;uJNjpH=EXR-oDpjS$2Hbb`-%mvw5EH5kLV(*+v2&7=N zEZz1gDyR%yp&v(!f_u_bjis!ykp8<~O;98)lhX5L3g1)5L)!Z2e>rRnr+@3E^dnHa ztP#lK%bA-u_3(IM_c0BcXSMtbf_;%QMy^660$xkFkpYi8{vMQ;ln;5X?R{bU9g1(m z4nYF1xY`%4Tt|lLZZhqsh>PLP4Z$*RIZqv(>?W7WX&U_vj^B3aO?iuQuw0%i^VY7~MVrvf3gY2^wx63m=gDr2&vCW;LjQ~4Tk3B4Xp7X2{@AdLrV4uKKCV%RdiM z4Q{drX7y&KZEhPmg)+Cb&{+iRPs;?dW94^_1@36H!bW7Iq1wX{Q3p@N&e}v6Ovd{{5GDq3jQg*1jpBmekTT|Jdt14Lh8_ca%~FOhsP# z*8N`Db$N?fs}HK}c$(m9i88hTAVZ?|v&5QBThG){f^RALwab_1$e!6FeUy+tHMCv| zQbF#r9qf$eKMR&}EpT8Kk>_w)uXeT|}EDd=0-i)Qy9)gunPkMm&^)PkU zbI%z;>?MwwFms4Fd$UANafxwbA66$#h!?tdu((YiR)x*rf-^g*_e4i%1{?rP_5ell zZGxc#rp|qryuxr}9prm(xNAl@8H~c9N49rxzE3}0@+zT4KdUtphP|KCm9nTE=DrhY z<3j-bAnr3B?srl>RONS#wHqkkH8TWhBo3<-{sa4$`LnIw=FTblaFqUl6qlaBF^oQN z=$tlXmGctg`snKBhbYa9T29UR5tL-XS$Kl*b9-@@m^Kul9#hOcV7vtxcpfQ+c$jT{ z+L4%YV>aJDd^BU6nKpZPHJfQA2F6gSIq9;n_<3I=Mm<@foioxx6GEh9jyv-4=d$P- zEIo8@G)l-II!3R-_Q~zMsYr~YP;`T@WxF$(i4#|S%P8zJBoJ}1b`;w9yme^1t9d1< zOY%|=l#iQ*xb=Nui2S0gFaUy~_~>IjYa=GIzyc@(*ahC;k+;g)#3Wdwa~LD{z4RqD z#8h&3c8Sg^ym{)r)c?g!S9dP=Unx?|TxWTMhy($x=@dtU~R`Ie2Rp-}; ztUJRWzO>9NJ?mr+Utr8#Vyheda~E@hFsWs~J;6_HGQvem9FjB1j@$SDIvQv?9N|36 zk-;J3+_}p~GF`-1_lLO0iUdDu2MMi%UAFD0_j00oPlVQ^2EHGkc>PT?*6#vRn&y*z ze2evcKv2jA7TdwW;ey)G+1bRxi+%4)&x%9I+cZA=mX_C{0`2TeU|LZ zv{LAZk|Zzv?tb*2;7=Q3w}&k#ipoP9rYz0JHW|)1U||Q)Ow)UU$vGoiS2K|H!Yj~P ztHvZjuC{*r$b8C8DOKs=$Qnz8F{Ct~+}ve@CSe`eHkHS>kXBZ17*%L>4kAlk`K|;(Y24 zTBCi-!uyFxT~Hv)O?X{!l+P0_!{IFtNrA@|Z)1dUiSdiiQc`(&K$LvV-9G@<#Pg*u}b&6wgnStK1 zvf>SYE&qI0I7=ogXfI>G5cS8z2%;Plz1JZyaOm2(6m%(yMBhWzz+b);y^SOVFA$Th zw->;?gh`Y~fFnLc7ozS8M_FxwC!M#@z3FtB!ijJ$1&4SzCjIe z*HHY?3Sl%W(Xah%2Gh zn(cq9r86D{%_D;fep=amJbZl-|9+0eZqm62-s`lyd>@=w zpN9gpbIyjh>X8~fQDG5I?aM42v~;3;ao>Az11pBl?1OPnsVDQUD!LV31p||`m`uts z<_bZ*qkndAhMWfeZZeY#`P1w+ROw8ji?BTKa1mSano>&$=EDQwR z-gQl#QQzp-WtfNw)RRkkd)t%IPKNV6!XqK@1^rDwcA~F*D|Ui!Ehsjnj|;X=zZcPN z_0@fv$ur^g>!$^eg*?SBDdzEno?d)*({sE8emkY(FV)_r_KbKl17t!JfwDjjvIKFu z4E7}yOCNCWG$t=wpQhq~{HCn5WvH0t3%&d#d4mWA^+>xJ4mZu75px{IV*2`V*`DNs z2*dpRt(t z+LF6DR`n(_gnZW}7UWD*?6q>dxu5qYsFJPKLZM~vlC8hxDc{!Vq%URcQzwMh5`OE| zwL@@j!kes?=xzjDMp%u{X2pB(As8hQkI971n{fmvTxZ|j>Gl!d8}68Rcs)qiP%kd? z7h4bdliV@m&kvjJETQdYJ<27Zo$I>TEzr(=`OHQL$1KnS$6(Qd554~#ex&hqiQw{= z4QdHS9^BGA!JH+-5-Tm+rtERJN4DRKwwqCbETaRl!SYs!5ReG4 zplr+cmPSuldnnjVW=!}EcKTReKeDN~JOLRJfm1LS_)$D;bT)|-?^firtW_xZYD8mT zF(d53sObrc1#lMU5qflWGKR0;&BMsmPAy%L;gqS*ds^1(c!}&5OiQKJr;4#tq4`d zic9|HeRJOv3#R^3l;2)LS3VIc@7v_g%X^$;dzO!8t{t|zF%v-LIQF6(m7l+1AWtHR z_*sJWo3A(^_aztU&ux^{zfHz3c{LU2?8g`xRqn*ZNGSV&;xgkV?puV5dXC##pAjoj<23AR+w z0+Es3zTpEBnem^j!Yk9LRvf|8>51?8sZ3k1dBlHz0_W5 z|MRB573F>tad$X;F!^%gO!*icB03ir6MW>S!jkufqyx#R_Y2gtw6V%&^d$dO1CNhl zrPcaT^D6R)E^LK%2`cf&w9Pp|%q%Yk(Lsh1W7e17F0UOoNx_1Y-@cuCcWJsxecqGP z{P0RnF_@W4u^&|$x6yg%+`)S{jVwSVV%zemsn>L=b4AkCZa+rc?y+&}%#Oq7l0PpF z&+HaxS{*~(c(iZNFcIDyAdG9g1Tzz&uz;;)Cu`-v)j349mI?FWj2d5%Qr2)zyekYz zl==|B_Uzraw&@jMN0l&<(Xv3|qH4mty)HxI-g^QWxQ!}Jk+Ts2#qsSjF^f-MTooK0 z`h|N;P$BzXj`COfFTxc{aF6spdrdWVtgTcN$SoBc(#SyYCXV(6=I!-S@i4E~?yRwge{gP92EagwoSLM5nGue4vsC(G;krUW|QZjVLP{c5d{`}|U z={uB4a?|1G6Xs_4X58Fw)FOMXb0uk)b2gIvJSZj01x*4%``-tAm+7@ z_x7XrM>jhs%jr=|Eq`)M9Z~X=rP#RxDj=V)x~7sahfeebcR5aJ99`GiDlO)O!2>8J zDyj!xEbD$z;_bw<;Fo%~_pGF#8DBrav5pRYFjT2fhrusEe~tvi8N19Q(A=$`K4zCT zp+N>B@9E7i%0sCs)D55UG)+ZpMcZ7Cm|XOzgm4pZ^YaTS!$&J$fqL3Q7G8&>zast- zrs#VYSl2R=%3hA@#RclY4Na)6rRT!ClVp+g`xR9xmRXxBwx&WKjppnX5`wg*-jWa| z_6h+~Myp}o_HLJywq++cBQetd9+Pok_Qc2|H}n6_0(36#P|yy^Kz(Cx{Px}yyZq|x zns|nbEg=wwb9}B8Wyax?7B9DL^3=1*8IG*RmIz|vJdH#MG9LMMN$+O|eC5ZAp1fV- zZ)+NnVKZW#(W?cS%?3pr752Oa_Tk{S0V9)cfetJzvy_Sy7MT&Vt4-Mo#~e~F=Ry>716 z|D>(&7UYMKwzBHb_iMeS0Ci;oe-slz&@=vv#zE2bV?=v%_ijknELBud3XPxY0%+Hp zQu%V%&(MtHEE-jM;MZvNZZLN{fSRA$CXs7Rte_`VQDZmv-DqEa6jfuG_MgvI*~MR& zFVpRp3fJs+#}~BLW_`~+15P7CnvB>pggP(}`aA_f(Pg;?-w9rzqeYnf1v_Y1J zQ$BQ;H3IQ(XgNydYNNN%-W(+_BL8z}+goZMNJ=$CUFCfhF_AM$BmZewks|U5O&oszT=x(WYVst&s?&oD4#Sis&9nz*7xbMyb&QIO~ z%+z)&(=>D&x33Ap!kb_NcR{h}ngg5$M4}YXFuP!Gd$=-EFv-(b0R%$6xbYWtDySrg zJ3w>=aaCBq~|R7LH-T2B#NM?fRE2sy4I|w87jMsL{0lJ%+SBS zx{EI=^=PC{;h&gz0{WgW#GHYXL+no2puBrF@K1ft^N@e)p3XBD@(WAclM`;!xiS-= zWqmHtlPuqhKY_Q-Jt6>H4vfZ<8ssF#g`;U~E?%WBca$s!(83#Q&FGzX{N^PU4C`TF z`i*$Lz1M(-dSUe`C<}@#rgRhvX5w6@Zu$FzTTx_y_XY#e$#VQxl=auofjIFG*e2l6}8-rnTeFIQ=5; z0&&U1ji*-qEA@HqypgBa(&@Dt^j~S2;iKC$0}6fZGzIWMS(8G6X;-PZuuSc{T!XQj z$15(fJ$Y^VzDEbAO^rvXP{dQ+x8FPlm!^56H%86oLczwMTGhbow`71I{U>%mQPFzW znpg7PWr!XB$>oW9qQ-i3(z*VNQF&Q&hH5qqljlyw1oHOEq^Nh|Zo$>gcC8=V?~m1b z1$9&vq$3NTSlsgPr%@-=GZFXK7czOO(RmP%SUTQ@ly!__gBql<<0rIVq+iT5 zHO~M3cJ-#*$YRWY?dxZG+-TY;Z=0N<(fDcF{^HI=+K?~plJ}Jsu+icLqj>%J_ONsb zZ{>w=PxA$<(DG%ryD8t{wi6i0hu7?N;aAri0PBLa%sl42!msG+^S+II3y|T`pun*WUbrKTzz`s0Xz!d!A zaGjcc%z97w$A>aJ#qY21ctb>KD#T){K3Z*zM)(I>smh`XZYbAW%NPy?e2$B7yRcAl?$**y}Z;wzFDM==WLkKwsN%mBJ|FoFCJ)( zemyXJ4hW6UZ>I9|Umybw$F}pl7uPCv?aZ6+rYNHSk{Q-iIVF&L)Rtv`cLP#=Emn_Z z`$3|_;aCBUXB$b$un9;E5Az5|d@J1}e1~JWF#89W^Aytnza5TZuhCTEdqu^pX3@>{ z-7q4Vu;cRg^PyK{i0gM%V{4Hq=h)Ko8Loez3UB=dK16?T=U}0ARBgx9a*aHs508*+ z#Vi2Mz{%t9C4JJ^%)a_4LuWr6|B+f;RNS?g;0tfAwBb|xo|cHvuh?J(+U!85={*0M zr_-1p#{I0(4A0)sW;cL~ukdnGq7bP|BZZ-)kCrkZqM{S2=(s{N53-!_+T30H8cAH8 z-tISI!F0%RmC>cOWr$U;=zqi7{iw^|>kfz{$LGgb5;`mi#1D6Wf|;Pa!vt6!Z;0Q< zLVw!Dui?r1Po}p6#brHGMQ`PjbLcUy@jHcH!v2_%(-{+V7&Cp~z_UlbQli=^9Zxqw z_pGliFtf`Va+M>4&~$-Mg54H6T;zY7*Wc*^C+QjSjC3n8(7nI0LX+yZ*vs=opsq_F zsb!OZi3nke3Fm-Fdc^(r{_%wAPt4kA5Sm2DB`tr66j$>WYRq>D)c?#!s2F~GE-p#r zdWv0ys*L~L9^?%{={5}aGTzP_v=aq-Tv@u!5&EeNUF-JhUi(Z+CLeHuqV}aX?0cLf#h;T{ zoaCKCL-N{B!Q2>uC+rSrz=2Z(#4rkt@H>e6k7qm-fT0U5#8OwLTICO2s#D@kAo9-pBiaB6pFGTjPES@g?{;- zl3M?c(RYjqXe1iwcrDG^%|5rjab5ZmFS{t+cO=RO_~>~3-@1upY1wP5;Tg>1^(}tf zBl<{QrnN+l!_q{-Z=3on{lFX!;VYEC;ztCjeId-IoIb3^ciqK%`mT*)=SRW!VEZ>b z0~e2nXb}Y4Kk_HM=Aoq(0#;{nZ*EBG_M)pMqB>_+pB`VW{{oXa-Fi{C`#Yjt2K!>k zw{iOWF)}7~ZxjYreWssdk)=AfY7)|*eSDRa^Kj=J8&~-^ee@njX_#9RA;otp%sqdz zwp&_B#5OtWFgvD}22LdYSLhgCyBp|PVJIf#qbW@flPECwK$qb5^W_-y(p5e6nof;9AJl9F@Vgc?5 zVEXNY=~-YuS>90r$A2wf3t8@~dk55{ew$g{r;i`t9@j88hbiMKM@r{xX!=n%V)|9I zb0D$V;WyZuW;&mCf^MGNG5~XipmHu3bKGaWWv25Z>ldPXz@4dpKlMxR1@#}M81N}v z=2M<`d`{Z|oxjDBgC)DY=#S{MAN!~d{nbciHvI5(vqj{3@;mXcM~_P24=4&RTA9hQ zWp7r1!x#b98JgDPrjMFncK52(%Z*L_KTy1ka(#mJ0?toF6FxGQ#yt6qs0M_}_)|2W z3wu8$qWqKox-K?xa#d<3>;D`xbkESwa<-%}aChd3gW~AEcYzIHT-R!?IcAqagyfy@)51zCGE2K|&& zoj0Wx&Vw2Qr1Fdm)MozUQ^Z5Eo~90GhkOT0m-%r|@s#j2f|`ixAnce>N?utl)l`%~ zxOtDNCdf3uwV7^{)O*55MZ#mzNb8w#cu^6yxZ9|9G2%J>M)RivU0&W^PpKN25-MAjK zjI+${_Ma{`+JqBWg8v2wc-R6yN{erx(~#sRrfw+ugXU<`W_1aHmI)3)t#)6s5Kz?r zoC#D72h+Ka%e>3W9WmMbd*HEa$J1r~;q%-C_)A|3Ux3+A!TrR9B9l2kY8EzlD~dUJ z$?Rhh4L)*&RrvL>{3t@%zmq%|-V0Zdqh-|ZM$IJ$4yW?{#7{BHsJC0`H;Wtcz0Qj4 zo%sChgl*D1J2>&28}Zq)bM*5+uxO+1%|MO|Yl-6APowd7{ar9h$mG=;e(+s_0WJkh zUov_7us_W?AGj|PF!y;}<#d0nit)x8NrW-Lnm=9hv|P((B0E}WaY&47y?OIXzChP} z7gV5Lo(uH%doSxlkeFe{s1xbOS~=q?N5`}z2Ga>J%ueSuq!_qX!f$B~!@AMcF$@ci zIIVhbNXc4SjG2{@qJp`)HHVw55*E6R+F77fdL0C*-Fv3{asDb@^J2*&;WD!D_KExS z(S~=6u>6bgFvtiU0N0UN7jiw}_q=Zd3hlmG4pnd*EapV_VO(Tjh!8idff$;N={AgF z$^6IQ3@mR#^e&*-Q=zGt6J&6ZtABz94H?Vu!8%c*eQd=Ajs-lBv9J;}R=M$6^QO*x zi|0TxE#A51wL8ZT66{jJy((@rE4`m8Sxff&I?(=wS% zv={@%y|9Bqj~UfQ@cydl0$uTEsOus7A$&h$jLPiOmqV9{zo9sW)c)DBD*1e- z2F4H?uSs90YW|g<40^ZQUm>-zzy%tL*9@v5CQ_mNK1V}c?Q-bC+3%y$4f{>Dd>l`t z9c7;`wicj>72)m}gQh6Q4WDzKVNRnEO@uro8te~BBcOkhqCRiPrV;4nuKifg^<=7; zkKFfO>j~C@Vf7HyhtNo|V#sf#0he0x5=-1q)mT|8exJm=D^Tg}tcrOy>_g{t0Y#_s zuK#(_`?{mChF-_gkTY{sVtQ*?Qq|mMQNLFR3zL zO0Iv{1BTliqY{Y_kb%_I3vME3!Y542G&OZzlA|iT6i^=JN(u`N&@NRs%%*=n{P+H? zw4U%{oOh;H^JaLa(>$63Sjo@}m~0y5HidF6dJX?`2CJh>A8NcWULNS`VAl#}-0*3v zLbKYSY{zO}J(WmwGzacm0f4@tPhC?;H0Mt`Fz`xAPo1*5Abx)@z~x(K`8YAB>07rX zNtf8KDJNp&pbLyz9sh35T)J}JLbrV#(12M|sO7iLeh%o1MqML#lzvDHy?5Yi8ND@< zc47Yd_^=mIGS(Adnc8Kt7D!dRnH2xD!Z|zpWj$U_HqJfX3g^!Q-mu1 zPgiFi2zB>8{y`{9mLfapsg!KV78zVFNLrClojNv}m(Q|q+A zuX){*am$(SuFx^14Z4P^&EXd7+;Eeaz3p{s14AW5&h0y0&cCL%_41V@zgM-0rMr+s zrfa%?newIWl7-1Jk^94}dK!>Gt9~aNoq1)2u%y72V|LXdq<+G&LPSKYKrJBPHD8gL z$=_#pDr2`U&x_#ZZFN@cgy@qgna56@hwC09bZQ#_NoJCl|~K6qnQ6je-^ZqRiX>L>X;Ep^E_--F-_KL5K-dCNUmCwO7icW%|HKxk&znYXa&CyLw!?Vy5AF z9D`Qv(~*|xy_^Z{MluFpv=%Hmv43v_q@JBSiE+PB*jlvdZO8S#wKNB{n{I$gEv?wU zb#t(PTL^#eI9k~v?&hlhJ3`76LBbP5wNRF%&4bB1&N6(h&h5iNM_nmA0qRFXaM^N* zgI%$3^Su<^VCh@uFSFworQw9eUn(|QPn4tyxmS2>o33t`e&&B?r?tgt+iln9uUy;Z zOO1b%!JWZ5_+)L|_txuI)#-Y9(pg^cT$!{fX>m&0SfBMNb^bCDKw6_LwTdKG&|Z`8@st@do03>#in^7E(khP2%T}3-r=$-9_9}Ga6+Ff zbKcbqRoq&?zUZ4^H+}OKol4GHa{e_ft=U(1wY1iqL+)IBy&xnw?e(_HqOgg0cP&Er zSdQ|SX_H6dtmAZZqmyn6YzaTrS!lL5A=OZgO0V~t-Kxx+z#fC3`Ci7tm4fJ!d5J44 zl!c!yk~8&WNCt#e*jF#CSle>=+7T?D-MeaPV|($(HzzKy0JMQXjJ)!-zIy)P{%Ot! zJ2?qMHtjV$iZr>b*+nB=WpZC)B#Uox+5Y5H8I)DI?pa-YUe9HON@-O}XS=t8 z_6CHuu9i|6e@#>@I%DQ`Y@WH3e7uw@@>D4t_8Fn(na7sz>(g^D4Aa>w(hKqB6%EsA z60baF;F-I;*N>h}u4+U2H06geKRMsmRy=UyTeZi>{g9dmBd=z%K(~3vrmA)I>w)&B zk+N?M_N!GL)(iP|?}})x97;G*!_HKDCps+csFHd(g6io8O-hZO_ja2oE_0YUCAVn*F$WiM|8~S&sIwgv!`HBtlNS?c&;rePf%x;+GklkK5Lz`C;43dBID4t(IUs+Blec@|u7bAsMWBkyBsO2)RoXtSKic^jW=-NUxrvp#qNIj>vyghMHO-Yu}(?pZWH=O zKDsC^vn&EPY$vPSCK`xIcYtBMNK%(wrt>3FvfAiF{rVRvWrD^Fc_`zTc~Q4NDDCNU z9Oik4r*%8&lvQ3gVoTHJxHSi#cRH+C!Q~CdN0E=5^f{aN42XxcczVYvXH(=osU|3Jb!>Rx$Mo3;x*-YWcNPl_Qvwk^T8Q! z!|uh%q`HPShT7Ru)RlZxA15lUPnxGa>AwfZdNL8!H2d)Vdy=KxRbyWYtUy2!)K zWSPH9I=@l~u!#>pQa;{ppw*=OFXTSh~R<*lz{+ z%-QbTK8xeE_Yat%7Oy)|6(<=}Ny;Srs3E;dV2ua~nx3}48#6kxs_9^Me#ly+ut-ga zX^mTmp6e)Yd!S%dO>-!^b?f*_H0Sv_{}BJBN)|#>ZNx4looK>dpSsW%t!RBSu6>Y@ z5Z+%wr90HOJ8x+}NNqoB%~+ea-u?3ySlfWC{yRb8=t!$q!*QB}%!T)(Rn^iCKN3N` z<;nRWoaQM(?f#fz4q@`8IMHOa=A={h*WGM|o(FEd^Ph>qef$Pn2_bjd{EOQY{HJxsXM7lY&DgecIZ<%_YF*H4@-!&BRvq0SUP=v;WqtV`lR`5A3B8mPEm zMfZtr+D;}D8+$N+J$urU zU3_R&rsx}A$%onz{DWBM_(hMCBi8Q`d&vV2nfMOleLuIY`zOQxcny_!LCE#x zfL8UeuP91ywMr5e5Ahs0#Z1QAJP{Q3@P%)+DEV}Kfy*__c??V}wUcuK=Z#MdlGtNc z;e6)YL{4u_L#9sY`JD3^suY%FH14A<9f=ij$yMHdQH~GqjjLkJWB;{kG)dR@VfIoRvWJgS&+c4 zNPj6fG8&p}k1FXeKde80R}tQ z2R6#$c~aEM$g95E@n|{>CLiH|ES4N42>;YoE}r>JZ#VH?F$=!%9loIEe7*X>oF!?$ z=^6KNd2`Ctc~&;$1-ci#Pmlw`WiJFy(5X+h#ko+kv6;a0O*^aGl$4t0J$cK_#JVHN z;x}?7*z^>EH^G*yRK(*@c{)`MPg+tAZ7Rg0+4LX|FjEylgB2t@94($HwR5el-4~#W z!SENNQdWue6Go|OPYM!(xP8p2g@xZEz>hDVvH5D#$_r{cMDAyyeiqft7>q3(TA#I} zbx^xQS=LFBQDvjr+>NA%^x*3-AV0Npv?wwD+bHV#oqs>F{l02Bu^G)|A7xC`;dzb{ z7XAf6gtNDVXEYvUkqdUPMN=VI4CXcr_DpWI+gy%)@tN~l_q7FZbIHc=^jT@-54i!e zw_3$^NnxATp9b-eQ2a^(cp}Z10?J05BcYq}xGa#FqARd{MEEQ(JpT|Hx@9jZr=rzZ zgZhpS1_xYehDDmVC{3!`Q8`$L=egy3*(%%HhFLf~ppKc(ctiI-edoEeI1XgF~l4-Qfy%@l>G0Y%(+h5`8a9us;kzS5x~> z>(~zr$Z;F&FyO$ge|@xaBRPpc9s>Qf|c*ANW=MfM`R^AKpKZbQWe~!6d*80E+M3FAnp{~#c`zd zChFRzDq^i-N90nYr1|dHrD*(1KxxZZLq^8}CqVhra}pl#8wFg9|;z zK$7eGfu}M7S8G$xlrfoFElG_{YoFEiwex=5z=H!N8~NRN$S{0z$h&e@R<>AK=<8LuKReZoHWj+E zW4Sx?41~S_A&m5qfNP4qWEV2Y3qEm%{)EEre0X#*oRg_MMdH|3K^kFUgzs}@Wnl`}Y;rXK`%UQl*NJ?!u4q_E@%KGnm6j|u{f z!0A)^={-_ud(Rc3YE%IfBJ6ApRF@GyZQkRQ1NR%pJI%6?-b1Y82`fJZaVHq_DP7di z`BE4tcGKCdWF;lGBbflgU~}MQL>*shASFz7qs7XJ5nD))V1)L7C>cy8t4xzqJHCp>`jW??z$tiX9GWb>^bF!r+wW=?hD_LXesU^a4+1f&|(xd=_1P9 z^OY$yTkH||jkV7yThjJ*u=$LA#6vGE?7;^ucDIX`+S?P`pHMKE==-!mA*Rb~j`np; z_?Mx$bT{M;@Xa9EfDgIdtxy=K! z6ZwFqZ)I!0?cx`8*C%!q{P=t_?jtvd0!5J6-D6tY4`Nt_n1lN$#q|ix!?*;fAzYVd*>!c zu~pmd5b3zE2gJJ1IIiPs&O58|`zFNiwpaP_zz|e`8<#S$Vw&qq*0R4hvqd;C1G?uS z4o}-RjT$zXyQ;cA*&q|fKS64*T_d%M&85}PKN_&6l3F}F3P{7?Z=AVVp*|7VG*4l0 z*a9n?2!DSsBSS-VXWGH=rIHHTR{%OQf^CGNBTB)RqgALz=%JbYGE*KAW6Ea(kX05 zaF8T+>z5!r3zwIwopyP7ANk1(n!3T|^f4qp`RDWO{0{;Ibbizm%HeOS@$M~0K~%l9 zu$9Z(88q!SAO?=bxa~e+gqHI|VAFNtlX%n5_JR;j%tlmWSg$z>#KI4oqECvHGCtGg zoGebTe&&o24RN#YlG<76_qJEXOMT)x=P3_zErGo$Ef^|+;)(lN`LKvq2*5dwQ-dG+ zh^MP;u~p;RHKxQ4*i26Og2yi8SxrkJg?JOm1(YrWq+gBHDf9uvb_w0+1$qUx%U8wSa9@(i8g;St_QZ+3I!rY1MsZHmN$K7f%U=y7 z>@z2mLD%aw@=@>yS@?tBcu()607({8DW6Y2e*0^GKHuIfNVXY#dA);yqKp0-u;c-b z5kPc4fs0(Y^QJxZs2ur+BUjZrbP{k!!MQXQcNUE#F8p43SG*2@hX3^rrEBbbS1c)s zQsuv?qve7n8^P0#4b&<(+DFz-g94==ZtBWOk?|BP=Y!L1q!n13&RdXM2RnF@RRq7n?G5OkS2MK=RVBVZ{NXv1 zj+b-E<|sj|>%X6ilHXQmQ2MC8d9;Hb3WK#w%!1o)eEpB<@+x)}EJ#O$)A9>Z@;WVL z85eG;TgWnnh|hyq)AE7JqqEh*m?|TO_GHH4a330|$CM1>dQ`OX1J^ELGB98}v%v>! z)4#5EuGo64G0=FEqyU{f3rMP;7qfbsv0aGB`YZ;fj2^YEkX@KEA)cSgS}VE0QEe z#v-d_*$=nNb^fd?l!HQa`jLuJ+>~DF-R7uBjJ`8DK@#o9($_C({71BVwqS!=!$jS z8O>s;ow)@{w*Jr{fo1$YUc4nzkl4U24_C&o^~5i?|3HruvHp<(+-fryJ?0l(?b5P# zh}D&Bu?P2qkCM!#&q>)mpoE{WyY~I%_^HH2P8CyCg+;7`3~)goDgxWs&B1MPpV7I~ zhZK7j)CSssw_>DoJ*xWwJ;<`#;F?|tC4V4NaN{kBus$+zMilib0I$d|cWzf|WZUuh zsi?~{`zVCrN>v`u&wQN2zWbyCNve<`iEPYCt2&NIw z%a#%C`B+ntLF?Dau(VFJ0qm`?!bvX2m3>Ri?`Gd19-_UYZs`o-_$Vz+zXAbP8x>P0ze)bo>4f38iv59IFtG zH{Uv?&9K068MZQ7BjXXExg~s`#Ogc+DGy&)Waz#u=D40C_p88UeY5pe>S3T|*#3VY zzKOwuaWMU9;h7YdDN6-0huk@mP)j+Tvv7e@~x`5DcVDj z-=DF?Rb?uCVJF$-TOv9BO#b>;i#3yun^bP~DQ3yc0{>vxPG!VdBLEYi66ROj5$Rq7 zr{LYMEw7Kt@jlIjW#*N-ua#LhrmuHR@xneVy=tSTKWYO%Nb>{$#4Bz-7p~s!7@*<) zAt&HJS?6K@G!qo6OCa#*_d%^P@vQXIbPNynt+xQB%ckDS%WON+`JQu*k9E%4No|S6 zCG9m<94}W$JXI{0e5#<`4hD0Qot^~HF^;ewGPZ-I!B-o(kVn7BCJ$m3TZ|0s>TcaCn_%Sl z(8N=HJFoBtSz_T-zVFQDgAJ206+SrOl>P_Tjn&A5Mgs28so%j^Tw z|AqY^iP&4IYj@S#$=mZl@Ae?e*PG4@&!^W;zkEW8L^Y|G^L`_)}SZ|0PR_C|i7!D@A#o)-y#LcKOof zM%U@+3z$c~*U4a%b^yf5J9W!x@0#=uv8%;3@FI^ghlStb-78m7#!(+8IS5bhG(6}J zd-g!=a;xq@NgPuHbJzWbzK~D9H<-ei8w=}-1{Z&I`S}VdpLR1bE!`U7FBmalIX-@k zRwdzITJ^E&6k(44h1sQzl=-gSI7RZPJb5u7Ex`4LNyB`yWw|u(mF5$mDQS6bK&KrJ zUg2pX5Z50tH$BcdmWIup&|raHGXYTer=b((<_+4XoE-B;vu}EbQ%Z={mkHo6e zMFg-)0oMeFl>JVf->rsK|JMPaaK*N>_&wG6wYt$G7e>3KRGOx+%I&aYfA4@NnR+br z|A1Z2Vaygu{%xGMucbt0Lz?R|Xf7Brrp%k_Vt z!tC*aIwi_p6e#MZFHFp1sl(LvkjC>xnCM?mZ$_i z{B2*yrnS-%!nt^Z8kH|n)#t)!RTKeFTv*$pzfC6pP4ff z%`D5v_`BwH8NmRy$cn(>69C=_{)Hcvb;AybwQ3%?I0Q`r1x(ufL8bAk>)kx zK~B{5M>gTV0dZ#IIhpAb>JGP51|C2VN%U>Lw9Ri4<`(PI>5sqrQtN;#t=CJn_|ml^ z+N^rxULf768mBWiV+2vc!bWL^+691{{*AxSfBmO%JA2}5*DF*n<8s&(6WKW)l7PX2 zATn`@FMdf?FBun4x0gAgUU4!86{We%FFhUqw7JeP**JU+!o>#^GB;LR`#ePGFL{>6 zk0jMokMRLuPIGI$=>PvgyvA^^*QQ{G6>gG2Z;WU?s_z}P!hhFGEs{1mh z<#$8pZp>b7R};9{KK=aKsEwojwH^HG-6wS8&%dbG^=8wv#lD=1ACWOD3?xa3vTK~V4%V4Q$*gB|40JmL#f?47%hob+& z(FV7M;&I7eupd8r0E}0Zgpbwwu^%?R>wf^jrA67lTtJt=rRGV!Iq!x`mUBrsM<=bm zLCdD+qrA0@8ZpJ|m&A`C7(6uI%}zGEw-D9AC1Vr*A2_h;vg>8c{%EP{{BYW!VyOi)9db%GWcH^wjJHA9?ZGq=d27>uk zU}5u`VUJ{H?=#ro3kiGoS-9Oz0VTagXkj&r=rRw~CctO2cg@)>IYuxoV{fW? zZreDG&KbiNTm-$$k~Yq&qt{ zuSF88)n6|jYhF7`58;#a_G*b4jr;UF*Yo;r6drY(%sA4NB#6^A^93%Ro%{C-&_5;U z-(l!q(|E0!gkV(vBz+LBcs#%%nNscQEu9(H#sF~-Kx<}D!g{m68OlqY){N%Aayb$* zf}@j(p!|6vBKmIvTt#mJ?({D7&^OOzHTlk*K#6s@8ck~D^ut3wL8_Q@7XZX4WNC9D zC3(CoSvhc33*J3sj4Iy??_NKvbL!>elVzNoV)r97`Og9R5Qx-1Ejd0tA&C&eJ@guM z&Nzt0C0T~9OMRV+?ZjMHa;{W$$vU87A(hfiPuROWBg5fu)w`F44$4X@GXbqLh+GI% z?X#936?IyY!3Ur`+GrQtf;h+L=i*+LGqFJDE^LU4nmt~ych45LD#q?OK9@%#pp+ZW zcTFm@3|79`Q5rF!IAH5vi{dBt%()F0kC!h#z9E#>Wuq;z4p#g2V;n!RceY0Nb)Fk+ zJfNt;ewb`QxOub6AA5n&eEC_t#kZO=sGo7n#`g^bao-;vV33PX7y;||KV~F6!&|!m zq2HGhGTIQX^vRa}mN4r%|G~kUKf}!i++F8%>U*_60|3caAjIsY1qB7ruIF4{NL>}D z!YbhM5l3mH(Vu#KO{cB)KqkDvVmr*w7wyIflAo!xV5jP zStaw)lZ|W50Q~xsP%0xWW43aJYkKc|3m-wkUlJ| zkpasTuw22Awf?Ow@k$hAIf5~}k#e_!r?6qy`hxrDP6XG}+S2+d`D-z=VJ*TS7=$H$ z5Lux9%T#FQz-yvI;jH5lNaIh5gh)00`=8kwF{X#Ded7Wu4&=!4hjzkajsT*63;1&c z@!83E@@gtlbHPEz>Qp9^4xOGW0)kgKX>SQ0ms9iFV;&fWx6?|?J-NrUCMz8q_p@A# ztbI)T|d_#<+Tu~5bGfJLD65b9S15j)mXxRzH$KC zmQMsz3H^$e$@GzU?%J1gN4p>)p}gl%)wq@67B$F*6Y>=UNr}(cX!Zs>Ym#8V^>c!<=MD zr_A|oc&9nO2JW;QBNm|TNKQ6;d%%jcjc>#N=*s?!#T)`q=f+h1M$`8whGdqd|FQus z2rBs>KX(SH<^aD6B1J;svg@%%BwC)^qTM+D=>5wrNF@$nw+gistz^;M(Qk?$x#T%n z?@IA)(PLE*K`Krl!}f2XoAa`(IqZA(moDljj2Mh#!*Err(YEB9NJ->MZ_ z=WAYLN0K9_*ujMS_x2CH(QB@B5dTt{8mi}{3&2j`^DnVYi?Dc$*KzqWvR^3~KU%q2 zPG`oMQ$ZDu6aVergzP>n_nPK1S1uLQG(oJszTw^0KP8(E)j>s*$G~Pjp;u?@@jdLf zEx5Fal+XD9t1;#hoB@hElMlF#n3>p+p%Bd{MutjTTfvH0UEq$-%xC;5%k!4K%kOOf zK};g}qZ$lOoE%b}>bKj&rsxE@X#;OA{Fu2>4aFer5cd!&^VAGw0s;m+T?8QQG^&*k z>hx_N_@Ww6a{=ho?#oNIy1igcx;-In%1hWf+zjEDe>1$Jd3DcXGFp}A;dzy;ohmdV z<$8(Q;&j#-5L`c;Y<(1{$a9-$0N0_Nr&qA7l`jNk+3O;j3u=TftNBxt3tKhofIJST zgWb%C)4*g=?!LX(Tt;nK=t3)u;-7*vxfa?4q^eTPDZg?$j4hmOrAep266q9=UQF_% z0y1uhDCiJ!1OugW{&+dY<|aQG!=mPSqv$ySfk17SjsV<8ktoUyMBPB1w@d#+d@gKY z6z6%EVE6EG+1^A|mZml3X2Aa=2bZ;a1ZVfwfC>a|JGlsab_Q;IEQqGHu$zs4! zbO?l__#j8%EO{c}vn9WfhaO}wUPFXpkN6D=bX^)7s;&pVRe^W=n!w$qHDwRO_eQgE zzZONp?1_hQ&P|*qxS_afy?;UndZON& z|6K-ZO{WOy!h&a|0OiFhHX5Q_0v4jBzjj+v0 z*#BBzbrgQOcCWaFGSXWKdNqYX8esosSpKPVGau~T1^u*lwsn}lFo~mOhOkBftTx_4*a&}owTZl&dEGi3(GY*+Kiq7J zL&rzTwIfRFoq=F$m%s_CfSel%d0d5HOp0pRgt<-llh?sSgO|LA-h%>o!t9-zYLoRg zZd~*IeX5Aybh(t?+fnu1+Ow*#Tu<~SUz z2Ykz_ze_$m*SFi+{-?H|3=FwBlH_;j%)+?Q)eLS5eO_zcec;1axh7S33gSLy>|mii l3WX9^83zBLLmH+*opv6Qe^V><0X{)pRJ(%7KW`QM{{Vux+Ef4l literal 0 HcmV?d00001 From c6ab58807b933b728b5777545b5041bb5f973856 Mon Sep 17 00:00:00 2001 From: Fernando Alarid-Escudero Date: Fri, 13 Nov 2026 19:56:22 -0800 Subject: [PATCH 10/12] README: promote the regression-metamodel EVPPI and Gaussian-approximation EVSI - 'What it does' credits the methods: EVPPI from regression metamodels (Strong et al. 2014; Jalal and Alarid-Escudero 2018) and EVSI with the Gaussian approximation approach (Jalal and Alarid-Escudero 2018, doi:10.1177/0272989X17715627), both from the existing PSA sample. - New 'Value of research: EVPPI and EVSI from one PSA' section with a runnable example on cancer_psa_data (EVPI, EVPPI of two failure probabilities, EVSI by study size), with numbers computed inline and a hidden check that EVSI rises with n and EVSI < EVPPI < EVPI. - The Gaussian approximation paper is listed under Citation. --- README.Rmd | 36 +++++++++++++++++++++++++- README.md | 76 +++++++++++++++++++++++++++++++++++++++++++++++++++--- 2 files changed, 107 insertions(+), 5 deletions(-) diff --git a/README.Rmd b/README.Rmd index 258e6f0..b792e58 100644 --- a/README.Rmd +++ b/README.Rmd @@ -37,7 +37,7 @@ dampack does these analyses in R for any model. It only needs the model's output - **Cost-effectiveness analysis.** `calculate_icers()` identifies dominated and extendedly dominated strategies and computes ICERs along the efficient frontier; `plot()` draws the frontier. - **Probabilistic sensitivity analysis.** `make_psa_obj()` stores the PSA output of any model. Its `plot()` and `summary()` methods show the cost-effectiveness plane, `ceac()` computes cost-effectiveness acceptability curves (CEACs) together with the acceptability frontier (CEAF), and `calc_exp_loss()` computes expected loss curves (ELCs), which show in one figure the expected foregone benefit of each strategy, the optimal strategy, and the expected value of perfect information ([Alarid-Escudero et al., 2019](https://doi.org/10.1016/j.jval.2019.02.008)). -- **Value of information.** `calc_evpi()` computes the expected value of perfect information. `calc_evppi()` and `calc_evsi()` estimate the expected value of partial perfect information and of sample information from regression metamodels of the PSA, so no further model runs are needed. +- **Value of information.** `calc_evpi()` computes the expected value of perfect information (EVPI). `calc_evppi()` estimates the expected value of partial perfect information (EVPPI) with regression metamodels of the PSA ([Strong et al., 2014](https://doi.org/10.1177/0272989X13505910); [Jalal and Alarid-Escudero, 2018](https://doi.org/10.1177/0272989X17715627)), and `calc_evsi()` estimates the expected value of sample information (EVSI) of a future study with the Gaussian approximation approach ([Jalal and Alarid-Escudero, 2018](https://doi.org/10.1177/0272989X17715627)). Both work from the PSA sample you already have, so no further model runs are needed. - **Deterministic sensitivity analysis.** One-way and two-way analyses come either from the PSA, through metamodels (`owsa()`, `twsa()`), or from running your model directly (`run_owsa_det()`, `run_twsa_det()`), with tornado diagrams and optimal-strategy plots. - **PSA inputs.** `gen_psa_samp()` draws parameter samples from common distributions, helpers such as `beta_params()` and `gamma_params()` convert means and standard deviations into distribution parameters, and `run_psa()` runs your model over the samples. @@ -110,6 +110,36 @@ stopifnot(isTRUE(all.equal(loss_optimal, evpi_50k))) At $50,000 per QALY, `r optimal` has the lowest expected loss, $`r format(round(loss_optimal), big.mark = ",")` per patient, which is also the EVPI at that threshold. +### Value of research: EVPPI and EVSI from one PSA + +The EVPI is the most a decision maker should pay to remove all uncertainty. To decide which research is worth doing, it helps to know how much of that value comes from a few parameters, the EVPPI, and how much a study of a given size would capture, the EVSI. Both usually need nested simulations: running the model again for every possible study result. + +dampack avoids them. `calc_evppi()` fits a regression metamodel of the net benefit on the parameters of interest across the PSA samples. `calc_evsi()` uses the Gaussian approximation approach of Jalal and Alarid-Escudero ([2018](https://doi.org/10.1177/0272989X17715627)). A regression metamodel relates the net benefit to the parameters, and a Gaussian approximation of the preposterior distribution tells how much a study of `n` patients would narrow the uncertainty around them. It needs only the PSA sample, and it works for correlated parameters, non-Gaussian distributions, and several parameters informed by one study. + +Here the parameters of interest are the probabilities of treatment failure with chemotherapy and with surgery, each estimated from an earlier study of `n0 = 50` patients: + +```{r voi} +v_params <- c("pFailChemo", "pFailSurg") +calc_evpi(l_psa, wtp = 100000) +l_evppi <- calc_evppi(l_psa, wtp = 100000, params = v_params, + outcome = "nmb", k = 3, progress = FALSE) +l_evppi$df_evppi +l_evsi <- calc_evsi(l_psa, wtp = 100000, params = v_params, + outcome = "nmb", k = 3, n0 = 50, + n = c(10, 50, 100, 500, 1000), progress = FALSE) +l_evsi$df_evsi +``` + +```{r voi-check, include = FALSE} +evpi <- calc_evpi(l_psa, wtp = 100000)$EVPI +evppi <- l_evppi$df_evppi$EVPPI +v_evsi <- l_evsi$df_evsi$EVSI +stopifnot(all(diff(v_evsi) > 0), max(v_evsi) < evppi, evppi < evpi) +fmt <- function(x) format(round(x), big.mark = ",") +``` + +At $100,000 per QALY, removing all uncertainty is worth $`r fmt(evpi)` per patient, and `r round(100 * evppi / evpi)`% of that comes from the two failure probabilities. A new study of 10 patients would capture $`r fmt(v_evsi[1])` per patient, and one of 1,000 patients would capture $`r fmt(max(v_evsi))`, close to the EVPPI. Scaled to the population affected by the decision and compared with the cost of each study design, these values show whether a study is worth running, and how large it should be. + ## Learn more The [package website](https://darth-git.github.io/dampack/) has the full [function reference](https://darth-git.github.io/dampack/reference/) and these articles, best read in this order: @@ -130,6 +160,10 @@ The expected loss curves in dampack, and why they complement CEACs and the CEAF, > Alarid-Escudero F, Enns EA, Kuntz KM, Michaud TL, Jalal H. "Time Traveling Is Just Too Dangerous" but Some Methods Are Worth Revisiting: The Advantages of Expected Loss Curves Over Cost-Effectiveness Acceptability Curves and Frontier. *Value in Health*. 2019;22(5):611-618. [doi:10.1016/j.jval.2019.02.008](https://doi.org/10.1016/j.jval.2019.02.008) +The Gaussian approximation approach that `calc_evsi()` uses to compute the EVSI from a single PSA sample is described in: + +> Jalal H, Alarid-Escudero F. A Gaussian Approximation Approach for Value of Information Analysis. *Medical Decision Making*. 2018;38(2):174-188. [doi:10.1177/0272989X17715627](https://doi.org/10.1177/0272989X17715627) + ## Getting help and contributing Please report bugs and suggest features in the [issue tracker](https://github.com/DARTH-git/dampack/issues). To contribute code, see the [contributing guide](https://github.com/DARTH-git/dampack/blob/main/.github/CONTRIBUTING.md). Everyone taking part in the project agrees to follow its [code of conduct](https://github.com/DARTH-git/dampack/blob/main/CODE_OF_CONDUCT.md). diff --git a/README.md b/README.md index 2e23250..6431c58 100644 --- a/README.md +++ b/README.md @@ -50,10 +50,16 @@ scientists, health economists, and students. information ([Alarid-Escudero et al., 2019](https://doi.org/10.1016/j.jval.2019.02.008)). - **Value of information.** `calc_evpi()` computes the expected value of - perfect information. `calc_evppi()` and `calc_evsi()` estimate the - expected value of partial perfect information and of sample - information from regression metamodels of the PSA, so no further model - runs are needed. + perfect information (EVPI). `calc_evppi()` estimates the expected + value of partial perfect information (EVPPI) with regression + metamodels of the PSA ([Strong et al., + 2014](https://doi.org/10.1177/0272989X13505910); [Jalal and + Alarid-Escudero, 2018](https://doi.org/10.1177/0272989X17715627)), and + `calc_evsi()` estimates the expected value of sample information + (EVSI) of a future study with the Gaussian approximation approach + ([Jalal and Alarid-Escudero, + 2018](https://doi.org/10.1177/0272989X17715627)). Both work from the + PSA sample you already have, so no further model runs are needed. - **Deterministic sensitivity analysis.** One-way and two-way analyses come either from the PSA, through metamodels (`owsa()`, `twsa()`), or from running your model directly (`run_owsa_det()`, `run_twsa_det()`), @@ -152,6 +158,60 @@ plot(df_elc, n_x_ticks = 10) At \$50,000 per QALY, Chemo has the lowest expected loss, \$14,954 per patient, which is also the EVPI at that threshold. +### Value of research: EVPPI and EVSI from one PSA + +The EVPI is the most a decision maker should pay to remove all +uncertainty. To decide which research is worth doing, it helps to know +how much of that value comes from a few parameters, the EVPPI, and how +much a study of a given size would capture, the EVSI. Both usually need +nested simulations: running the model again for every possible study +result. + +dampack avoids them. `calc_evppi()` fits a regression metamodel of the +net benefit on the parameters of interest across the PSA samples. +`calc_evsi()` uses the Gaussian approximation approach of Jalal and +Alarid-Escudero ([2018](https://doi.org/10.1177/0272989X17715627)). A +regression metamodel relates the net benefit to the parameters, and a +Gaussian approximation of the preposterior distribution tells how much a +study of `n` patients would narrow the uncertainty around them. It needs +only the PSA sample, and it works for correlated parameters, +non-Gaussian distributions, and several parameters informed by one +study. + +Here the parameters of interest are the probabilities of treatment +failure with chemotherapy and with surgery, each estimated from an +earlier study of `n0 = 50` patients: + +``` r +v_params <- c("pFailChemo", "pFailSurg") +calc_evpi(l_psa, wtp = 100000) +#> WTP EVPI +#> 1 1e+05 22683.95 +l_evppi <- calc_evppi(l_psa, wtp = 100000, params = v_params, + outcome = "nmb", k = 3, progress = FALSE) +l_evppi$df_evppi +#> WTP EVPPI +#> 1 1e+05 8213.325 +l_evsi <- calc_evsi(l_psa, wtp = 100000, params = v_params, + outcome = "nmb", k = 3, n0 = 50, + n = c(10, 50, 100, 500, 1000), progress = FALSE) +l_evsi$df_evsi +#> WTP n EVSI +#> 1 1e+05 10 1331.363 +#> 2 1e+05 50 4492.632 +#> 3 1e+05 100 5837.748 +#> 4 1e+05 500 7599.690 +#> 5 1e+05 1000 7894.617 +``` + +At \$100,000 per QALY, removing all uncertainty is worth \$22,684 per +patient, and 36% of that comes from the two failure probabilities. A new +study of 10 patients would capture \$1,331 per patient, and one of 1,000 +patients would capture \$7,895, close to the EVPPI. Scaled to the +population affected by the decision and compared with the cost of each +study design, these values show whether a study is worth running, and +how large it should be. + ## Learn more The [package website](https://darth-git.github.io/dampack/) has the full @@ -187,6 +247,14 @@ the CEAF, are described in: > Health*. 2019;22(5):611-618. > [doi:10.1016/j.jval.2019.02.008](https://doi.org/10.1016/j.jval.2019.02.008) +The Gaussian approximation approach that `calc_evsi()` uses to compute +the EVSI from a single PSA sample is described in: + +> Jalal H, Alarid-Escudero F. A Gaussian Approximation Approach for +> Value of Information Analysis. *Medical Decision Making*. +> 2018;38(2):174-188. +> [doi:10.1177/0272989X17715627](https://doi.org/10.1177/0272989X17715627) + ## Getting help and contributing Please report bugs and suggest features in the [issue From 48e032dd2bad03033b6e16686140b50611665194 Mon Sep 17 00:00:00 2001 From: Fernando Alarid-Escudero Date: Tue, 17 Nov 2026 15:39:06 -0600 Subject: [PATCH 11/12] Keep calculate_icers() fully incremental; add calculate_cers_ref() for comparisons with a reference As agreed at the DARTH meeting of Nov 17, 2026: - Remove ref_str from calculate_icers(), print.icers(), plot.icers(), calculate_icers_psa() and summary.psa(). They keep wtp/nhb (NMB, NHB) and currency. - plot.icers(incremental = TRUE) puts the first non-dominated strategy at the origin. - New calculate_cers_ref() and calculate_cers_ref_psa(), with print() and plot() methods, compare every strategy with a reference strategy, citing the NICE, CDA-AMC and Zorginstituut Nederland guidelines. - basic_cea vignette section, README bullet, pkgdown reference group. Co-Authored-By: Claude --- NAMESPACE | 4 + NEWS.md | 10 +- R/cers_ref.R | 460 +++++++++++++++++++ R/icers.R | 291 ++++-------- R/psa.R | 44 +- README.Rmd | 2 +- README.md | 6 +- _pkgdown.yml | 11 + man/add_cers_ref_uncertainty.Rd | 27 ++ man/{add_reference_and_nmb.Rd => add_nmb.Rd} | 13 +- man/add_reference_intervals.Rd | 30 -- man/calculate_cers_ref.Rd | 114 +++++ man/calculate_cers_ref_psa.Rd | 85 ++++ man/calculate_icers.Rd | 42 +- man/calculate_icers_psa.Rd | 22 +- man/format_cea_table.Rd | 21 + man/plot.cers_ref.Rd | 114 +++++ man/plot.icers.Rd | 23 +- man/print.cers_ref.Rd | 38 ++ man/print.icers.Rd | 21 +- man/status_vs_reference.Rd | 2 +- man/summary.psa.Rd | 20 +- tests/testthat/test_cers_ref.R | 178 +++++++ tests/testthat/test_icers_reference.R | 136 ------ tests/testthat/test_psa_reference.R | 77 ---- tests/testthat/test_reference_plots.R | 70 ++- vignettes/basic_cea.Rmd | 36 ++ 27 files changed, 1283 insertions(+), 614 deletions(-) create mode 100644 R/cers_ref.R create mode 100644 man/add_cers_ref_uncertainty.Rd rename man/{add_reference_and_nmb.Rd => add_nmb.Rd} (63%) delete mode 100644 man/add_reference_intervals.Rd create mode 100644 man/calculate_cers_ref.Rd create mode 100644 man/calculate_cers_ref_psa.Rd create mode 100644 man/format_cea_table.Rd create mode 100644 man/plot.cers_ref.Rd create mode 100644 man/print.cers_ref.Rd create mode 100644 tests/testthat/test_cers_ref.R delete mode 100644 tests/testthat/test_icers_reference.R delete mode 100644 tests/testthat/test_psa_reference.R diff --git a/NAMESPACE b/NAMESPACE index 4fb5b30..1dca31c 100644 --- a/NAMESPACE +++ b/NAMESPACE @@ -1,6 +1,7 @@ # Generated by roxygen2: do not edit by hand S3method(plot,ceac) +S3method(plot,cers_ref) S3method(plot,evpi) S3method(plot,evppi) S3method(plot,evsi) @@ -10,6 +11,7 @@ S3method(plot,owsa) S3method(plot,psa) S3method(plot,twsa) S3method(predict,metamodel) +S3method(print,cers_ref) S3method(print,icers) S3method(print,metamodel) S3method(print,sa) @@ -21,6 +23,8 @@ export(calc_evpi) export(calc_evppi) export(calc_evsi) export(calc_exp_loss) +export(calculate_cers_ref) +export(calculate_cers_ref_psa) export(calculate_icers) export(calculate_icers_psa) export(ceac) diff --git a/NEWS.md b/NEWS.md index 5367521..9d802f2 100644 --- a/NEWS.md +++ b/NEWS.md @@ -5,12 +5,10 @@ - `plot.psa()` gains `xtrans` and `ytrans` to put the effectiveness and cost axes on, for example, a log scale. The ellipses are computed on the transformed scale, so they keep their shape instead of breaking where they cross zero; values a log axis cannot show give an informative error (#201). - Plot arguments are more consistent across methods (#214): `plot.owsa()` and `plot.exp_loss()` take `linewidth` (as in ggplot2); `plot.owsa()` and `plot.twsa()` gain `xlim` and `ylim`, and `owsa_opt_strat()` gains `xlim`; `effect_units` defaults to `"QALYs"` in every plot, and willingness-to-pay labels use its singular ("$/QALY"). The naming rules are recorded in an API style section of CONTRIBUTING.md. - The currency of a `psa` object now carries through to its analyses: `ceac()`, `calc_exp_loss()`, `calc_evpi()`, `calc_evppi()`, `calc_evsi()`, and `calculate_icers_psa()` keep it, and their `plot()` and `print()` methods use it unless `currency` is given. `plot.ceac()` gains `effect_units` (#203). -- `summary.psa()` gains `ref_str` and `wtp`. With a reference strategy it adds the mean incremental cost and effect and the proportion of PSA samples in each quadrant of the incremental cost-effectiveness plane (`propNE`, `propSE`, `propSW`, `propNW`); with `wtp` it adds the mean NMB and, with a reference, `meanIncNMB` and the proportion of samples that are cost-effective compared with the reference (`propCostEffective`) (#209, #202). -- `calculate_icers_psa()` gains `ref_str`, `wtp`, and `nhb`. With `uncertainty = TRUE` and a reference strategy, it adds 95% intervals for the differences from the reference, computed sample by sample (#209). -- `plot.icers()` gains `ref_str`, which plots the incremental cost-effectiveness plane with the reference strategy at the origin, and `wtp`, which draws a willingness-to-pay line. `plot.psa()` gains `wtp` for the same line, drawn through the reference strategy or, without one, through the least costly strategy. `plot.icers()` uses the currency stored by `calculate_icers()` (#209, #202). -- `calculate_icers()` can compare every strategy with a user-selected reference strategy: with `ref_str = "Standard of care"`, it adds `Inc_Cost_ref`, `Inc_Effect_ref`, `ICER_ref`, and `Status_ref` ("Reference", "Dominant", "Dominated", "More costly, more effective", "Less costly, less effective", or "Equivalent"). The frontier-based results are unchanged, and without `ref_str` no columns are added (#209). -- `calculate_icers()` gains `wtp`, which adds the net monetary benefit (`NMB`, and `Inc_NMB_ref` with a reference strategy), and `nhb = TRUE`, which also adds the net health benefit (`NHB`, `Inc_NHB_ref`) (#200). -- `calculate_icers()` gains `currency`, used by `print.icers()`. `print.icers()` gains `ref`: `"sequential"` (the default) prints the usual table, and `"reference"` or `"both"` print the comparison with the reference strategy (#209). +- New `calculate_cers_ref()` and `calculate_cers_ref_psa()` compare every strategy with a reference strategy chosen by the user, such as the standard of care: the difference in cost and effect from the reference, the cost-effectiveness ratio (CER) of that difference, and the status relative to the reference ("Dominant", "Dominated", "More costly, more effective", "Less costly, less effective", or "Equivalent"). With `wtp`, they add the net monetary benefit and its difference from the reference. From a PSA, `uncertainty = TRUE` adds 95% intervals for the differences and the proportion of samples in each quadrant of the incremental cost-effectiveness plane and, with `wtp`, the proportion that is cost-effective compared with the reference. `plot()` draws the incremental plane centered at the reference, with a segment to each strategy and an optional willingness-to-pay line. These comparisons complement, and do not replace, the fully incremental analysis of `calculate_icers()`, which HTA guidelines require; the documentation cites the NICE, CDA-AMC, and Zorginstituut Nederland guidelines (#209). +- `plot.icers()` gains `incremental = TRUE`, which plots the incremental cost-effectiveness plane with the first non-dominated strategy (the least costly strategy on the frontier) at the origin, and `wtp`, which draws a willingness-to-pay line through that strategy. `plot.psa()` gains `wtp` for the same line, drawn through the reference strategy or, without one, through the least costly strategy (#209, #202). +- `calculate_icers()` and `calculate_icers_psa()` gain `wtp`, which adds the net monetary benefit (`NMB`) of each strategy, and `nhb = TRUE`, which also adds the net health benefit (`NHB`) (#200). +- `calculate_icers()` gains `currency`, which `print.icers()` and `plot.icers()` use (#203). - `print.icers()` prints a formatted ICER table, with options for the currency symbol and the number of digits (#184, #171). - `plot.psa()` gains a `ref_str` argument that plots the PSA on the incremental cost-effectiveness plane relative to a chosen reference strategy (#188). - Plotting functions now default to an accessible, colorblind-friendly palette (`col = "access"`, based on viridis). The previous palettes remain available as `col = "full"` and `col = "bw"`. diff --git a/R/cers_ref.R b/R/cers_ref.R new file mode 100644 index 0000000..a734571 --- /dev/null +++ b/R/cers_ref.R @@ -0,0 +1,460 @@ +# nolint start: line_length_linter. (guideline URL in the references) +#' Compare every strategy with a reference strategy +#' +#' @description +#' Compares each strategy with a single reference strategy chosen by the user, such as the +#' current standard of care: the difference in cost and in effect from the reference, the +#' cost-effectiveness ratio (CER) of that difference, and where each strategy lies on the +#' incremental cost-effectiveness plane centered at the reference. +#' +#' These comparisons are \emph{not} the incremental cost-effectiveness ratios (ICERs) of a +#' fully incremental analysis. When more than two strategies are compared, health technology +#' assessment guidelines require a fully incremental (sequential) analysis, in which dominated +#' and extendedly dominated strategies are removed and each strategy on the cost-effectiveness +#' frontier is compared with the next less costly one: use \code{\link{calculate_icers}} for that. +#' Comparisons with a single reference are a complement for settings where they are relevant +#' and justified. The NICE manual (2022), for example, states that "pairwise comparisons may be +#' presented when relevant and justified (for example, when the technology is expected to +#' specifically displace individual comparators)", after requiring a fully incremental +#' analysis. The CDA-AMC (2017) and Zorginstituut Nederland (2024) guidelines also require a +#' sequential analysis when more than two interventions are compared. +#' +#' @param cost vector of cost for each strategy. +#' @param effect vector of effect for each strategy. +#' @param strategies string vector of strategy names. +#' @param ref_str name of the reference strategy, one of \code{strategies}, for example +#' \code{"Standard of care"}. +#' @param wtp a single willingness-to-pay threshold (currency per unit of effect). If given, the +#' net monetary benefit of each strategy and its difference from the reference are added. +#' The default (\code{NULL}) adds no net-benefit columns. +#' @param nhb if \code{TRUE}, also add the net health benefit and its difference from the +#' reference. Requires \code{wtp}. +#' @param currency currency symbol used when printing and plotting the results, for example +#' \code{"€"}. The default (\code{NULL}) uses \code{"$"}. +#' +#' @return A data frame of class \code{cers_ref}, with the reference strategy in the first row +#' and the other strategies by increasing cost, and these columns: +#' \describe{ +#' \item{Strategy, Cost, Effect}{as given.} +#' \item{Inc_Cost, Inc_Effect}{cost and effect minus those of the reference.} +#' \item{CER}{\code{Inc_Cost / Inc_Effect}, the cost-effectiveness ratio compared with the +#' reference. It is \code{NA} where a ratio is not meaningful: for the reference itself and for +#' dominant, dominated and equivalent strategies.} +#' \item{Status}{one of \code{"Reference"}, \code{"Dominant"} (not more costly and at least as +#' effective), \code{"Dominated"} (not less costly and no more effective), +#' \code{"More costly, more effective"}, \code{"Less costly, less effective"}, or +#' \code{"Equivalent"} (same cost and effect).} +#' \item{NMB, Inc_NMB}{with \code{wtp}: the net monetary benefit, \code{wtp * Effect - Cost}, +#' and its difference from the reference. A strategy with a positive \code{Inc_NMB} is +#' cost-effective compared with the reference at this threshold.} +#' \item{NHB, Inc_NHB}{with \code{nhb = TRUE}: the net health benefit, +#' \code{Effect - Cost / wtp}, and its difference from the reference.} +#' } +#' The reference strategy, threshold and currency are stored as attributes, which the +#' \code{print} and \code{plot} methods use. +#' +#' For a strategy that is more costly and more effective than the reference, a \code{CER} below +#' the willingness-to-pay threshold means it is cost-effective compared with the reference. For a +#' strategy that is less costly and less effective, the \code{CER} is the cost saved per unit of +#' effect lost, and the strategy is cost-effective compared with the reference when its +#' \code{CER} is \emph{above} the threshold. +#' +#' @references +#' National Institute for Health and Care Excellence (NICE). NICE health technology evaluations: +#' the manual (PMG36), section 4.6, "Economic evaluation". London: NICE; 2022. +#' \url{https://www.nice.org.uk/process/pmg36/chapter/economic-evaluation-2} +#' +#' CDA-AMC (formerly CADTH). Guidelines for the Economic Evaluation of Health Technologies: +#' Canada. 4th ed. Ottawa: CADTH; 2017. +#' \url{https://www.cda-amc.ca/sites/default/files/pdf/guidelines_for_the_economic_evaluation_of_health_technologies_canada_4th_ed.pdf} +#' +#' Zorginstituut Nederland. Guideline for economic evaluations in healthcare (2024 version). +#' Diemen: Zorginstituut Nederland; 2024. +#' \url{https://english.zorginstituutnederland.nl/documents/2024/01/16/guideline-for-economic-evaluations-in-healthcare} +#' +#' @seealso \code{\link{plot.cers_ref}}, \code{\link{calculate_cers_ref_psa}} for the same +#' comparison from a PSA, and \code{\link{calculate_icers}} for the fully incremental analysis. +#' +#' @examples +#' # three strategies; "Standard of care" is neither the cheapest nor the least effective +#' df_ref <- calculate_cers_ref(cost = c(12000, 9000, 20000), +#' effect = c(8.0, 7.6, 8.5), +#' strategies = c("Standard of care", "Generic drug", "New drug"), +#' ref_str = "Standard of care", +#' wtp = 50000) +#' df_ref +#' plot(df_ref) +#' +#' # the fully incremental analysis of the same strategies +#' calculate_icers(cost = c(12000, 9000, 20000), +#' effect = c(8.0, 7.6, 8.5), +#' strategies = c("Standard of care", "Generic drug", "New drug")) +#' @export +calculate_cers_ref <- function(cost, effect, strategies, ref_str, wtp = NULL, nhb = FALSE, + currency = NULL) { + if (length(cost) != length(effect) || length(effect) != length(strategies)) { + stop("cost, effect, and strategies must all be vectors of the same length", call. = FALSE) + } + if (length(strategies) == 0) { + stop("cost, effect, and strategies must contain at least one strategy", call. = FALSE) + } + if (missing(ref_str) || is.null(ref_str)) { + stop("`ref_str` must name the reference strategy.", call. = FALSE) + } + v_strat <- as.character(strategies) + if (anyDuplicated(v_strat)) { + stop("Strategy names must be unique.", call. = FALSE) + } + ref_str <- check_ref_str(ref_str, v_strat) + check_wtp_nhb(wtp, nhb) + + # reference first, then the other strategies by increasing cost (and decreasing effect) + v_order <- order(v_strat != ref_str, cost, -effect) + df_ref <- data.frame(Strategy = v_strat[v_order], Cost = cost[v_order], + Effect = effect[v_order], stringsAsFactors = FALSE) + is_ref <- df_ref$Strategy == ref_str + + # differences from the reference; its own are reported as NA + df_ref$Inc_Cost <- df_ref$Cost - df_ref$Cost[is_ref] + df_ref$Inc_Effect <- df_ref$Effect - df_ref$Effect[is_ref] + v_status <- status_vs_reference(df_ref$Inc_Cost, df_ref$Inc_Effect) + v_status[is_ref] <- "Reference" + df_ref$Inc_Cost[is_ref] <- NA_real_ + df_ref$Inc_Effect[is_ref] <- NA_real_ + # a ratio is only meaningful when cost and effect move in the same direction (NE and SW + # quadrants); for dominant, dominated, and equivalent strategies the CER stays NA + v_has_ratio <- v_status %in% c("More costly, more effective", "Less costly, less effective") + df_ref$CER <- ifelse(v_has_ratio, df_ref$Inc_Cost / df_ref$Inc_Effect, NA_real_) + df_ref$Status <- v_status + + if (!is.null(wtp)) { + df_ref$NMB <- wtp * df_ref$Effect - df_ref$Cost + df_ref$Inc_NMB <- df_ref$NMB - df_ref$NMB[is_ref] + df_ref$Inc_NMB[is_ref] <- NA_real_ + if (nhb) { + df_ref$NHB <- df_ref$Effect - df_ref$Cost / wtp + df_ref$Inc_NHB <- df_ref$NHB - df_ref$NHB[is_ref] + df_ref$Inc_NHB[is_ref] <- NA_real_ + } + } + + structure(df_ref, class = c("cers_ref", "data.frame"), ref_str = ref_str, wtp = wtp, + currency = currency) +} + +# nolint end + +#' Compare every strategy with a reference strategy, from a \code{psa} object +#' +#' @description +#' Compares each strategy with a reference strategy using the mean costs and effects of a +#' probabilistic sensitivity analysis (PSA), as \code{\link{calculate_cers_ref}} does for a base +#' case. With \code{uncertainty = TRUE}, it also summarizes the uncertainty in each comparison +#' across the PSA samples. See \code{\link{calculate_cers_ref}} for when a comparison with a +#' single reference is appropriate, and \code{\link{calculate_icers_psa}} for the fully +#' incremental analysis. +#' +#' @param psa \code{psa} object from \code{\link{make_psa_obj}}. +#' @param uncertainty if \code{TRUE}, add the 95\% intervals and quadrant proportions described +#' in Value. Defaults to \code{FALSE}. +#' @inheritParams calculate_cers_ref +#' +#' @return A \code{cers_ref} object with the columns described in +#' \code{\link{calculate_cers_ref}}, computed from the mean cost and effect of each strategy. The +#' currency of \code{psa} is kept. +#' +#' With \code{uncertainty = TRUE}, the differences from the reference are computed in every PSA +#' sample, and these columns are added (\code{NA} for the reference itself): +#' \itemize{ +#' \item \code{Lower_95_Inc_Cost} and \code{Upper_95_Inc_Cost}, the 2.5th and 97.5th +#' percentiles of the difference in cost, placed after \code{Inc_Cost}; likewise for +#' \code{Inc_Effect} and, with \code{wtp}, \code{Inc_NMB} (and \code{Inc_NHB}). +#' \item \code{Prop_NE}, \code{Prop_SE}, \code{Prop_SW}, and \code{Prop_NW}, the proportion of +#' samples in each quadrant of the incremental cost-effectiveness plane: NE more effective and +#' more costly; SE more effective and not more costly; SW not more effective and not more +#' costly; NW not more effective and more costly. The four proportions sum to 1. +#' \item With \code{wtp}, \code{Prop_CE}, the proportion of samples in which the strategy is +#' cost-effective compared with the reference (it has a higher net monetary benefit). +#' } +#' +#' @inherit calculate_cers_ref references +#' @seealso \code{\link{calculate_cers_ref}}, \code{\link{plot.cers_ref}}, and +#' \code{\link{plot.psa}} with \code{ref_str} for the PSA samples on the incremental plane. +#' +#' @examples +#' data(cancer_psa_data) +#' l_psa <- make_psa_obj(cost = cancer_psa_data$cost, +#' effectiveness = cancer_psa_data$effectiveness, +#' parameters = cancer_psa_data$parameters, +#' strategies = cancer_psa_data$strategies) +#' df_ref <- calculate_cers_ref_psa(l_psa, ref_str = "Radio", wtp = 50000, +#' uncertainty = TRUE) +#' df_ref +#' plot(df_ref) +#' @export +calculate_cers_ref_psa <- function(psa, ref_str, wtp = NULL, nhb = FALSE, uncertainty = FALSE) { + check_psa_object(psa) + if (!isTRUE(uncertainty) && !isFALSE(uncertainty)) { + stop("`uncertainty` must be TRUE or FALSE.", call. = FALSE) + } + df_sum <- summary(psa) + df_ref <- calculate_cers_ref(cost = df_sum$meanCost, effect = df_sum$meanEffect, + strategies = df_sum$Strategy, ref_str = ref_str, wtp = wtp, + nhb = nhb, currency = psa$currency) + if (uncertainty) { + df_ref <- add_cers_ref_uncertainty(df_ref, psa, wtp, nhb) + } + df_ref +} + +#' Add PSA intervals and quadrant proportions to a \code{cers_ref} object +#' +#' @param df_ref a \code{cers_ref} object computed from the means of \code{psa}. +#' @param psa the \code{psa} object. +#' @inheritParams calculate_cers_ref +#' @return \code{df_ref} with the columns described in \code{\link{calculate_cers_ref_psa}}. +#' @keywords internal +add_cers_ref_uncertainty <- function(df_ref, psa, wtp, nhb) { + ref_str <- attr(df_ref, "ref_str") + v_strat <- df_ref$Strategy + is_ref <- v_strat == ref_str + # sample-by-sample differences from the reference: one row per PSA sample, one column per + # strategy in the row order of df_ref (subtracting a vector recycles it down each column) + m_inc_cost <- as.matrix(psa$cost)[, v_strat, drop = FALSE] - psa$cost[[ref_str]] + m_inc_effect <- as.matrix(psa$effectiveness)[, v_strat, drop = FALSE] - + psa$effectiveness[[ref_str]] + l_inc <- list(Inc_Cost = m_inc_cost, Inc_Effect = m_inc_effect) + if (!is.null(wtp)) { + m_inc_nmb <- wtp * m_inc_effect - m_inc_cost + l_inc$Inc_NMB <- m_inc_nmb + if (nhb) { + l_inc$Inc_NHB <- m_inc_effect - m_inc_cost / wtp + } + } + na_ref <- function(v) { + v <- unname(v) + v[is_ref] <- NA_real_ + v + } + + # 95% intervals, each placed right after its point estimate + v_cols <- names(df_ref) + for (col in names(l_inc)) { + df_ref[[paste0("Lower_95_", col)]] <- na_ref(apply(l_inc[[col]], 2, quantile, probs = 0.025, + names = FALSE)) + df_ref[[paste0("Upper_95_", col)]] <- na_ref(apply(l_inc[[col]], 2, quantile, probs = 0.975, + names = FALSE)) + v_cols <- append(v_cols, paste0(c("Lower_95_", "Upper_95_"), col), after = match(col, v_cols)) + } + + # share of samples in each quadrant of the incremental plane, and cost-effective ones + df_ref$Prop_NE <- na_ref(colMeans(m_inc_effect > 0 & m_inc_cost > 0)) + df_ref$Prop_SE <- na_ref(colMeans(m_inc_effect > 0 & m_inc_cost <= 0)) + df_ref$Prop_SW <- na_ref(colMeans(m_inc_effect <= 0 & m_inc_cost <= 0)) + df_ref$Prop_NW <- na_ref(colMeans(m_inc_effect <= 0 & m_inc_cost > 0)) + v_cols <- c(v_cols, "Prop_NE", "Prop_SE", "Prop_SW", "Prop_NW") + if (!is.null(wtp)) { + df_ref$Prop_CE <- na_ref(colMeans(m_inc_nmb > 0)) + v_cols <- c(v_cols, "Prop_CE") + } + + l_attr <- attributes(df_ref)[c("class", "ref_str", "wtp", "currency")] + df_ref <- as.data.frame(unclass(df_ref)[v_cols], check.names = FALSE, stringsAsFactors = FALSE) + for (a in names(l_attr)) { + attr(df_ref, a) <- l_attr[[a]] + } + df_ref +} + +#' Status of each strategy relative to a reference strategy +#' +#' @param v_inc_cost,v_inc_effect cost and effect of each strategy minus those of the reference. +#' @return character vector of statuses (the reference itself is labeled by the caller). +#' @keywords internal +status_vs_reference <- function(v_inc_cost, v_inc_effect) { + ifelse(v_inc_cost == 0 & v_inc_effect == 0, "Equivalent", + ifelse(v_inc_cost <= 0 & v_inc_effect >= 0, "Dominant", + ifelse(v_inc_cost >= 0 & v_inc_effect <= 0, "Dominated", + ifelse(v_inc_cost > 0, "More costly, more effective", + "Less costly, less effective")))) +} + +#' Print a comparison with a reference strategy +#' +#' Prints the table of a \code{\link{calculate_cers_ref}} object with formatted numbers and +#' headers that name the reference strategy. +#' +#' @param x a \code{cers_ref} object. +#' @param currency currency symbol. Defaults to the currency stored in \code{x}, or \code{"$"}. +#' @param dig_cost number of digits to round costs and net monetary benefits to. +#' @param dig_eff number of digits to round effects and net health benefits to. +#' @param dig_cer number of digits to round the cost-effectiveness ratios to. +#' @param dig_prop number of digits to round the proportions of PSA samples to. +#' @param ... further arguments (not used). +#' @return the formatted table, invisibly. +#' @export +print.cers_ref <- function(x, currency = NULL, dig_cost = 0, dig_eff = 2, dig_cer = 0, + dig_prop = 3, ...) { + currency <- resolve_currency(currency, x) + vs <- paste0(" vs ", attr(x, "ref_str")) + cat("Comparison of each strategy with the reference strategy, ", attr(x, "ref_str"), + "\n(not a fully incremental analysis; see calculate_icers())\n\n", sep = "") + df_table <- as.data.frame(unclass(x), check.names = FALSE, stringsAsFactors = FALSE) + l_format <- list( + Cost = list(dig_cost, paste0("Costs (", currency, ")")), + Effect = list(dig_eff, "QALYs"), + Inc_Cost = list(dig_cost, paste0("Incremental Costs", vs, " (", currency, ")")), + Inc_Effect = list(dig_eff, paste0("Incremental QALYs", vs)), + CER = list(dig_cer, paste0("CER", vs, " (", currency, "/QALY)")), + Status = list(NA, paste0("Status", vs)), + NMB = list(dig_cost, paste0("NMB (", currency, ")")), + Inc_NMB = list(dig_cost, paste0("Incremental NMB", vs, " (", currency, ")")), + NHB = list(dig_eff, "NHB (QALYs)"), + Inc_NHB = list(dig_eff, paste0("Incremental NHB", vs, " (QALYs)")) + ) + # interval columns use the digits of their point estimate; proportions their own + for (col in grep("^(Lower|Upper)_95_", names(df_table), value = TRUE)) { + base <- sub("^(Lower|Upper)_95_", "", col) + l_format[[col]] <- list(l_format[[base]][[1]], col) + } + # proportions with a fixed number of decimals, so they line up + for (col in grep("^Prop_", names(df_table), value = TRUE)) { + v <- df_table[[col]] + df_table[[col]] <- ifelse(is.na(v), "-", formatC(v, format = "f", digits = dig_prop)) + } + df_table <- format_cea_table(df_table, l_format) + print(df_table) + invisible(df_table) +} + +#' Plot a comparison with a reference strategy +#' +#' Draws the incremental cost-effectiveness plane centered at the reference strategy: each +#' strategy is placed at its difference in effect (x) and cost (y) from the reference, which sits +#' at the origin. Optional segments join the reference to each strategy, forming a star, and an +#' optional willingness-to-pay line goes through the origin: strategies below the line are +#' cost-effective compared with the reference at that threshold. +#' +#' @param x a \code{cers_ref} object from \code{\link{calculate_cers_ref}} or +#' \code{\link{calculate_cers_ref_psa}}. +#' @param wtp a willingness-to-pay threshold for the dotted line through the origin. Defaults to +#' the threshold stored in \code{x}, if any; \code{NULL} draws no line. +#' @param lines whether to draw a segment from the reference to each strategy. The default +#' (\code{NULL}) draws them when there are at most 20 strategies besides the reference, since +#' with many strategies they make the plot hard to read. +#' @param label whether to label all strategies (the default) or none. +#' @param label_max_char maximum number of characters of the strategy labels; longer names are +#' truncated. The default (\code{NULL}) does not truncate. +#' @param ref_axis draw dashed axes through the origin. Defaults to \code{TRUE}. +#' @param ref_caption add a caption naming the reference strategy and the WTP threshold. +#' Defaults to \code{TRUE}. +#' @param currency currency symbol. Defaults to the currency stored in \code{x}, or \code{"$"}. +#' @param effect_units units of effectiveness. Defaults to \code{"QALYs"}. +#' @param col color palette for the status of each strategy: \code{"access"} (the default, an +#' accessible palette), \code{"full"}, or \code{"bw"}. +#' @param alpha opacity of the points. +#' @inheritParams add_common_aes +#' @inheritParams ggrepel::geom_label_repel +#' @param ... further arguments (not used). +#' +#' @return a \code{ggplot2} object, which can be modified by adding further layers. +#' @seealso \code{\link{calculate_cers_ref}} +#' @examples +#' df_ref <- calculate_cers_ref(cost = c(12000, 9000, 20000), +#' effect = c(8.0, 7.6, 8.5), +#' strategies = c("Standard of care", "Generic drug", "New drug"), +#' ref_str = "Standard of care") +#' plot(df_ref, wtp = 50000) +#' +#' # many strategies: no segments, no labels +#' data(hund_strat) +#' df_hund <- calculate_cers_ref(hund_strat$Cost, hund_strat$QALYs, hund_strat$Strategy, +#' ref_str = "1") +#' plot(df_hund, label = "none") +#' @importFrom ggrepel geom_label_repel +#' @export +plot.cers_ref <- function(x, + wtp = attr(x, "wtp"), + lines = NULL, + label = c("all", "none"), + label_max_char = NULL, + ref_axis = TRUE, + ref_caption = TRUE, + currency = NULL, + effect_units = "QALYs", + col = c("access", "full", "bw"), + alpha = 1, + txtsize = 12, + n_x_ticks = 6, + n_y_ticks = 6, + xbreaks = NULL, + ybreaks = NULL, + xlim = NULL, + ylim = NULL, + xexpand = expansion(0.1), + yexpand = expansion(0.1), + max.iter = 20000, + ...) { + Inc_Effect <- Inc_Cost <- Status <- Strategy <- NULL + label <- match.arg(label) + col <- match.arg(col) + currency <- resolve_currency(currency, x) + check_wtp_nhb(wtp, nhb = FALSE) + ref_str <- attr(x, "ref_str") + is_ref <- x$Strategy == ref_str + n_comp <- sum(!is_ref) + if (is.null(lines)) { + lines <- n_comp <= 20 + } + + # the reference sits at the origin of the incremental plane + df_plot <- data.frame(Strategy = x$Strategy, Status = x$Status, + Inc_Cost = ifelse(is_ref, 0, x$Inc_Cost), + Inc_Effect = ifelse(is_ref, 0, x$Inc_Effect), + stringsAsFactors = FALSE) + if (!is.null(label_max_char)) { + df_plot$Strategy <- substr(df_plot$Strategy, 1, label_max_char) + } + v_levels <- c("Dominant", "More costly, more effective", "Less costly, less effective", + "Dominated", "Equivalent") + df_comp <- df_plot[!is_ref, , drop = FALSE] + # all levels, so each status keeps its color whichever statuses are present + df_comp$Status <- factor(df_comp$Status, levels = v_levels) + df_ref <- df_plot[is_ref, , drop = FALSE] + + p <- ggplot(df_comp, aes(x = Inc_Effect, y = Inc_Cost)) + if (ref_axis) { + p <- p + + geom_hline(yintercept = 0, linetype = "dashed", color = "grey50") + + geom_vline(xintercept = 0, linetype = "dashed", color = "grey50") + } + if (lines && n_comp > 0) { + p <- p + geom_segment(aes(x = 0, y = 0, xend = Inc_Effect, yend = Inc_Cost), + color = "grey60", linewidth = 0.4) + } + if (!is.null(wtp)) { + p <- p + geom_abline(intercept = 0, slope = wtp, linetype = "dotted", color = "grey30") + } + p <- p + + geom_point(aes(color = Status), size = 2.5, alpha = alpha) + + geom_point(data = df_ref, shape = 23, size = 3.5, fill = "black", color = "black") + + labs(x = paste0("Incremental Effect (", effect_units, ")"), + y = paste0("Incremental Cost (", currency, ")")) + v_caption <- c(paste0("Reference Strategy (diamond): ", ref_str), + if (!is.null(wtp)) wtp_caption(wtp, currency, effect_units)) + if (ref_caption) { + p <- p + labs(caption = paste(v_caption, collapse = "\n")) + } + p <- add_common_aes(p, txtsize, col = col, col_aes = "color", scale_name = NULL, + continuous = c("x", "y"), + n_x_ticks = n_x_ticks, n_y_ticks = n_y_ticks, + xbreaks = xbreaks, ybreaks = ybreaks, + xlim = xlim, ylim = ylim, + xexpand = xexpand, yexpand = yexpand, + v_str = v_levels) + if (label == "all") { + p <- p + geom_label_repel(data = df_plot, aes(label = Strategy), size = 3, + show.legend = FALSE, max.iter = max.iter, direction = "both") + } + p +} diff --git a/R/icers.R b/R/icers.R index 449bbe5..36392c5 100644 --- a/R/icers.R +++ b/R/icers.R @@ -14,10 +14,6 @@ #' @param cost vector of cost for each strategy #' @param effect vector of effect for each strategy #' @param strategies string vector of strategy names -#' @param ref_str name of a reference strategy (one of \code{strategies}) to compare every -#' strategy against, for example \code{"Standard of care"}. The default (\code{NULL}) adds no -#' reference columns. The frontier-based results (\code{Status}, \code{Inc_Cost}, -#' \code{Inc_Effect}, \code{ICER}) never depend on the reference. #' @param wtp a single willingness-to-pay threshold (currency per unit of effect). If given, #' the net monetary benefit (\code{NMB}) of each strategy is added. The default (\code{NULL}) #' adds no net-benefit columns. @@ -29,20 +25,16 @@ #' status, incremental cost, incremental effect, and ICER, computed sequentially along the #' cost-effectiveness frontier. #' -#' If \code{ref_str} is given, four columns compare each strategy with the reference: -#' \code{Inc_Cost_ref} and \code{Inc_Effect_ref} (cost and effect minus those of the -#' reference), \code{ICER_ref}, and \code{Status_ref}, one of \code{"Reference"}, -#' \code{"Dominant"} (less costly and at least as effective), \code{"Dominated"} (more costly -#' and no more effective), \code{"More costly, more effective"}, -#' \code{"Less costly, less effective"}, or \code{"Equivalent"} (same cost and effect). -#' \code{ICER_ref} is \code{NA} where a ratio is not meaningful: for the reference itself and -#' for dominant, dominated, and equivalent strategies. +#' If \code{wtp} is given, \code{NMB} (\code{wtp * Effect - Cost}) is added. With +#' \code{nhb = TRUE}, \code{NHB} (\code{Effect - Cost / wtp}) is added too. #' -#' If \code{wtp} is given, \code{NMB} (\code{wtp * Effect - Cost}) is added, and with -#' \code{ref_str} also \code{Inc_NMB_ref}, the NMB minus that of the reference. With -#' \code{nhb = TRUE}, \code{NHB} (\code{Effect - Cost / wtp}) and \code{Inc_NHB_ref} are added too. +#' The ICERs are always sequential: each strategy on the frontier is compared with the next less +#' costly one, after removing dominated and extendedly dominated strategies. This is the fully +#' incremental analysis that health technology assessment guidelines require when more than two +#' strategies are compared. To compare every strategy with a single reference strategy instead, +#' use \code{\link{calculate_cers_ref}}. #' -#' @seealso \code{\link{plot.icers}} +#' @seealso \code{\link{plot.icers}}, \code{\link{calculate_cers_ref}} #' #' @examples #' ## Base Case @@ -77,15 +69,15 @@ #' # note that longer strategy names will get truncated #' plot(icers, label = "all") #' -#' ## Comparing every strategy with a reference strategy -#' hund_ref <- calculate_icers(hund_strat$Cost, -#' hund_strat$QALYs, -#' hund_strat$Strategy, -#' ref_str = hund_strat$Strategy[1], -#' wtp = 50000) -#' print(hund_ref, ref = "reference") +#' # the incremental cost-effectiveness plane, with the least costly strategy on the +#' # frontier at the origin +#' plot(icers, incremental = TRUE) +#' +#' ## Net monetary benefit at a willingness-to-pay threshold +#' calculate_icers(sum_cdiff$meanCost, sum_cdiff$meanEffect, sum_cdiff$Strategy, +#' wtp = 50000) #' @export -calculate_icers <- function(cost, effect, strategies, ref_str = NULL, wtp = NULL, +calculate_icers <- function(cost, effect, strategies, wtp = NULL, nhb = FALSE, currency = NULL) { # checks on input n_cost <- length(cost) @@ -97,7 +89,6 @@ calculate_icers <- function(cost, effect, strategies, ref_str = NULL, wtp = NULL if (n_strat == 0) { stop("cost, effect, and strategies must contain at least one strategy", call. = FALSE) } - ref_str <- check_ref_str(ref_str, as.character(strategies)) check_wtp_nhb(wtp, nhb) # column names used unquoted in the dplyr code below; declaring them avoids the @@ -122,7 +113,7 @@ calculate_icers <- function(cost, effect, strategies, ref_str = NULL, wtp = NULL df_cea$ICER <- NA_real_ df_cea$Status <- "ND" class(df_cea) <- c("icers", "data.frame") - return(add_reference_and_nmb(df_cea, ref_str, wtp, nhb, currency)) + return(add_nmb(df_cea, wtp, nhb, currency)) } # three statuses: dominated, extended dominated, and non-dominated @@ -211,7 +202,7 @@ calculate_icers <- function(cost, effect, strategies, ref_str = NULL, wtp = NULL # declare class of results class(df_results) <- c("icers", "data.frame") - add_reference_and_nmb(df_results, ref_str, wtp, nhb, currency) + add_nmb(df_results, wtp, nhb, currency) } #' Caption text for a willingness-to-pay line @@ -266,20 +257,7 @@ check_wtp_nhb <- function(wtp, nhb) { invisible(NULL) } -#' Status of each strategy relative to a reference strategy -#' -#' @param v_inc_cost,v_inc_effect cost and effect of each strategy minus those of the reference. -#' @return character vector of statuses (the reference itself is labeled by the caller). -#' @keywords internal -status_vs_reference <- function(v_inc_cost, v_inc_effect) { - ifelse(v_inc_cost == 0 & v_inc_effect == 0, "Equivalent", - ifelse(v_inc_cost <= 0 & v_inc_effect >= 0, "Dominant", - ifelse(v_inc_cost >= 0 & v_inc_effect <= 0, "Dominated", - ifelse(v_inc_cost > 0, "More costly, more effective", - "Less costly, less effective")))) -} - -#' Add reference-strategy and net-benefit columns to an icers object +#' Add net-benefit columns and stored attributes to an icers object #' #' The frontier-based columns are not modified. Columns and attributes are only added #' for the arguments that are given, so with all of them \code{NULL} the object is unchanged. @@ -288,44 +266,13 @@ status_vs_reference <- function(v_inc_cost, v_inc_effect) { #' @inheritParams calculate_icers #' @return \code{df_icers} with the requested columns and attributes added. #' @keywords internal -add_reference_and_nmb <- function(df_icers, ref_str, wtp, nhb, currency) { - if (!is.null(ref_str)) { - # row of the reference strategy; check_ref_str() has already confirmed it exists - i_ref <- match(ref_str, df_icers$Strategy) - # differences of every strategy from the reference (the reference's own are 0) - v_inc_cost <- df_icers$Cost - df_icers$Cost[i_ref] - v_inc_effect <- df_icers$Effect - df_icers$Effect[i_ref] - # classify each strategy by its quadrant of the incremental CE plane - v_status <- status_vs_reference(v_inc_cost, v_inc_effect) - # the reference has no comparison with itself: label it and blank its differences - v_status[i_ref] <- "Reference" - v_inc_cost[i_ref] <- NA_real_ - v_inc_effect[i_ref] <- NA_real_ - # a ratio is only meaningful when cost and effect move in the same direction (NE and SW - # quadrants); for dominant, dominated, and equivalent strategies ICER_ref stays NA - v_has_ratio <- v_status %in% c("More costly, more effective", "Less costly, less effective") - df_icers$Inc_Cost_ref <- v_inc_cost - df_icers$Inc_Effect_ref <- v_inc_effect - df_icers$ICER_ref <- ifelse(v_has_ratio, v_inc_cost / v_inc_effect, NA_real_) - df_icers$Status_ref <- v_status - # print.icers() and plot.icers() read the reference from this attribute - attr(df_icers, "ref_str") <- ref_str - } +add_nmb <- function(df_icers, wtp, nhb, currency) { if (!is.null(wtp)) { # net monetary benefit of each strategy at the given threshold df_icers$NMB <- wtp * df_icers$Effect - df_icers$Cost - if (!is.null(ref_str)) { - # incremental NMB versus the reference; positive means cost-effective versus the reference - df_icers$Inc_NMB_ref <- df_icers$NMB - df_icers$NMB[i_ref] - df_icers$Inc_NMB_ref[i_ref] <- NA_real_ - } if (nhb) { # net health benefit: the same comparison expressed in units of effect df_icers$NHB <- df_icers$Effect - df_icers$Cost / wtp - if (!is.null(ref_str)) { - df_icers$Inc_NHB_ref <- df_icers$NHB - df_icers$NHB[i_ref] - df_icers$Inc_NHB_ref[i_ref] <- NA_real_ - } } attr(df_icers, "wtp") <- wtp } @@ -361,16 +308,14 @@ add_reference_and_nmb <- function(df_icers, ref_str, wtp, nhb, currency) { #' set to \code{TRUE}, four additional columns are provided for the 2.5th and 97.5th quantiles for #' each strategy's cost and effect. #' -#' With \code{ref_str}, \code{wtp}, or \code{nhb}, the reference-strategy and net-benefit columns -#' described in \code{\link{calculate_icers}} are added. With \code{uncertainty = TRUE} and -#' \code{ref_str}, the 2.5th and 97.5th quantiles of the sample-by-sample differences from the reference -#' are also given for \code{Inc_Cost_ref}, \code{Inc_Effect_ref}, and (with \code{wtp}) -#' \code{Inc_NMB_ref}, in columns such as \code{Lower_95_Inc_Cost_ref} and \code{Upper_95_Inc_Cost_ref}. +#' With \code{wtp} or \code{nhb}, the net-benefit columns described in +#' \code{\link{calculate_icers}} are added. #' @seealso \code{\link{plot.icers}} #' @seealso \code{\link{calculate_icers}} +#' @seealso \code{\link{calculate_cers_ref_psa}} to compare every strategy with a reference strategy #' @importFrom tidyr pivot_longer #' @export -calculate_icers_psa <- function(psa, uncertainty = FALSE, ref_str = NULL, wtp = NULL, nhb = FALSE) { +calculate_icers_psa <- function(psa, uncertainty = FALSE, wtp = NULL, nhb = FALSE) { # check that psa has class 'psa' check_psa_object(psa) @@ -387,12 +332,12 @@ calculate_icers_psa <- function(psa, uncertainty = FALSE, ref_str = NULL, wtp = df_icers <- calculate_icers(cost = df_psa_sum$meanCost, effect = df_psa_sum$meanEffect, strategies = df_psa_sum$Strategy, - ref_str = ref_str, wtp = wtp, nhb = nhb, + wtp = wtp, nhb = nhb, currency = psa$currency) if (uncertainty == TRUE) { # the joins and column selection below drop the icers class and attributes; keep them - l_attr <- attributes(df_icers)[c("class", "ref_str", "wtp", "currency")] + l_attr <- attributes(df_icers)[c("class", "wtp", "currency")] l_attr <- l_attr[!vapply(l_attr, is.null, logical(1))] # extract cost and effect data.frames from psa object @@ -417,14 +362,11 @@ calculate_icers_psa <- function(psa, uncertainty = FALSE, ref_str = NULL, wtp = left_join(df_cost_bounds, by = "Strategy") %>% left_join(df_effect_bounds, by = "Strategy") %>% # put each interval next to its estimate; everything() keeps the remaining columns - # (the reference and net-benefit columns), which select() would otherwise drop + # (the net-benefit columns), which select() would otherwise drop select(Strategy, Cost, Lower_95_Cost, Upper_95_Cost, Effect, Lower_95_Effect, Upper_95_Effect, Inc_Cost, Inc_Effect, ICER, Status, everything()) - if (!is.null(ref_str)) { - df_icers <- add_reference_intervals(df_icers, psa, ref_str, wtp) - } df_icers <- as.data.frame(df_icers) for (a in names(l_attr)) { attr(df_icers, a) <- l_attr[[a]] @@ -434,45 +376,6 @@ calculate_icers_psa <- function(psa, uncertainty = FALSE, ref_str = NULL, wtp = df_icers } -#' Add 95\% intervals for the differences from a reference strategy -#' -#' Quantiles of the sample-by-sample differences between each strategy and the reference, placed -#' after the corresponding point estimate. \code{NA} for the reference itself. -#' -#' @param df_icers an \code{icers} object with reference columns. -#' @param psa the \code{psa} object it was computed from. -#' @inheritParams calculate_icers -#' @return \code{df_icers} with the interval columns added. -#' @keywords internal -add_reference_intervals <- function(df_icers, psa, ref_str, wtp) { - # sample-by-sample differences from the reference: one row per PSA sample, one column per - # strategy (subtracting a vector from a matrix recycles it down each column) - m_inc_cost <- as.matrix(psa$cost) - psa$cost[[ref_str]] - m_inc_effect <- as.matrix(psa$effectiveness) - psa$effectiveness[[ref_str]] - # one matrix of differences per point-estimate column that gets an interval - l_inc <- list(Inc_Cost_ref = m_inc_cost, Inc_Effect_ref = m_inc_effect) - if (!is.null(wtp)) { - l_inc$Inc_NMB_ref <- wtp * m_inc_effect - m_inc_cost - } - v_strat <- df_icers$Strategy - # running column order, so each interval lands right after its point estimate - v_cols <- names(df_icers) - for (col in names(l_inc)) { - # reorder the matrix columns to match the rows of df_icers (frontier order) - m_inc <- l_inc[[col]][, v_strat, drop = FALSE] - v_lower <- unname(apply(m_inc, 2, quantile, probs = 0.025, names = FALSE)) - v_upper <- unname(apply(m_inc, 2, quantile, probs = 0.975, names = FALSE)) - # the reference's differences are all 0, so its interval is reported as NA - v_lower[v_strat == ref_str] <- NA_real_ - v_upper[v_strat == ref_str] <- NA_real_ - df_icers[[paste0("Lower_95_", col)]] <- v_lower - df_icers[[paste0("Upper_95_", col)]] <- v_upper - i <- match(col, v_cols) - v_cols <- append(v_cols, paste0(c("Lower_95_", "Upper_95_"), col), after = i) - } - df_icers[, v_cols] -} - #' compute icers for non-dominated strategies #' #' @param non_d a data frame of non-dominated strategies, with columns @@ -512,16 +415,19 @@ compute_icers <- function(non_d) { #' longer strategies are truncated to save space. #' @param plot_frontier_only only plot the efficient frontier #' @param alpha opacity of points -#' @param ref_str name of a reference strategy. If given, the plot shows the incremental -#' cost-effectiveness plane, with the reference strategy at the origin. The frontier and the -#' status of each strategy do not change. Defaults to \code{NULL} (the cost-effectiveness plane). +#' @param incremental if \code{TRUE}, plot the incremental cost-effectiveness plane: the costs +#' and effects of every strategy are shown relative to those of the first non-dominated strategy +#' (the least costly strategy on the frontier), which is placed at the origin (0, 0). The frontier +#' and the status of each strategy do not change. Defaults to \code{FALSE} (absolute costs and +#' effects). To compare every strategy with a reference strategy of your choice, use +#' \code{\link{calculate_cers_ref}} and its \code{plot} method. #' @param wtp a willingness-to-pay threshold. If given, a dotted line with slope \code{wtp} is -#' drawn through the reference strategy (or, without \code{ref_str}, through the least costly -#' strategy on the frontier). Strategies below the line are cost-effective compared with that -#' strategy at this threshold. -#' @param ref_axis with \code{ref_str}, draw dashed axes through the origin. Defaults to \code{TRUE}. -#' @param ref_caption add a caption naming the reference strategy and the WTP threshold. +#' drawn through the least costly strategy on the frontier. Strategies below the line are +#' cost-effective compared with that strategy at this threshold. +#' @param ref_axis with \code{incremental = TRUE}, draw dashed axes through the origin. #' Defaults to \code{TRUE}. +#' @param ref_caption add a caption naming the strategy at the origin (with +#' \code{incremental = TRUE}) and the WTP threshold (with \code{wtp}). Defaults to \code{TRUE}. #' @inheritParams ggrepel::geom_label_repel #' #' @return a ggplot2 object which can be modified by adding additional geoms @@ -548,7 +454,7 @@ plot.icers <- function(x, xexpand = expansion(0.1), yexpand = expansion(0.1), max.iter = 20000, - ref_str = NULL, + incremental = FALSE, wtp = NULL, ref_axis = TRUE, ref_caption = TRUE, @@ -558,11 +464,13 @@ plot.icers <- function(x, if (is.null(currency)) { currency <- if (is.null(attr(x, "currency"))) "$" else attr(x, "currency") } - ref_str <- check_ref_str(ref_str, x$Strategy) + if (!isTRUE(incremental) && !isFALSE(incremental)) { + stop("`incremental` must be TRUE or FALSE.", call. = FALSE) + } check_wtp_nhb(wtp, nhb = FALSE) # keep only the columns the plot uses, as a plain data frame, so the plot data are the same - # whether or not x has uncertainty bounds, reference columns, or stored attributes + # whether or not x has uncertainty bounds, net-benefit columns, or stored attributes v_plot_cols <- c("Strategy", "Cost", "Effect", "Inc_Cost", "Inc_Effect", "ICER", "Status") x <- data.frame(unclass(x)[v_plot_cols], check.names = FALSE, stringsAsFactors = FALSE) @@ -593,19 +501,16 @@ plot.icers <- function(x, eff_name <- "Effect" cost_name <- "Cost" - # point the WTP line goes through: the reference, or the least costly frontier strategy - if (!is.null(wtp)) { - df_front <- x[x$Status == nd_name, ] - df_anchor <- if (is.null(ref_str)) df_front[which.min(df_front$Cost), ] else x[x$Strategy == ref_str, ] - } + # first non-dominated strategy: the least costly strategy on the frontier. It is the origin of + # the incremental plane and the point the WTP line goes through + df_front <- x[x$Status == nd_name, ] + df_anchor <- df_front[which.min(df_front$Cost), ] - # incremental plane: put the reference strategy at the origin x_lab <- paste0("Effect (", effect_units, ")") y_lab <- paste0("Cost (", currency, ")") - if (!is.null(ref_str)) { - i_ref <- which(x$Strategy == ref_str) - x$Cost <- x$Cost - x$Cost[i_ref] - x$Effect <- x$Effect - x$Effect[i_ref] + if (incremental) { + x$Cost <- x$Cost - df_anchor$Cost + x$Effect <- x$Effect - df_anchor$Effect x_lab <- paste0("Incremental Effect (", effect_units, ")") y_lab <- paste0("Incremental Cost (", currency, ")") } @@ -626,17 +531,19 @@ plot.icers <- function(x, scale_shape_discrete(name = NULL) + labs(x = x_lab, y = y_lab) - if (!is.null(ref_str) && ref_axis) { + if (incremental && ref_axis) { icer_plot <- icer_plot + geom_hline(yintercept = 0, linetype = "dashed", color = "grey50") + geom_vline(xintercept = 0, linetype = "dashed", color = "grey50") } if (!is.null(wtp)) { - intercept <- if (is.null(ref_str)) df_anchor$Cost - wtp * df_anchor$Effect else 0 + # through the anchor strategy, which is at the origin on the incremental plane + intercept <- if (incremental) 0 else df_anchor$Cost - wtp * df_anchor$Effect icer_plot <- icer_plot + geom_abline(intercept = intercept, slope = wtp, linetype = "dotted", color = "grey30") } - v_caption <- c(if (!is.null(ref_str)) paste0("Reference Strategy: ", ref_str), + v_caption <- c(if (incremental) paste0("Origin: ", df_anchor$Strategy, + " (least costly strategy on the frontier)"), if (!is.null(wtp)) wtp_caption(wtp, currency, effect_units)) if (ref_caption && length(v_caption)) { icer_plot <- icer_plot + labs(caption = paste(v_caption, collapse = "; ")) @@ -676,12 +583,8 @@ plot.icers <- function(x, #' Print the formatted CEA table #' -#' By default, prints the sequential cost-effectiveness table: each strategy on the frontier is -#' compared with the previous one. If the \code{icers} object was created with a reference -#' strategy (\code{ref_str} in \code{\link{calculate_icers}}), \code{ref = "reference"} shows the -#' comparison of every strategy with the reference instead, listing the reference first and the -#' other strategies by increasing cost. \code{ref = "both"} shows both comparisons, in the -#' frontier order of the sequential table. +#' Prints the sequential cost-effectiveness table: each strategy on the frontier is compared with +#' the previous one. #' #' @param x Object of class \code{icers} #' @param currency string. with currency used in the cost-effectiveness analysis (CEA). @@ -689,8 +592,6 @@ plot.icers <- function(x, #' @param dig_cost integer. number of digits to round costs to #' @param dig_eff integer. number of digits to round effects to #' @param dig_icer integer. number of digits to round ICER to -#' @param ref which comparison to show: \code{"sequential"} (the default), \code{"reference"}, -#' or \code{"both"}. #' @param ... Additional arguments passed to other methods. #' @return a dataframe object - formatted CEA table #' @export @@ -699,22 +600,42 @@ print.icers <- function(x, dig_cost = 0, dig_eff = 2, dig_icer = 0, - ref = c("sequential", "reference", "both"), ...) { # check if icers object if (!inherits(x, "icers")) { stop("x must be an object of class icers") } - ref <- match.arg(ref) if (is.null(currency)) { currency <- if (is.null(attr(x, "currency"))) "$" else attr(x, "currency") } - ref_str <- attr(x, "ref_str") - if (ref != "sequential" && is.null(ref_str)) { - stop("x has no reference strategy; create it with calculate_icers(..., ref_str = ) ", - "to use ref = \"", ref, "\".", call. = FALSE) - } + df_table_cea <- x + class(df_table_cea) <- "data.frame" + + # display digits and header for each known column + l_format <- list( + Cost = list(dig_cost, paste0("Costs (", currency, ")")), + Effect = list(dig_eff, "QALYs"), + Inc_Cost = list(dig_cost, paste0("Incremental Costs (", currency, ")")), + Inc_Effect = list(dig_eff, "Incremental QALYs"), + ICER = list(dig_icer, paste0("ICER (", currency, "/QALY)")), + NMB = list(dig_cost, paste0("NMB (", currency, ")")), + NHB = list(dig_eff, "NHB (QALYs)") + ) + df_table_cea <- format_cea_table(df_table_cea, l_format) + + print(df_table_cea) + invisible(df_table_cea) +} + +#' Format the numeric columns of a CEA table for printing +#' +#' @param df_table data frame to format. +#' @param l_format named list, one element per column to format: a list of the number of digits +#' (\code{NA} to leave the column as it is) and the header to print. +#' @return \code{df_table} with the formatted columns, as character, and renamed. +#' @keywords internal +format_cea_table <- function(df_table, l_format) { format_numbers <- function(col, n_digits) { v_new_col <- rep("-", length(col)) v_non_na <- !is.na(col) @@ -725,49 +646,13 @@ print.icers <- function(x, v_new_col[v_non_na] <- format(round(col[v_non_na], n_digits), big.mark = ",", nsmall = n_digits, scientific = FALSE) } - v_new_col } - - # columns that belong to each view - v_seq_cols <- c("Inc_Cost", "Inc_Effect", "ICER", "Status") - v_ref_cols <- grep("_ref$", names(x), value = TRUE) - v_drop <- switch(ref, sequential = v_ref_cols, reference = v_seq_cols, both = character(0)) - df_table_cea <- x - class(df_table_cea) <- "data.frame" - df_table_cea <- df_table_cea[, setdiff(names(df_table_cea), v_drop), drop = FALSE] - # the reference view lists the reference first, then the other strategies by cost; the - # sequential and combined views keep the frontier order, in which the sequential ICERs are read - if (ref == "reference") { - v_order <- order(df_table_cea$Strategy != ref_str, df_table_cea$Cost) - df_table_cea <- df_table_cea[v_order, , drop = FALSE] - rownames(df_table_cea) <- NULL - } - - # display digits and header for each known column - vs <- paste0(" vs ", ref_str) - l_format <- list( - Cost = list(dig_cost, paste0("Costs (", currency, ")")), - Effect = list(dig_eff, "QALYs"), - Inc_Cost = list(dig_cost, paste0("Incremental Costs (", currency, ")")), - Inc_Effect = list(dig_eff, "Incremental QALYs"), - ICER = list(dig_icer, paste0("ICER (", currency, "/QALY)")), - Inc_Cost_ref = list(dig_cost, paste0("Incremental Costs", vs, " (", currency, ")")), - Inc_Effect_ref = list(dig_eff, paste0("Incremental QALYs", vs)), - ICER_ref = list(dig_icer, paste0("ICER", vs, " (", currency, "/QALY)")), - Status_ref = list(NA, paste0("Status", vs)), - NMB = list(dig_cost, paste0("NMB (", currency, ")")), - Inc_NMB_ref = list(dig_cost, paste0("Incremental NMB", vs, " (", currency, ")")), - NHB = list(dig_eff, "NHB (QALYs)"), - Inc_NHB_ref = list(dig_eff, paste0("Incremental NHB", vs, " (QALYs)")) - ) - for (col in intersect(names(l_format), names(df_table_cea))) { + for (col in intersect(names(l_format), names(df_table))) { if (!is.na(l_format[[col]][[1]])) { - df_table_cea[[col]] <- format_numbers(df_table_cea[[col]], n_digits = l_format[[col]][[1]]) + df_table[[col]] <- format_numbers(df_table[[col]], n_digits = l_format[[col]][[1]]) } - names(df_table_cea)[names(df_table_cea) == col] <- l_format[[col]][[2]] + names(df_table)[names(df_table) == col] <- l_format[[col]][[2]] } - - print(df_table_cea) - invisible(df_table_cea) + df_table } diff --git a/R/psa.R b/R/psa.R index 4bfc556..2013e66 100644 --- a/R/psa.R +++ b/R/psa.R @@ -104,31 +104,13 @@ check_df_and_coerce <- function(obj) { #' #' @param object the psa object #' @param calc_sds whether or not to calculate the standard deviations. Defaults to FALSE -#' @param ref_str name of a reference strategy. If given, each strategy is compared with it -#' sample by sample (see Value). Defaults to \code{NULL}. -#' @param wtp a single willingness-to-pay threshold. If given, the mean net monetary benefit is -#' added (see Value). Defaults to \code{NULL}. #' @param ... further arguments to summary (not used) #' #' @importFrom stats sd #' @return a \code{data.frame} containing the mean cost and effectiveness for each strategy and, if requested, #' the standard deviations of the cost and effectiveness for each strategy. -#' -#' With \code{ref_str}, the differences between each strategy and the reference are computed in -#' every PSA sample, and these columns are added (\code{NA} for the reference itself): -#' \code{meanIncCost} and \code{meanIncEffect}, the mean differences in cost and effect; and -#' \code{propNE}, \code{propSE}, \code{propSW}, and \code{propNW}, the proportion of samples in each -#' quadrant of the incremental cost-effectiveness plane: NE more effective and more costly, SE more -#' effective and not more costly (dominant), SW not more effective and not more costly, and NW not more -#' effective and more costly (dominated). The four proportions sum to 1. -#' -#' With \code{wtp}, \code{meanNMB} (\code{wtp * effect - cost}, averaged over samples) is added; with -#' \code{ref_str} too, also \code{meanIncNMB} and \code{propCostEffective}, the proportion of samples in -#' which the strategy has a higher NMB than the reference (it lies below the willingness-to-pay line). #' @export -summary.psa <- function(object, calc_sds = FALSE, ref_str = NULL, wtp = NULL, ...) { - ref_str <- check_ref_str(ref_str, object$strategies) - check_wtp_nhb(wtp, nhb = FALSE) +summary.psa <- function(object, calc_sds = FALSE, ...) { v_mean_cost <- colMeans(object$cost) v_mean_effect <- colMeans(object$effectiveness) @@ -143,30 +125,6 @@ summary.psa <- function(object, calc_sds = FALSE, ref_str = NULL, wtp = NULL, .. df_sum_psa[, "sdCost"] <- v_sd_cost df_sum_psa[, "sdEffect"] <- v_sd_effect } - if (!is.null(wtp)) { - df_sum_psa[, "meanNMB"] <- unname(colMeans(wtp * object$effectiveness - object$cost)) - } - if (!is.null(ref_str)) { - # sample-by-sample differences from the reference strategy - m_inc_cost <- as.matrix(object$cost) - object$cost[[ref_str]] - m_inc_effect <- as.matrix(object$effectiveness) - object$effectiveness[[ref_str]] - v_is_ref <- v_strat == ref_str - na_ref <- function(v) { - v[v_is_ref] <- NA_real_ - unname(v) - } - df_sum_psa[, "meanIncCost"] <- na_ref(colMeans(m_inc_cost)) - df_sum_psa[, "meanIncEffect"] <- na_ref(colMeans(m_inc_effect)) - df_sum_psa[, "propNE"] <- na_ref(colMeans(m_inc_effect > 0 & m_inc_cost > 0)) - df_sum_psa[, "propSE"] <- na_ref(colMeans(m_inc_effect > 0 & m_inc_cost <= 0)) - df_sum_psa[, "propSW"] <- na_ref(colMeans(m_inc_effect <= 0 & m_inc_cost <= 0)) - df_sum_psa[, "propNW"] <- na_ref(colMeans(m_inc_effect <= 0 & m_inc_cost > 0)) - if (!is.null(wtp)) { - m_inc_nmb <- wtp * m_inc_effect - m_inc_cost - df_sum_psa[, "meanIncNMB"] <- na_ref(colMeans(m_inc_nmb)) - df_sum_psa[, "propCostEffective"] <- na_ref(colMeans(m_inc_nmb > 0)) - } - } rownames(df_sum_psa) <- seq_len(nrow(df_sum_psa)) df_sum_psa } diff --git a/README.Rmd b/README.Rmd index b792e58..9237d51 100644 --- a/README.Rmd +++ b/README.Rmd @@ -35,7 +35,7 @@ dampack does these analyses in R for any model. It only needs the model's output ## What it does -- **Cost-effectiveness analysis.** `calculate_icers()` identifies dominated and extendedly dominated strategies and computes ICERs along the efficient frontier; `plot()` draws the frontier. +- **Cost-effectiveness analysis.** `calculate_icers()` identifies dominated and extendedly dominated strategies and computes ICERs along the efficient frontier; `plot()` draws the frontier on absolute or incremental axes. Where guidelines allow it alongside this fully incremental analysis, `calculate_cers_ref()` also compares every strategy with a single reference strategy, such as the standard of care. - **Probabilistic sensitivity analysis.** `make_psa_obj()` stores the PSA output of any model. Its `plot()` and `summary()` methods show the cost-effectiveness plane, `ceac()` computes cost-effectiveness acceptability curves (CEACs) together with the acceptability frontier (CEAF), and `calc_exp_loss()` computes expected loss curves (ELCs), which show in one figure the expected foregone benefit of each strategy, the optimal strategy, and the expected value of perfect information ([Alarid-Escudero et al., 2019](https://doi.org/10.1016/j.jval.2019.02.008)). - **Value of information.** `calc_evpi()` computes the expected value of perfect information (EVPI). `calc_evppi()` estimates the expected value of partial perfect information (EVPPI) with regression metamodels of the PSA ([Strong et al., 2014](https://doi.org/10.1177/0272989X13505910); [Jalal and Alarid-Escudero, 2018](https://doi.org/10.1177/0272989X17715627)), and `calc_evsi()` estimates the expected value of sample information (EVSI) of a future study with the Gaussian approximation approach ([Jalal and Alarid-Escudero, 2018](https://doi.org/10.1177/0272989X17715627)). Both work from the PSA sample you already have, so no further model runs are needed. - **Deterministic sensitivity analysis.** One-way and two-way analyses come either from the PSA, through metamodels (`owsa()`, `twsa()`), or from running your model directly (`run_owsa_det()`, `run_twsa_det()`), with tornado diagrams and optimal-strategy plots. diff --git a/README.md b/README.md index 6431c58..2569fa8 100644 --- a/README.md +++ b/README.md @@ -39,7 +39,11 @@ scientists, health economists, and students. - **Cost-effectiveness analysis.** `calculate_icers()` identifies dominated and extendedly dominated strategies and computes ICERs along - the efficient frontier; `plot()` draws the frontier. + the efficient frontier; `plot()` draws the frontier on absolute or + incremental axes. Where guidelines allow it alongside this fully + incremental analysis, `calculate_cers_ref()` also compares every + strategy with a single reference strategy, such as the standard of + care. - **Probabilistic sensitivity analysis.** `make_psa_obj()` stores the PSA output of any model. Its `plot()` and `summary()` methods show the cost-effectiveness plane, `ceac()` computes cost-effectiveness diff --git a/_pkgdown.yml b/_pkgdown.yml index 2219e97..f4e24c1 100644 --- a/_pkgdown.yml +++ b/_pkgdown.yml @@ -50,6 +50,17 @@ reference: - print.icers - plot.icers + - title: Comparisons with a reference strategy + desc: > + Compare every strategy with a single reference strategy, such as + the standard of care, where guidelines allow it alongside the fully + incremental analysis. + contents: + - calculate_cers_ref + - calculate_cers_ref_psa + - print.cers_ref + - plot.cers_ref + - title: Probabilistic sensitivity analysis (PSA) objects desc: > Build a `psa` object from the costs, effects, and parameter draws of any diff --git a/man/add_cers_ref_uncertainty.Rd b/man/add_cers_ref_uncertainty.Rd new file mode 100644 index 0000000..053f30b --- /dev/null +++ b/man/add_cers_ref_uncertainty.Rd @@ -0,0 +1,27 @@ +% Generated by roxygen2: do not edit by hand +% Please edit documentation in R/cers_ref.R +\name{add_cers_ref_uncertainty} +\alias{add_cers_ref_uncertainty} +\title{Add PSA intervals and quadrant proportions to a \code{cers_ref} object} +\usage{ +add_cers_ref_uncertainty(df_ref, psa, wtp, nhb) +} +\arguments{ +\item{df_ref}{a \code{cers_ref} object computed from the means of \code{psa}.} + +\item{psa}{the \code{psa} object.} + +\item{wtp}{a single willingness-to-pay threshold (currency per unit of effect). If given, the +net monetary benefit of each strategy and its difference from the reference are added. +The default (\code{NULL}) adds no net-benefit columns.} + +\item{nhb}{if \code{TRUE}, also add the net health benefit and its difference from the +reference. Requires \code{wtp}.} +} +\value{ +\code{df_ref} with the columns described in \code{\link{calculate_cers_ref_psa}}. +} +\description{ +Add PSA intervals and quadrant proportions to a \code{cers_ref} object +} +\keyword{internal} diff --git a/man/add_reference_and_nmb.Rd b/man/add_nmb.Rd similarity index 63% rename from man/add_reference_and_nmb.Rd rename to man/add_nmb.Rd index 7d4f97b..c7764b4 100644 --- a/man/add_reference_and_nmb.Rd +++ b/man/add_nmb.Rd @@ -1,19 +1,14 @@ % Generated by roxygen2: do not edit by hand % Please edit documentation in R/icers.R -\name{add_reference_and_nmb} -\alias{add_reference_and_nmb} -\title{Add reference-strategy and net-benefit columns to an icers object} +\name{add_nmb} +\alias{add_nmb} +\title{Add net-benefit columns and stored attributes to an icers object} \usage{ -add_reference_and_nmb(df_icers, ref_str, wtp, nhb, currency) +add_nmb(df_icers, wtp, nhb, currency) } \arguments{ \item{df_icers}{an \code{icers} object.} -\item{ref_str}{name of a reference strategy (one of \code{strategies}) to compare every -strategy against, for example \code{"Standard of care"}. The default (\code{NULL}) adds no -reference columns. The frontier-based results (\code{Status}, \code{Inc_Cost}, -\code{Inc_Effect}, \code{ICER}) never depend on the reference.} - \item{wtp}{a single willingness-to-pay threshold (currency per unit of effect). If given, the net monetary benefit (\code{NMB}) of each strategy is added. The default (\code{NULL}) adds no net-benefit columns.} diff --git a/man/add_reference_intervals.Rd b/man/add_reference_intervals.Rd deleted file mode 100644 index 45abbf9..0000000 --- a/man/add_reference_intervals.Rd +++ /dev/null @@ -1,30 +0,0 @@ -% Generated by roxygen2: do not edit by hand -% Please edit documentation in R/icers.R -\name{add_reference_intervals} -\alias{add_reference_intervals} -\title{Add 95\% intervals for the differences from a reference strategy} -\usage{ -add_reference_intervals(df_icers, psa, ref_str, wtp) -} -\arguments{ -\item{df_icers}{an \code{icers} object with reference columns.} - -\item{psa}{the \code{psa} object it was computed from.} - -\item{ref_str}{name of a reference strategy (one of \code{strategies}) to compare every -strategy against, for example \code{"Standard of care"}. The default (\code{NULL}) adds no -reference columns. The frontier-based results (\code{Status}, \code{Inc_Cost}, -\code{Inc_Effect}, \code{ICER}) never depend on the reference.} - -\item{wtp}{a single willingness-to-pay threshold (currency per unit of effect). If given, -the net monetary benefit (\code{NMB}) of each strategy is added. The default (\code{NULL}) -adds no net-benefit columns.} -} -\value{ -\code{df_icers} with the interval columns added. -} -\description{ -Quantiles of the sample-by-sample differences between each strategy and the reference, placed -after the corresponding point estimate. \code{NA} for the reference itself. -} -\keyword{internal} diff --git a/man/calculate_cers_ref.Rd b/man/calculate_cers_ref.Rd new file mode 100644 index 0000000..144c920 --- /dev/null +++ b/man/calculate_cers_ref.Rd @@ -0,0 +1,114 @@ +% Generated by roxygen2: do not edit by hand +% Please edit documentation in R/cers_ref.R +\name{calculate_cers_ref} +\alias{calculate_cers_ref} +\title{Compare every strategy with a reference strategy} +\usage{ +calculate_cers_ref( + cost, + effect, + strategies, + ref_str, + wtp = NULL, + nhb = FALSE, + currency = NULL +) +} +\arguments{ +\item{cost}{vector of cost for each strategy.} + +\item{effect}{vector of effect for each strategy.} + +\item{strategies}{string vector of strategy names.} + +\item{ref_str}{name of the reference strategy, one of \code{strategies}, for example +\code{"Standard of care"}.} + +\item{wtp}{a single willingness-to-pay threshold (currency per unit of effect). If given, the +net monetary benefit of each strategy and its difference from the reference are added. +The default (\code{NULL}) adds no net-benefit columns.} + +\item{nhb}{if \code{TRUE}, also add the net health benefit and its difference from the +reference. Requires \code{wtp}.} + +\item{currency}{currency symbol used when printing and plotting the results, for example +\code{"€"}. The default (\code{NULL}) uses \code{"$"}.} +} +\value{ +A data frame of class \code{cers_ref}, with the reference strategy in the first row +and the other strategies by increasing cost, and these columns: +\describe{ + \item{Strategy, Cost, Effect}{as given.} + \item{Inc_Cost, Inc_Effect}{cost and effect minus those of the reference.} + \item{CER}{\code{Inc_Cost / Inc_Effect}, the cost-effectiveness ratio compared with the + reference. It is \code{NA} where a ratio is not meaningful: for the reference itself and for + dominant, dominated and equivalent strategies.} + \item{Status}{one of \code{"Reference"}, \code{"Dominant"} (not more costly and at least as + effective), \code{"Dominated"} (not less costly and no more effective), + \code{"More costly, more effective"}, \code{"Less costly, less effective"}, or + \code{"Equivalent"} (same cost and effect).} + \item{NMB, Inc_NMB}{with \code{wtp}: the net monetary benefit, \code{wtp * Effect - Cost}, + and its difference from the reference. A strategy with a positive \code{Inc_NMB} is + cost-effective compared with the reference at this threshold.} + \item{NHB, Inc_NHB}{with \code{nhb = TRUE}: the net health benefit, + \code{Effect - Cost / wtp}, and its difference from the reference.} +} +The reference strategy, threshold and currency are stored as attributes, which the +\code{print} and \code{plot} methods use. + +For a strategy that is more costly and more effective than the reference, a \code{CER} below +the willingness-to-pay threshold means it is cost-effective compared with the reference. For a +strategy that is less costly and less effective, the \code{CER} is the cost saved per unit of +effect lost, and the strategy is cost-effective compared with the reference when its +\code{CER} is \emph{above} the threshold. +} +\description{ +Compares each strategy with a single reference strategy chosen by the user, such as the +current standard of care: the difference in cost and in effect from the reference, the +cost-effectiveness ratio (CER) of that difference, and where each strategy lies on the +incremental cost-effectiveness plane centered at the reference. + +These comparisons are \emph{not} the incremental cost-effectiveness ratios (ICERs) of a +fully incremental analysis. When more than two strategies are compared, health technology +assessment guidelines require a fully incremental (sequential) analysis, in which dominated +and extendedly dominated strategies are removed and each strategy on the cost-effectiveness +frontier is compared with the next less costly one: use \code{\link{calculate_icers}} for that. +Comparisons with a single reference are a complement for settings where they are relevant +and justified. The NICE manual (2022), for example, states that "pairwise comparisons may be +presented when relevant and justified (for example, when the technology is expected to +specifically displace individual comparators)", after requiring a fully incremental +analysis. The CDA-AMC (2017) and Zorginstituut Nederland (2024) guidelines also require a +sequential analysis when more than two interventions are compared. +} +\examples{ +# three strategies; "Standard of care" is neither the cheapest nor the least effective +df_ref <- calculate_cers_ref(cost = c(12000, 9000, 20000), + effect = c(8.0, 7.6, 8.5), + strategies = c("Standard of care", "Generic drug", "New drug"), + ref_str = "Standard of care", + wtp = 50000) +df_ref +plot(df_ref) + +# the fully incremental analysis of the same strategies +calculate_icers(cost = c(12000, 9000, 20000), + effect = c(8.0, 7.6, 8.5), + strategies = c("Standard of care", "Generic drug", "New drug")) +} +\references{ +National Institute for Health and Care Excellence (NICE). NICE health technology evaluations: +the manual (PMG36), section 4.6, "Economic evaluation". London: NICE; 2022. +\url{https://www.nice.org.uk/process/pmg36/chapter/economic-evaluation-2} + +CDA-AMC (formerly CADTH). Guidelines for the Economic Evaluation of Health Technologies: +Canada. 4th ed. Ottawa: CADTH; 2017. +\url{https://www.cda-amc.ca/sites/default/files/pdf/guidelines_for_the_economic_evaluation_of_health_technologies_canada_4th_ed.pdf} + +Zorginstituut Nederland. Guideline for economic evaluations in healthcare (2024 version). +Diemen: Zorginstituut Nederland; 2024. +\url{https://english.zorginstituutnederland.nl/documents/2024/01/16/guideline-for-economic-evaluations-in-healthcare} +} +\seealso{ +\code{\link{plot.cers_ref}}, \code{\link{calculate_cers_ref_psa}} for the same +comparison from a PSA, and \code{\link{calculate_icers}} for the fully incremental analysis. +} diff --git a/man/calculate_cers_ref_psa.Rd b/man/calculate_cers_ref_psa.Rd new file mode 100644 index 0000000..0a30074 --- /dev/null +++ b/man/calculate_cers_ref_psa.Rd @@ -0,0 +1,85 @@ +% Generated by roxygen2: do not edit by hand +% Please edit documentation in R/cers_ref.R +\name{calculate_cers_ref_psa} +\alias{calculate_cers_ref_psa} +\title{Compare every strategy with a reference strategy, from a \code{psa} object} +\usage{ +calculate_cers_ref_psa( + psa, + ref_str, + wtp = NULL, + nhb = FALSE, + uncertainty = FALSE +) +} +\arguments{ +\item{psa}{\code{psa} object from \code{\link{make_psa_obj}}.} + +\item{ref_str}{name of the reference strategy, one of \code{strategies}, for example +\code{"Standard of care"}.} + +\item{wtp}{a single willingness-to-pay threshold (currency per unit of effect). If given, the +net monetary benefit of each strategy and its difference from the reference are added. +The default (\code{NULL}) adds no net-benefit columns.} + +\item{nhb}{if \code{TRUE}, also add the net health benefit and its difference from the +reference. Requires \code{wtp}.} + +\item{uncertainty}{if \code{TRUE}, add the 95\% intervals and quadrant proportions described +in Value. Defaults to \code{FALSE}.} +} +\value{ +A \code{cers_ref} object with the columns described in +\code{\link{calculate_cers_ref}}, computed from the mean cost and effect of each strategy. The +currency of \code{psa} is kept. + +With \code{uncertainty = TRUE}, the differences from the reference are computed in every PSA +sample, and these columns are added (\code{NA} for the reference itself): +\itemize{ + \item \code{Lower_95_Inc_Cost} and \code{Upper_95_Inc_Cost}, the 2.5th and 97.5th + percentiles of the difference in cost, placed after \code{Inc_Cost}; likewise for + \code{Inc_Effect} and, with \code{wtp}, \code{Inc_NMB} (and \code{Inc_NHB}). + \item \code{Prop_NE}, \code{Prop_SE}, \code{Prop_SW}, and \code{Prop_NW}, the proportion of + samples in each quadrant of the incremental cost-effectiveness plane: NE more effective and + more costly; SE more effective and not more costly; SW not more effective and not more + costly; NW not more effective and more costly. The four proportions sum to 1. + \item With \code{wtp}, \code{Prop_CE}, the proportion of samples in which the strategy is + cost-effective compared with the reference (it has a higher net monetary benefit). +} +} +\description{ +Compares each strategy with a reference strategy using the mean costs and effects of a +probabilistic sensitivity analysis (PSA), as \code{\link{calculate_cers_ref}} does for a base +case. With \code{uncertainty = TRUE}, it also summarizes the uncertainty in each comparison +across the PSA samples. See \code{\link{calculate_cers_ref}} for when a comparison with a +single reference is appropriate, and \code{\link{calculate_icers_psa}} for the fully +incremental analysis. +} +\examples{ +data(cancer_psa_data) +l_psa <- make_psa_obj(cost = cancer_psa_data$cost, + effectiveness = cancer_psa_data$effectiveness, + parameters = cancer_psa_data$parameters, + strategies = cancer_psa_data$strategies) +df_ref <- calculate_cers_ref_psa(l_psa, ref_str = "Radio", wtp = 50000, + uncertainty = TRUE) +df_ref +plot(df_ref) +} +\references{ +National Institute for Health and Care Excellence (NICE). NICE health technology evaluations: +the manual (PMG36), section 4.6, "Economic evaluation". London: NICE; 2022. +\url{https://www.nice.org.uk/process/pmg36/chapter/economic-evaluation-2} + +CDA-AMC (formerly CADTH). Guidelines for the Economic Evaluation of Health Technologies: +Canada. 4th ed. Ottawa: CADTH; 2017. +\url{https://www.cda-amc.ca/sites/default/files/pdf/guidelines_for_the_economic_evaluation_of_health_technologies_canada_4th_ed.pdf} + +Zorginstituut Nederland. Guideline for economic evaluations in healthcare (2024 version). +Diemen: Zorginstituut Nederland; 2024. +\url{https://english.zorginstituutnederland.nl/documents/2024/01/16/guideline-for-economic-evaluations-in-healthcare} +} +\seealso{ +\code{\link{calculate_cers_ref}}, \code{\link{plot.cers_ref}}, and +\code{\link{plot.psa}} with \code{ref_str} for the PSA samples on the incremental plane. +} diff --git a/man/calculate_icers.Rd b/man/calculate_icers.Rd index e6e2fbb..c42d49e 100644 --- a/man/calculate_icers.Rd +++ b/man/calculate_icers.Rd @@ -8,7 +8,6 @@ calculate_icers( cost, effect, strategies, - ref_str = NULL, wtp = NULL, nhb = FALSE, currency = NULL @@ -21,11 +20,6 @@ calculate_icers( \item{strategies}{string vector of strategy names} -\item{ref_str}{name of a reference strategy (one of \code{strategies}) to compare every -strategy against, for example \code{"Standard of care"}. The default (\code{NULL}) adds no -reference columns. The frontier-based results (\code{Status}, \code{Inc_Cost}, -\code{Inc_Effect}, \code{ICER}) never depend on the reference.} - \item{wtp}{a single willingness-to-pay threshold (currency per unit of effect). If given, the net monetary benefit (\code{NMB}) of each strategy is added. The default (\code{NULL}) adds no net-benefit columns.} @@ -40,18 +34,14 @@ A data frame and \code{icers} object of strategies and their associated status, incremental cost, incremental effect, and ICER, computed sequentially along the cost-effectiveness frontier. -If \code{ref_str} is given, four columns compare each strategy with the reference: -\code{Inc_Cost_ref} and \code{Inc_Effect_ref} (cost and effect minus those of the -reference), \code{ICER_ref}, and \code{Status_ref}, one of \code{"Reference"}, -\code{"Dominant"} (less costly and at least as effective), \code{"Dominated"} (more costly -and no more effective), \code{"More costly, more effective"}, -\code{"Less costly, less effective"}, or \code{"Equivalent"} (same cost and effect). -\code{ICER_ref} is \code{NA} where a ratio is not meaningful: for the reference itself and -for dominant, dominated, and equivalent strategies. - -If \code{wtp} is given, \code{NMB} (\code{wtp * Effect - Cost}) is added, and with -\code{ref_str} also \code{Inc_NMB_ref}, the NMB minus that of the reference. With -\code{nhb = TRUE}, \code{NHB} (\code{Effect - Cost / wtp}) and \code{Inc_NHB_ref} are added too. +If \code{wtp} is given, \code{NMB} (\code{wtp * Effect - Cost}) is added. With +\code{nhb = TRUE}, \code{NHB} (\code{Effect - Cost / wtp}) is added too. + +The ICERs are always sequential: each strategy on the frontier is compared with the next less +costly one, after removing dominated and extendedly dominated strategies. This is the fully +incremental analysis that health technology assessment guidelines require when more than two +strategies are compared. To compare every strategy with a single reference strategy instead, +use \code{\link{calculate_cers_ref}}. } \description{ This function takes in strategies and their associated cost and effect, assigns them @@ -97,14 +87,14 @@ plot(icers) # note that longer strategy names will get truncated plot(icers, label = "all") -## Comparing every strategy with a reference strategy -hund_ref <- calculate_icers(hund_strat$Cost, - hund_strat$QALYs, - hund_strat$Strategy, - ref_str = hund_strat$Strategy[1], - wtp = 50000) -print(hund_ref, ref = "reference") +# the incremental cost-effectiveness plane, with the least costly strategy on the +# frontier at the origin +plot(icers, incremental = TRUE) + +## Net monetary benefit at a willingness-to-pay threshold +calculate_icers(sum_cdiff$meanCost, sum_cdiff$meanEffect, sum_cdiff$Strategy, + wtp = 50000) } \seealso{ -\code{\link{plot.icers}} +\code{\link{plot.icers}}, \code{\link{calculate_cers_ref}} } diff --git a/man/calculate_icers_psa.Rd b/man/calculate_icers_psa.Rd index 3766cea..a672a51 100644 --- a/man/calculate_icers_psa.Rd +++ b/man/calculate_icers_psa.Rd @@ -4,13 +4,7 @@ \alias{calculate_icers_psa} \title{Calculate incremental cost-effectiveness ratios from a \code{psa} object.} \usage{ -calculate_icers_psa( - psa, - uncertainty = FALSE, - ref_str = NULL, - wtp = NULL, - nhb = FALSE -) +calculate_icers_psa(psa, uncertainty = FALSE, wtp = NULL, nhb = FALSE) } \arguments{ \item{psa}{\code{psa} object from \code{link{make_psa_object}}} @@ -18,11 +12,6 @@ calculate_icers_psa( \item{uncertainty}{whether or not 95\% quantiles for the cost and QALY outcomes should be included in the resulting ICER table. Defaults to \code{FALSE}.} -\item{ref_str}{name of a reference strategy (one of \code{strategies}) to compare every -strategy against, for example \code{"Standard of care"}. The default (\code{NULL}) adds no -reference columns. The frontier-based results (\code{Status}, \code{Inc_Cost}, -\code{Inc_Effect}, \code{ICER}) never depend on the reference.} - \item{wtp}{a single willingness-to-pay threshold (currency per unit of effect). If given, the net monetary benefit (\code{NMB}) of each strategy is added. The default (\code{NULL}) adds no net-benefit columns.} @@ -35,11 +24,8 @@ status, cost, effect, incremental cost, incremental effect, and ICER. If \code{u set to \code{TRUE}, four additional columns are provided for the 2.5th and 97.5th quantiles for each strategy's cost and effect. -With \code{ref_str}, \code{wtp}, or \code{nhb}, the reference-strategy and net-benefit columns -described in \code{\link{calculate_icers}} are added. With \code{uncertainty = TRUE} and -\code{ref_str}, the 2.5th and 97.5th quantiles of the sample-by-sample differences from the reference -are also given for \code{Inc_Cost_ref}, \code{Inc_Effect_ref}, and (with \code{wtp}) -\code{Inc_NMB_ref}, in columns such as \code{Lower_95_Inc_Cost_ref} and \code{Upper_95_Inc_Cost_ref}. +With \code{wtp} or \code{nhb}, the net-benefit columns described in +\code{\link{calculate_icers}} are added. } \description{ The mean costs and QALYs for each strategy in a PSA are used @@ -60,4 +46,6 @@ even if present in the ICER table. \code{\link{plot.icers}} \code{\link{calculate_icers}} + +\code{\link{calculate_cers_ref_psa}} to compare every strategy with a reference strategy } diff --git a/man/format_cea_table.Rd b/man/format_cea_table.Rd new file mode 100644 index 0000000..d5bec25 --- /dev/null +++ b/man/format_cea_table.Rd @@ -0,0 +1,21 @@ +% Generated by roxygen2: do not edit by hand +% Please edit documentation in R/icers.R +\name{format_cea_table} +\alias{format_cea_table} +\title{Format the numeric columns of a CEA table for printing} +\usage{ +format_cea_table(df_table, l_format) +} +\arguments{ +\item{df_table}{data frame to format.} + +\item{l_format}{named list, one element per column to format: a list of the number of digits +(\code{NA} to leave the column as it is) and the header to print.} +} +\value{ +\code{df_table} with the formatted columns, as character, and renamed. +} +\description{ +Format the numeric columns of a CEA table for printing +} +\keyword{internal} diff --git a/man/plot.cers_ref.Rd b/man/plot.cers_ref.Rd new file mode 100644 index 0000000..22a3d92 --- /dev/null +++ b/man/plot.cers_ref.Rd @@ -0,0 +1,114 @@ +% Generated by roxygen2: do not edit by hand +% Please edit documentation in R/cers_ref.R +\name{plot.cers_ref} +\alias{plot.cers_ref} +\title{Plot a comparison with a reference strategy} +\usage{ +\method{plot}{cers_ref}( + x, + wtp = attr(x, "wtp"), + lines = NULL, + label = c("all", "none"), + label_max_char = NULL, + ref_axis = TRUE, + ref_caption = TRUE, + currency = NULL, + effect_units = "QALYs", + col = c("access", "full", "bw"), + alpha = 1, + txtsize = 12, + n_x_ticks = 6, + n_y_ticks = 6, + xbreaks = NULL, + ybreaks = NULL, + xlim = NULL, + ylim = NULL, + xexpand = expansion(0.1), + yexpand = expansion(0.1), + max.iter = 20000, + ... +) +} +\arguments{ +\item{x}{a \code{cers_ref} object from \code{\link{calculate_cers_ref}} or +\code{\link{calculate_cers_ref_psa}}.} + +\item{wtp}{a willingness-to-pay threshold for the dotted line through the origin. Defaults to +the threshold stored in \code{x}, if any; \code{NULL} draws no line.} + +\item{lines}{whether to draw a segment from the reference to each strategy. The default +(\code{NULL}) draws them when there are at most 20 strategies besides the reference, since +with many strategies they make the plot hard to read.} + +\item{label}{whether to label all strategies (the default) or none.} + +\item{label_max_char}{maximum number of characters of the strategy labels; longer names are +truncated. The default (\code{NULL}) does not truncate.} + +\item{ref_axis}{draw dashed axes through the origin. Defaults to \code{TRUE}.} + +\item{ref_caption}{add a caption naming the reference strategy and the WTP threshold. +Defaults to \code{TRUE}.} + +\item{currency}{currency symbol. Defaults to the currency stored in \code{x}, or \code{"$"}.} + +\item{effect_units}{units of effectiveness. Defaults to \code{"QALYs"}.} + +\item{col}{color palette for the status of each strategy: \code{"access"} (the default, an +accessible palette), \code{"full"}, or \code{"bw"}.} + +\item{alpha}{opacity of the points.} + +\item{txtsize}{base text size} + +\item{n_x_ticks}{number of x-axis ticks} + +\item{n_y_ticks}{number of y-axis ticks} + +\item{xbreaks}{vector of x-axis breaks. +will override \code{n_x_ticks} if provided.} + +\item{ybreaks}{vector of y-axis breaks. +will override \code{n_y_ticks} if provided.} + +\item{xlim}{vector of x-axis limits, or NULL, which sets limits automatically} + +\item{ylim}{vector of y-axis limits, or NULL, which sets limits automatically} + +\item{xexpand}{Padding around data. See \code{\link[ggplot2]{scale_continuous}} for details.} + +\item{yexpand}{Padding around data. See \code{\link[ggplot2]{scale_continuous}} for details. +The default behavior in ggplot2 is \code{expansion(0.05)}. See \code{\link[ggplot2]{expansion}} +for how to modify this.} + +\item{max.iter}{Maximum number of iterations to try to resolve overlaps. +Defaults to 10000.} + +\item{...}{further arguments (not used).} +} +\value{ +a \code{ggplot2} object, which can be modified by adding further layers. +} +\description{ +Draws the incremental cost-effectiveness plane centered at the reference strategy: each +strategy is placed at its difference in effect (x) and cost (y) from the reference, which sits +at the origin. Optional segments join the reference to each strategy, forming a star, and an +optional willingness-to-pay line goes through the origin: strategies below the line are +cost-effective compared with the reference at that threshold. +} +\examples{ +df_ref <- calculate_cers_ref(cost = c(12000, 9000, 20000), + effect = c(8.0, 7.6, 8.5), + strategies = c("Standard of care", "Generic drug", "New drug"), + ref_str = "Standard of care") +plot(df_ref, wtp = 50000) + +# many strategies: no segments, no labels +data(hund_strat) +df_hund <- calculate_cers_ref(hund_strat$Cost, hund_strat$QALYs, hund_strat$Strategy, + ref_str = "1") +plot(df_hund, label = "none") +} +\seealso{ +\code{\link{calculate_cers_ref}} +} diff --git a/man/plot.icers.Rd b/man/plot.icers.Rd index af2dfdd..bd06faf 100644 --- a/man/plot.icers.Rd +++ b/man/plot.icers.Rd @@ -22,7 +22,7 @@ xexpand = expansion(0.1), yexpand = expansion(0.1), max.iter = 20000, - ref_str = NULL, + incremental = FALSE, wtp = NULL, ref_axis = TRUE, ref_caption = TRUE, @@ -71,20 +71,23 @@ for how to modify this.} \item{max.iter}{Maximum number of iterations to try to resolve overlaps. Defaults to 10000.} -\item{ref_str}{name of a reference strategy. If given, the plot shows the incremental -cost-effectiveness plane, with the reference strategy at the origin. The frontier and the -status of each strategy do not change. Defaults to \code{NULL} (the cost-effectiveness plane).} +\item{incremental}{if \code{TRUE}, plot the incremental cost-effectiveness plane: the costs +and effects of every strategy are shown relative to those of the first non-dominated strategy +(the least costly strategy on the frontier), which is placed at the origin (0, 0). The frontier +and the status of each strategy do not change. Defaults to \code{FALSE} (absolute costs and +effects). To compare every strategy with a reference strategy of your choice, use +\code{\link{calculate_cers_ref}} and its \code{plot} method.} \item{wtp}{a willingness-to-pay threshold. If given, a dotted line with slope \code{wtp} is -drawn through the reference strategy (or, without \code{ref_str}, through the least costly -strategy on the frontier). Strategies below the line are cost-effective compared with that -strategy at this threshold.} +drawn through the least costly strategy on the frontier. Strategies below the line are +cost-effective compared with that strategy at this threshold.} -\item{ref_axis}{with \code{ref_str}, draw dashed axes through the origin. Defaults to \code{TRUE}.} - -\item{ref_caption}{add a caption naming the reference strategy and the WTP threshold. +\item{ref_axis}{with \code{incremental = TRUE}, draw dashed axes through the origin. Defaults to \code{TRUE}.} +\item{ref_caption}{add a caption naming the strategy at the origin (with +\code{incremental = TRUE}) and the WTP threshold (with \code{wtp}). Defaults to \code{TRUE}.} + \item{...}{further arguments to plot. This is not used by \code{dampack} but required for generic consistency.} } diff --git a/man/print.cers_ref.Rd b/man/print.cers_ref.Rd new file mode 100644 index 0000000..4ebffff --- /dev/null +++ b/man/print.cers_ref.Rd @@ -0,0 +1,38 @@ +% Generated by roxygen2: do not edit by hand +% Please edit documentation in R/cers_ref.R +\name{print.cers_ref} +\alias{print.cers_ref} +\title{Print a comparison with a reference strategy} +\usage{ +\method{print}{cers_ref}( + x, + currency = NULL, + dig_cost = 0, + dig_eff = 2, + dig_cer = 0, + dig_prop = 3, + ... +) +} +\arguments{ +\item{x}{a \code{cers_ref} object.} + +\item{currency}{currency symbol. Defaults to the currency stored in \code{x}, or \code{"$"}.} + +\item{dig_cost}{number of digits to round costs and net monetary benefits to.} + +\item{dig_eff}{number of digits to round effects and net health benefits to.} + +\item{dig_cer}{number of digits to round the cost-effectiveness ratios to.} + +\item{dig_prop}{number of digits to round the proportions of PSA samples to.} + +\item{...}{further arguments (not used).} +} +\value{ +the formatted table, invisibly. +} +\description{ +Prints the table of a \code{\link{calculate_cers_ref}} object with formatted numbers and +headers that name the reference strategy. +} diff --git a/man/print.icers.Rd b/man/print.icers.Rd index 558f9aa..677b0f4 100644 --- a/man/print.icers.Rd +++ b/man/print.icers.Rd @@ -4,15 +4,7 @@ \alias{print.icers} \title{Print the formatted CEA table} \usage{ -\method{print}{icers}( - x, - currency = NULL, - dig_cost = 0, - dig_eff = 2, - dig_icer = 0, - ref = c("sequential", "reference", "both"), - ... -) +\method{print}{icers}(x, currency = NULL, dig_cost = 0, dig_eff = 2, dig_icer = 0, ...) } \arguments{ \item{x}{Object of class \code{icers}} @@ -26,19 +18,12 @@ Defaults to the currency stored in \code{x}, or \code{"$"}.} \item{dig_icer}{integer. number of digits to round ICER to} -\item{ref}{which comparison to show: \code{"sequential"} (the default), \code{"reference"}, -or \code{"both"}.} - \item{...}{Additional arguments passed to other methods.} } \value{ a dataframe object - formatted CEA table } \description{ -By default, prints the sequential cost-effectiveness table: each strategy on the frontier is -compared with the previous one. If the \code{icers} object was created with a reference -strategy (\code{ref_str} in \code{\link{calculate_icers}}), \code{ref = "reference"} shows the -comparison of every strategy with the reference instead, listing the reference first and the -other strategies by increasing cost. \code{ref = "both"} shows both comparisons, in the -frontier order of the sequential table. +Prints the sequential cost-effectiveness table: each strategy on the frontier is compared with +the previous one. } diff --git a/man/status_vs_reference.Rd b/man/status_vs_reference.Rd index 10def36..8fcd925 100644 --- a/man/status_vs_reference.Rd +++ b/man/status_vs_reference.Rd @@ -1,5 +1,5 @@ % Generated by roxygen2: do not edit by hand -% Please edit documentation in R/icers.R +% Please edit documentation in R/cers_ref.R \name{status_vs_reference} \alias{status_vs_reference} \title{Status of each strategy relative to a reference strategy} diff --git a/man/summary.psa.Rd b/man/summary.psa.Rd index ca0a145..9251f93 100644 --- a/man/summary.psa.Rd +++ b/man/summary.psa.Rd @@ -4,36 +4,18 @@ \alias{summary.psa} \title{summarize a psa object across all simulations} \usage{ -\method{summary}{psa}(object, calc_sds = FALSE, ref_str = NULL, wtp = NULL, ...) +\method{summary}{psa}(object, calc_sds = FALSE, ...) } \arguments{ \item{object}{the psa object} \item{calc_sds}{whether or not to calculate the standard deviations. Defaults to FALSE} -\item{ref_str}{name of a reference strategy. If given, each strategy is compared with it -sample by sample (see Value). Defaults to \code{NULL}.} - -\item{wtp}{a single willingness-to-pay threshold. If given, the mean net monetary benefit is -added (see Value). Defaults to \code{NULL}.} - \item{...}{further arguments to summary (not used)} } \value{ a \code{data.frame} containing the mean cost and effectiveness for each strategy and, if requested, the standard deviations of the cost and effectiveness for each strategy. - -With \code{ref_str}, the differences between each strategy and the reference are computed in -every PSA sample, and these columns are added (\code{NA} for the reference itself): -\code{meanIncCost} and \code{meanIncEffect}, the mean differences in cost and effect; and -\code{propNE}, \code{propSE}, \code{propSW}, and \code{propNW}, the proportion of samples in each -quadrant of the incremental cost-effectiveness plane: NE more effective and more costly, SE more -effective and not more costly (dominant), SW not more effective and not more costly, and NW not more -effective and more costly (dominated). The four proportions sum to 1. - -With \code{wtp}, \code{meanNMB} (\code{wtp * effect - cost}, averaged over samples) is added; with -\code{ref_str} too, also \code{meanIncNMB} and \code{propCostEffective}, the proportion of samples in -which the strategy has a higher NMB than the reference (it lies below the willingness-to-pay line). } \description{ summarize a psa object across all simulations diff --git a/tests/testthat/test_cers_ref.R b/tests/testthat/test_cers_ref.R new file mode 100644 index 0000000..4afebc2 --- /dev/null +++ b/tests/testthat/test_cers_ref.R @@ -0,0 +1,178 @@ +# calculate_cers_ref() and calculate_cers_ref_psa(): every strategy against a reference (#209) + +# Toy example with one strategy in each position relative to the reference "Ref": +# strategy cost effect vs Ref: inc cost inc effect status +# Ref 100 1.0 - - Reference +# More 200 2.0 100 1.0 More costly, more effective (CER 100) +# Cheaper 50 1.5 -50 0.5 Dominant +# Worse 300 0.5 200 -0.5 Dominated +# Less 80 0.8 -20 -0.2 Less costly, less effective (CER 100) +# Same 100 1.0 0 0.0 Equivalent +df_toy <- data.frame(strategy = c("Ref", "More", "Cheaper", "Worse", "Less", "Same"), + cost = c(100, 200, 50, 300, 80, 100), + effect = c(1, 2, 1.5, 0.5, 0.8, 1), + stringsAsFactors = FALSE) +toy_cers <- function(ref_str = "Ref", ...) { + calculate_cers_ref(df_toy$cost, df_toy$effect, df_toy$strategy, ref_str = ref_str, ...) +} +row_of <- function(x, s) x[x$Strategy == s, ] + +test_that("comparisons with the reference match hand calculations", { + x <- toy_cers() + expect_s3_class(x, "cers_ref") + expect_identical(names(x), c("Strategy", "Cost", "Effect", "Inc_Cost", "Inc_Effect", "CER", "Status")) + expect_identical(attr(x, "ref_str"), "Ref") + + expected <- list( + Ref = list(NA_real_, NA_real_, NA_real_, "Reference"), + More = list(100, 1, 100, "More costly, more effective"), + Cheaper = list(-50, 0.5, NA_real_, "Dominant"), + Worse = list(200, -0.5, NA_real_, "Dominated"), + Less = list(-20, -0.2, 100, "Less costly, less effective"), + Same = list(0, 0, NA_real_, "Equivalent") + ) + for (s in names(expected)) { + r <- row_of(x, s) + expect_equal(r$Inc_Cost, expected[[s]][[1]], info = s) + expect_equal(r$Inc_Effect, expected[[s]][[2]], info = s) + expect_equal(r$CER, expected[[s]][[3]], info = s) + expect_equal(r$Status, expected[[s]][[4]], info = s) + } + expect_type(x$CER, "double") +}) + +test_that("the reference comes first, then the other strategies by cost", { + x <- toy_cers(ref_str = "More") + expect_equal(x$Strategy, c("More", "Cheaper", "Less", "Ref", "Same", "Worse")) + + data(hund_strat) + y <- calculate_cers_ref(hund_strat$Cost, hund_strat$QALYs, hund_strat$Strategy, ref_str = "1") + v_rest <- hund_strat[hund_strat$Strategy != 1, ] + expect_equal(y$Strategy, c("1", as.character(v_rest$Strategy[order(v_rest$Cost)]))) + r0 <- row_of(y, "1") + others <- y$Strategy != "1" + expect_equal(y$Inc_Cost[others], y$Cost[others] - r0$Cost) + expect_equal(y$Inc_Effect[others], y$Effect[others] - r0$Effect) +}) + +test_that("a dominated or the most expensive strategy can be the reference", { + x <- toy_cers(ref_str = "Worse") # Worse is dominated and the most expensive strategy + expect_equal(row_of(x, "Worse")$Status, "Reference") + expect_equal(row_of(x, "Cheaper")$Status, "Dominant") + expect_equal(row_of(x, "Cheaper")$Inc_Cost, 50 - 300) + expect_true(all(x$Inc_Cost[x$Strategy != "Worse"] < 0)) +}) + +test_that("wtp adds NMB and its difference from the reference; nhb adds NHB", { + x <- toy_cers(wtp = 150) + expect_equal(x$NMB, 150 * x$Effect - x$Cost) + expect_equal(row_of(x, "More")$Inc_NMB, (150 * 2 - 200) - (150 * 1 - 100)) + expect_true(is.na(row_of(x, "Ref")$Inc_NMB)) + expect_false(any(c("NHB", "Inc_NHB") %in% names(x))) + expect_identical(attr(x, "wtp"), 150) + + z <- toy_cers(wtp = 150, nhb = TRUE) + expect_equal(z$NHB, z$Effect - z$Cost / 150) + expect_equal(row_of(z, "Cheaper")$Inc_NHB, (1.5 - 50 / 150) - (1 - 100 / 150)) +}) + +test_that("edge cases: one strategy, two strategies, names with spaces", { + one <- calculate_cers_ref(10, 1, "Standard of care", ref_str = "Standard of care", wtp = 100) + expect_equal(one$Status, "Reference") + expect_equal(one$NMB, 90) + + two <- calculate_cers_ref(c(10, 30), c(1, 2), c("Standard of care", "New drug (A)"), + ref_str = "Standard of care") + expect_equal(row_of(two, "New drug (A)")$CER, 20) + expect_equal(row_of(two, "New drug (A)")$Status, "More costly, more effective") +}) + +test_that("invalid inputs give informative errors", { + expect_error(calculate_cers_ref(df_toy$cost, df_toy$effect, df_toy$strategy), "must name the reference") + expect_error(toy_cers(ref_str = "Nope"), "must be one of the strategies") + expect_error(toy_cers(ref_str = c("Ref", "More")), "single strategy name") + expect_error(toy_cers(wtp = c(100, 200)), "single non-negative number") + expect_error(toy_cers(wtp = -1), "single non-negative number") + expect_error(toy_cers(nhb = TRUE), "requires `wtp`") + expect_error(calculate_cers_ref(1:2, 1:3, c("a", "b"), ref_str = "a"), "same length") + expect_error(calculate_cers_ref(1:2, 1:2, c("a", "a"), ref_str = "a"), "unique") +}) + +test_that("print() names the reference and formats the table", { + x <- toy_cers(wtp = 150, currency = "€") + out <- capture.output(tbl <- print(x)) + expect_true(any(grepl("not a fully incremental analysis", out, fixed = TRUE))) + expect_true("CER vs Ref (€/QALY)" %in% names(tbl)) + expect_true("Incremental NMB vs Ref (€)" %in% names(tbl)) + expect_equal(tbl[["CER vs Ref (€/QALY)"]][1], "-") +}) + +# calculate_icers() stays fully incremental: no reference strategy +test_that("calculate_icers() has no reference strategy", { + expect_false("ref_str" %in% names(formals(calculate_icers))) + expect_false("ref_str" %in% names(formals(calculate_icers_psa))) + expect_false("ref_str" %in% names(formals(summary.psa))) + x <- calculate_icers(df_toy$cost, df_toy$effect, df_toy$strategy) + expect_identical(names(x), c("Strategy", "Cost", "Effect", "Inc_Cost", "Inc_Effect", "ICER", "Status")) + y <- calculate_icers(df_toy$cost, df_toy$effect, df_toy$strategy, wtp = 150, nhb = TRUE) + expect_equal(y$NMB, 150 * y$Effect - y$Cost) + expect_equal(y$NHB, y$Effect - y$Cost / 150) + expect_identical(attr(y, "wtp"), 150) + expect_error(calculate_icers(df_toy$cost, df_toy$effect, df_toy$strategy, nhb = TRUE), + "requires `wtp`") +}) + +# PSA ---------------------------------------------------------------------------------------- + +# Four PSA samples; the difference "New" - "Ref" falls in a different quadrant in each: +# sample inc effect inc cost quadrant +# 1 +0.2 +100 NE (more effective, more costly) +# 2 +0.2 -50 SE (more effective, less costly) +# 3 -0.1 -50 SW (less effective, less costly) +# 4 -0.1 +100 NW (less effective, more costly) +df_cost <- data.frame(Ref = c(1000, 1000, 1000, 1000), New = c(1100, 950, 950, 1100)) +df_eff <- data.frame(Ref = c(1, 1, 1, 1), New = c(1.2, 1.2, 0.9, 0.9)) +psa_q <- make_psa_obj(df_cost, df_eff, strategies = c("Ref", "New"), currency = "€") + +test_that("calculate_cers_ref_psa() uses the means of the PSA and keeps its currency", { + s <- summary(psa_q) + expected <- calculate_cers_ref(s$meanCost, s$meanEffect, s$Strategy, ref_str = "Ref", + wtp = 1000, currency = "€") + expect_equal(calculate_cers_ref_psa(psa_q, ref_str = "Ref", wtp = 1000), expected) + expect_identical(attr(expected, "currency"), "€") +}) + +test_that("uncertainty = TRUE gives quadrant shares and the share of cost-effective samples", { + x <- calculate_cers_ref_psa(psa_q, ref_str = "Ref", wtp = 1000, uncertainty = TRUE) + r_new <- row_of(x, "New") + expect_equal(unlist(r_new[, c("Prop_NE", "Prop_SE", "Prop_SW", "Prop_NW")], use.names = FALSE), + rep(0.25, 4)) + inmb <- 1000 * (df_eff$New - df_eff$Ref) - (df_cost$New - df_cost$Ref) # 100, 250, -50, -200 + expect_equal(r_new$Prop_CE, mean(inmb > 0)) + expect_equal(r_new$Lower_95_Inc_NMB, unname(quantile(inmb, 0.025))) + r_ref <- row_of(x, "Ref") + expect_true(all(is.na(r_ref[, c("Lower_95_Inc_Cost", "Prop_NE", "Prop_CE")]))) + # each interval follows its point estimate + v_names <- names(x) + expect_equal(match("Lower_95_Inc_Cost", v_names), match("Inc_Cost", v_names) + 1) + expect_identical(attr(x, "ref_str"), "Ref") + expect_s3_class(x, "cers_ref") +}) + +test_that("uncertainty = TRUE gives intervals of the sample-by-sample differences", { + data("cancer_psa_data") + psa_big <- make_psa_obj(cancer_psa_data$cost, cancer_psa_data$effectiveness, + cancer_psa_data$parameters, cancer_psa_data$strategies) + x <- calculate_cers_ref_psa(psa_big, ref_str = "Surgery", wtp = 50000, uncertainty = TRUE) + i_chemo <- which(cancer_psa_data$strategies == "Chemo") + i_surg <- which(cancer_psa_data$strategies == "Surgery") + r <- row_of(x, "Chemo") + d_cost <- cancer_psa_data$cost[[i_chemo]] - cancer_psa_data$cost[[i_surg]] + d_eff <- cancer_psa_data$effectiveness[[i_chemo]] - cancer_psa_data$effectiveness[[i_surg]] + expect_equal(r$Lower_95_Inc_Cost, unname(quantile(d_cost, 0.025))) + expect_equal(r$Upper_95_Inc_Cost, unname(quantile(d_cost, 0.975))) + expect_equal(r$Lower_95_Inc_Effect, unname(quantile(d_eff, 0.025))) + expect_equal(r$Upper_95_Inc_NMB, unname(quantile(50000 * d_eff - d_cost, 0.975))) + out <- capture.output(tbl <- print(x)) + expect_match(tbl$Prop_NE[2], "^0\\.[0-9]{3}$") +}) diff --git a/tests/testthat/test_icers_reference.R b/tests/testthat/test_icers_reference.R deleted file mode 100644 index 89f9c4b..0000000 --- a/tests/testthat/test_icers_reference.R +++ /dev/null @@ -1,136 +0,0 @@ -# calculate_icers() against a reference strategy (#209) and net benefit (#200) - -# Toy example with one strategy in each position relative to the reference "Ref": -# strategy cost effect vs Ref: inc cost inc effect status -# Ref 100 1.0 - - Reference -# More 200 2.0 100 1.0 More costly, more effective (ICER 100) -# Cheaper 50 1.5 -50 0.5 Dominant -# Worse 300 0.5 200 -0.5 Dominated -# Less 80 0.8 -20 -0.2 Less costly, less effective (ICER 100) -# Same 100 1.0 0 0.0 Equivalent -df_toy <- data.frame(strategy = c("Ref", "More", "Cheaper", "Worse", "Less", "Same"), - cost = c(100, 200, 50, 300, 80, 100), - effect = c(1, 2, 1.5, 0.5, 0.8, 1), - stringsAsFactors = FALSE) -toy_icers <- function(...) { - calculate_icers(df_toy$cost, df_toy$effect, df_toy$strategy, ...) -} -row_of <- function(x, s) x[x$Strategy == s, ] - -test_that("without ref_str or wtp the output is unchanged", { - x <- toy_icers() - expect_identical(names(x), c("Strategy", "Cost", "Effect", "Inc_Cost", "Inc_Effect", "ICER", "Status")) - expect_null(attr(x, "ref_str")) -}) - -test_that("reference columns match hand calculations", { - x <- toy_icers(ref_str = "Ref") - # sequential (frontier) results are unchanged by ref_str - x0 <- toy_icers() - expect_equal(as.data.frame(x)[, names(x0)], as.data.frame(x0)) - expect_identical(attr(x, "ref_str"), "Ref") - - expected <- list( - Ref = list(NA_real_, NA_real_, NA_real_, "Reference"), - More = list(100, 1, 100, "More costly, more effective"), - Cheaper = list(-50, 0.5, NA_real_, "Dominant"), - Worse = list(200, -0.5, NA_real_, "Dominated"), - Less = list(-20, -0.2, 100, "Less costly, less effective"), - Same = list(0, 0, NA_real_, "Equivalent") - ) - for (s in names(expected)) { - r <- row_of(x, s) - expect_equal(r$Inc_Cost_ref, expected[[s]][[1]], info = s) - expect_equal(r$Inc_Effect_ref, expected[[s]][[2]], info = s) - expect_equal(r$ICER_ref, expected[[s]][[3]], info = s) - expect_equal(r$Status_ref, expected[[s]][[4]], info = s) - } - expect_type(x$ICER_ref, "double") -}) - -test_that("a dominated or the most expensive strategy can be the reference", { - x <- toy_icers(ref_str = "Worse") - expect_equal(row_of(x, "Worse")$Status_ref, "Reference") - expect_equal(row_of(x, "Cheaper")$Status_ref, "Dominant") - expect_equal(row_of(x, "Cheaper")$Inc_Cost_ref, 50 - 300) - # the frontier itself does not depend on which strategy is the reference - expect_equal(x$Status, toy_icers()$Status) - - y <- toy_icers(ref_str = "Worse") # Worse is also the most expensive strategy - expect_true(all(y$Inc_Cost_ref[y$Strategy != "Worse"] < 0)) -}) - -test_that("NMB is added by default with wtp, NHB only on request", { - x <- toy_icers(wtp = 150) - expect_equal(x$NMB, 150 * x$Effect - x$Cost) - expect_false(any(c("NHB", "Inc_NMB_ref", "Inc_NHB_ref") %in% names(x))) - expect_identical(attr(x, "wtp"), 150) - - y <- toy_icers(ref_str = "Ref", wtp = 150) - expect_equal(row_of(y, "More")$Inc_NMB_ref, (150 * 2 - 200) - (150 * 1 - 100)) - expect_true(is.na(row_of(y, "Ref")$Inc_NMB_ref)) - expect_false("NHB" %in% names(y)) - - z <- toy_icers(ref_str = "Ref", wtp = 150, nhb = TRUE) - expect_equal(z$NHB, z$Effect - z$Cost / 150) - expect_equal(row_of(z, "Cheaper")$Inc_NHB_ref, (1.5 - 50 / 150) - (1 - 100 / 150)) -}) - -test_that("hund_strat: reference columns equal the difference from the reference", { - data("hund_strat") - ref <- hund_strat$Strategy[3] - x <- calculate_icers(hund_strat$Cost, hund_strat$QALYs, hund_strat$Strategy, ref_str = ref) - r0 <- row_of(x, ref) - others <- x$Strategy != ref - expect_equal(x$Inc_Cost_ref[others], x$Cost[others] - r0$Cost) - expect_equal(x$Inc_Effect_ref[others], x$Effect[others] - r0$Effect) -}) - -test_that("edge cases: one strategy, two strategies, names with spaces", { - one <- calculate_icers(10, 1, "Standard of care", ref_str = "Standard of care", wtp = 100) - expect_equal(one$Status_ref, "Reference") - expect_equal(one$NMB, 90) - - two <- calculate_icers(c(10, 30), c(1, 2), c("Standard of care", "New drug (A)"), - ref_str = "Standard of care") - expect_equal(row_of(two, "New drug (A)")$ICER_ref, 20) - expect_equal(row_of(two, "New drug (A)")$Status_ref, "More costly, more effective") -}) - -test_that("invalid ref_str, wtp, and nhb give informative errors", { - expect_error(toy_icers(ref_str = "Nope"), "must be one of the strategies") - expect_error(toy_icers(ref_str = c("Ref", "More")), "single strategy name") - expect_error(toy_icers(wtp = c(100, 200)), "single non-negative number") - expect_error(toy_icers(wtp = -1), "single non-negative number") - expect_error(toy_icers(nhb = TRUE), "requires `wtp`") -}) - -test_that("print.icers shows the sequential table by default and the reference views on request", { - x <- toy_icers(ref_str = "Ref", wtp = 150, currency = "€") - out_seq <- capture.output(print(x)) - expect_true(any(grepl("ICER (€/QALY)", out_seq, fixed = TRUE))) - expect_false(any(grepl("vs Ref", out_seq, fixed = TRUE))) - - out_ref <- capture.output(print(x, ref = "reference")) - expect_true(any(grepl("vs Ref", out_ref, fixed = TRUE))) - expect_false(any(grepl("Incremental Costs (€)", out_ref, fixed = TRUE))) - - out_both <- capture.output(print(x, ref = "both")) - expect_true(any(grepl("vs Ref", out_both, fixed = TRUE))) - expect_true(any(grepl("Incremental Costs (€)", out_both, fixed = TRUE))) - - expect_error(print(toy_icers(), ref = "reference"), "no reference strategy") -}) - -test_that("print(ref = \"reference\") lists the reference first, then by cost", { - data(hund_strat) - x <- calculate_icers(hund_strat$Cost, hund_strat$QALYs, hund_strat$Strategy, ref_str = "1") - expect_true(which(x$Strategy == "1") > 1) - out <- utils::capture.output(tbl <- print(x, ref = "reference")) - expect_equal(tbl$Strategy[1], "1") - v_rest <- hund_strat[hund_strat$Strategy != 1, ] - expect_equal(tbl$Strategy[-1], as.character(v_rest$Strategy[order(v_rest$Cost)])) - # the sequential and combined views keep the frontier order of the object - out <- utils::capture.output(tbl_both <- print(x, ref = "both")) - expect_equal(tbl_both$Strategy, x$Strategy) -}) diff --git a/tests/testthat/test_psa_reference.R b/tests/testthat/test_psa_reference.R deleted file mode 100644 index 047a580..0000000 --- a/tests/testthat/test_psa_reference.R +++ /dev/null @@ -1,77 +0,0 @@ -# summary.psa() and calculate_icers_psa() against a reference strategy (#209, #202) - -# Four PSA samples; the difference "New" - "Ref" falls in a different quadrant in each: -# sample inc effect inc cost quadrant -# 1 +0.2 +100 NE (more effective, more costly) -# 2 +0.2 -50 SE (more effective, less costly) -# 3 -0.1 -50 SW (less effective, less costly) -# 4 -0.1 +100 NW (less effective, more costly) -df_cost <- data.frame(Ref = c(1000, 1000, 1000, 1000), New = c(1100, 950, 950, 1100)) -df_eff <- data.frame(Ref = c(1, 1, 1, 1), New = c(1.2, 1.2, 0.9, 0.9)) -psa_q <- make_psa_obj(df_cost, df_eff, strategies = c("Ref", "New")) - -test_that("summary.psa() is unchanged without ref_str and wtp", { - expect_identical(names(summary(psa_q)), c("Strategy", "meanCost", "meanEffect")) -}) - -test_that("summary.psa(ref_str = ) gives mean differences and quadrant shares", { - s <- summary(psa_q, ref_str = "Ref") - r_new <- s[s$Strategy == "New", ] - r_ref <- s[s$Strategy == "Ref", ] - expect_equal(r_new$meanIncCost, mean(df_cost$New - df_cost$Ref)) - expect_equal(r_new$meanIncEffect, mean(df_eff$New - df_eff$Ref)) - expect_equal(unlist(r_new[, c("propNE", "propSE", "propSW", "propNW")], use.names = FALSE), - rep(0.25, 4)) - expect_true(all(is.na(r_ref[, c("meanIncCost", "meanIncEffect", "propNE", "propSE", "propSW", "propNW")]))) - expect_error(summary(psa_q, ref_str = "Nope"), "must be one of the strategies") -}) - -test_that("summary.psa(wtp = ) gives mean NMB and the share of cost-effective samples", { - s <- summary(psa_q, wtp = 1000) - expect_equal(s$meanNMB, unname(1000 * colMeans(df_eff) - colMeans(df_cost))) - expect_false(any(c("meanIncNMB", "propCostEffective") %in% names(s))) - - s_ref <- summary(psa_q, ref_str = "Ref", wtp = 1000) - inmb <- 1000 * (df_eff$New - df_eff$Ref) - (df_cost$New - df_cost$Ref) # 100, 250, -50, -200 - r_new <- s_ref[s_ref$Strategy == "New", ] - expect_equal(r_new$meanIncNMB, mean(inmb)) - expect_equal(r_new$propCostEffective, mean(inmb > 0)) - expect_equal(r_new$propCostEffective, 0.5) -}) - -data("cancer_psa_data") -psa_big <- make_psa_obj(cancer_psa_data$cost, cancer_psa_data$effectiveness, - cancer_psa_data$parameters, c("Chemotherapy", "Radiation therapy", "Surgery")) - -test_that("calculate_icers_psa() passes ref_str, wtp, and nhb to calculate_icers()", { - s <- summary(psa_big) - expected <- calculate_icers(s$meanCost, s$meanEffect, s$Strategy, - ref_str = "Surgery", wtp = 50000, nhb = TRUE, currency = psa_big$currency) - obtained <- calculate_icers_psa(psa_big, ref_str = "Surgery", wtp = 50000, nhb = TRUE) - expect_equal(obtained, expected) -}) - -test_that("calculate_icers_psa(uncertainty = TRUE) adds intervals for the reference comparison", { - x <- calculate_icers_psa(psa_big, uncertainty = TRUE, ref_str = "Surgery", wtp = 50000) - expect_identical(attr(x, "ref_str"), "Surgery") - r <- x[x$Strategy == "Chemotherapy", ] - d_cost <- cancer_psa_data$cost[[1]] - cancer_psa_data$cost[[3]] - d_eff <- cancer_psa_data$effectiveness[[1]] - cancer_psa_data$effectiveness[[3]] - expect_equal(r$Lower_95_Inc_Cost_ref, unname(quantile(d_cost, 0.025))) - expect_equal(r$Upper_95_Inc_Cost_ref, unname(quantile(d_cost, 0.975))) - expect_equal(r$Lower_95_Inc_Effect_ref, unname(quantile(d_eff, 0.025))) - expect_equal(r$Upper_95_Inc_NMB_ref, unname(quantile(50000 * d_eff - d_cost, 0.975))) - r_ref <- x[x$Strategy == "Surgery", ] - expect_true(is.na(r_ref$Lower_95_Inc_Cost_ref)) - # the frontier-based columns and their existing intervals are unchanged - x0 <- calculate_icers_psa(psa_big, uncertainty = TRUE) - expect_equal(as.data.frame(x)[, names(x0)], as.data.frame(x0)[, names(x0)]) -}) - -test_that("print.icers() keeps reference intervals out of the sequential view", { - x <- calculate_icers_psa(psa_big, uncertainty = TRUE, ref_str = "Surgery") - out_seq <- capture.output(print(x)) - expect_false(any(grepl("Inc_Cost_ref", out_seq, fixed = TRUE))) - out_ref <- capture.output(print(x, ref = "reference")) - expect_true(any(grepl("Lower_95_Inc_Cost_ref", out_ref, fixed = TRUE))) -}) diff --git a/tests/testthat/test_reference_plots.R b/tests/testthat/test_reference_plots.R index f252a6e..f0101e4 100644 --- a/tests/testthat/test_reference_plots.R +++ b/tests/testthat/test_reference_plots.R @@ -1,4 +1,4 @@ -# reference strategy and WTP lines in plot.icers() and plot.psa() (#209, #202) +# incremental plane and WTP lines in plot.icers(), plot.psa() and plot.cers_ref() (#209, #202) df_toy <- data.frame(strategy = c("Standard of care", "New drug (A)", "Cheaper", "Worse"), cost = c(100, 200, 50, 300), @@ -21,30 +21,32 @@ test_that("plot.icers() is unchanged by default", { expect_setequal(point_layer(p)$y, df_toy$cost) }) -test_that("plot.icers(ref_str = ) puts the reference at the origin", { - p <- plot(x_toy, ref_str = "Standard of care") +test_that("plot.icers(incremental = TRUE) puts the first non-dominated strategy at the origin", { + # "Cheaper" (cost 50, effect 1.5) is the least costly strategy on the frontier + p <- plot(x_toy, incremental = TRUE) d <- point_layer(p) - expect_setequal(d$x, df_toy$effect - 1) - expect_setequal(d$y, df_toy$cost - 100) + expect_setequal(d$x, df_toy$effect - 1.5) + expect_setequal(d$y, df_toy$cost - 50) expect_equal(p$labels$y, "Incremental Cost ($)") expect_equal(p$labels$x, "Incremental Effect (QALYs)") - expect_match(p$labels$caption, "Reference Strategy: Standard of care", fixed = TRUE) - expect_error(plot(x_toy, ref_str = "Nope"), "must be one of the strategies") + expect_match(p$labels$caption, "Origin: Cheaper", fixed = TRUE) + # the frontier and statuses do not change + expect_equal(table(d$shape), table(point_layer(plot(x_toy))$shape)) + expect_error(plot(x_toy, incremental = "yes"), "TRUE or FALSE") + expect_false("ref_str" %in% names(formals(plot.icers))) }) -test_that("plot.icers(wtp = ) draws the WTP line", { - p_ref <- plot(x_toy, ref_str = "Standard of care", wtp = 150) - i <- abline_layers(p_ref) +test_that("plot.icers(wtp = ) draws the WTP line through the least costly frontier strategy", { + p_abs <- plot(x_toy, wtp = 150) + i <- abline_layers(p_abs) expect_length(i, 1) - d <- ggplot2::layer_data(p_ref, i) + d <- ggplot2::layer_data(p_abs, i) expect_equal(d$slope, 150) - expect_equal(d$intercept, 0) - expect_match(p_ref$labels$caption, "WTP = $150/QALY", fixed = TRUE) - - # without a reference, the line goes through the least costly strategy on the frontier - p_abs <- plot(x_toy, wtp = 150) - d <- ggplot2::layer_data(p_abs, abline_layers(p_abs)) expect_equal(d$intercept, 50 - 150 * 1.5) + expect_match(p_abs$labels$caption, "WTP = $150/QALY", fixed = TRUE) + + p_inc <- plot(x_toy, incremental = TRUE, wtp = 150) + expect_equal(ggplot2::layer_data(p_inc, abline_layers(p_inc))$intercept, 0) }) test_that("plot.icers() uses the currency stored by calculate_icers()", { @@ -87,3 +89,37 @@ test_that("plot.psa(wtp = ) without ref_str draws the line through the least cos expect_equal(d$intercept, mean(df_cost2$A) - 2000 * mean(df_eff2$A)) expect_match(p$labels$caption, "WTP = $2,000/QALY", fixed = TRUE) }) + +# plot.cers_ref(): the star around the reference strategy +segment_layers <- function(p) { + which(vapply(p$layers, function(l) inherits(l$geom, "GeomSegment"), logical(1))) +} + +test_that("plot.cers_ref() centers the plane at the reference and draws the star", { + x_ref <- calculate_cers_ref(df_toy$cost, df_toy$effect, df_toy$strategy, + ref_str = "Standard of care", wtp = 150) + p <- plot(x_ref) + expect_s3_class(p, "ggplot") + d <- point_layer(p) + others <- df_toy$strategy != "Standard of care" + expect_setequal(d$x, df_toy$effect[others] - 1) + expect_setequal(d$y, df_toy$cost[others] - 100) + expect_length(segment_layers(p), 1) + # the stored threshold draws the WTP line through the origin; NULL removes it + d_wtp <- ggplot2::layer_data(p, abline_layers(p)) + expect_equal(d_wtp$intercept, 0) + expect_equal(d_wtp$slope, 150) + expect_length(abline_layers(plot(x_ref, wtp = NULL)), 0) + expect_match(p$labels$caption, "Reference Strategy (diamond): Standard of care", fixed = TRUE) + expect_length(segment_layers(plot(x_ref, lines = FALSE)), 0) + expect_no_warning(ggplot2::ggplot_build(plot(x_ref, col = "bw", label = "none"))) +}) + +test_that("plot.cers_ref() leaves out the segments for many strategies", { + data(hund_strat) + x_hund <- calculate_cers_ref(hund_strat$Cost, hund_strat$QALYs, hund_strat$Strategy, ref_str = "1") + p <- plot(x_hund, label = "none") + expect_length(segment_layers(p), 0) + expect_length(segment_layers(plot(x_hund, label = "none", lines = TRUE)), 1) + expect_no_warning(ggplot2::ggplot_build(p)) +}) diff --git a/vignettes/basic_cea.Rmd b/vignettes/basic_cea.Rmd index 60a03f9..748e92e 100644 --- a/vignettes/basic_cea.Rmd +++ b/vignettes/basic_cea.Rmd @@ -112,6 +112,13 @@ plot(icer_hiv, ggtitle("Cost-effectiveness of HIV screening strategies") ``` +Decisions are based on incremental costs and effects, so it can help to plot them directly. With `incremental = TRUE`, the costs and effects of every strategy are shown relative to the first non-dominated strategy, the least costly strategy on the frontier (here, the status quo), which is placed at the origin. A willingness-to-pay line (`wtp`) through the origin shows which strategies are cost-effective compared with it at that threshold. +```{r} +plot(icer_hiv, + incremental = TRUE, + wtp = 100000) +``` + *** *** @@ -176,3 +183,32 @@ plot(icer_cdiff, label = "all") # can lead to a 'busy' plot plot(icer_cdiff, plot_frontier_only = TRUE) # completely removes dominated strategies from plot plot(icer_cdiff, currency = "USD", effect_units = "quality-adjusted life-years") # customize axis labels ``` + +*** +*** + +# Comparing every strategy with a reference strategy + +When more than two strategies are compared, health technology assessment guidelines require the fully incremental (sequential) analysis that `calculate_icers()` performs: dominated and extendedly dominated strategies are removed, and each remaining strategy is compared with the next less costly one on the frontier. This is the case in the guidelines of Canada's CDA-AMC ([2017](https://www.cda-amc.ca/sites/default/files/pdf/guidelines_for_the_economic_evaluation_of_health_technologies_canada_4th_ed.pdf)), England's NICE ([2022](https://www.nice.org.uk/process/pmg36/chapter/economic-evaluation-2)), and the Netherlands' Zorginstituut Nederland ([2024](https://english.zorginstituutnederland.nl/documents/2024/01/16/guideline-for-economic-evaluations-in-healthcare)). + +Some settings also call for comparing each strategy with a single reference strategy, such as the current standard of care. NICE, for example, states that "pairwise comparisons may be presented when relevant and justified (for example, when the technology is expected to specifically displace individual comparators)". These comparisons complement the fully incremental analysis; they do not replace it. + +`calculate_cers_ref()` compares every strategy with a reference strategy given in `ref_str`. For each strategy, it gives the difference in cost and effect from the reference, the cost-effectiveness ratio (CER) of that difference, and its status relative to the reference (dominant, dominated, or in one of the two quadrants where a ratio is meaningful). The reference is listed first, followed by the other strategies in order of increasing cost. With `wtp`, the net monetary benefit and its difference from the reference are added. Using the HIV screening example, with the status quo as the reference: +```{r} +cers_hiv <- calculate_cers_ref(cost = v_hiv_costs, + effect = v_hiv_qalys, + strategies = v_hiv_strat_names, + ref_str = "status quo", + wtp = 100000) +cers_hiv +``` + +Compared with the status quo, every screening strategy is more costly and more effective, so each has a CER. Annual screening has a CER below $100,000 per QALY and so looks cost-effective compared with the status quo, yet the fully incremental analysis above shows that it is strongly dominated by screening every 3 years, which costs less and gives more QALYs. This is why the comparison with a reference does not replace the fully incremental analysis. + +The `plot()` method draws the incremental cost-effectiveness plane centered at the reference strategy, with a segment from the reference to each strategy. The willingness-to-pay line stored in the object goes through the origin: strategies below it are cost-effective compared with the reference at that threshold. +```{r} +plot(cers_hiv) +``` + +With many strategies, the segments are left out by default; `lines = TRUE` or `lines = FALSE` overrides this. For a PSA, `calculate_cers_ref_psa()` makes the same comparison from the mean costs and effects, and with `uncertainty = TRUE` adds 95% intervals for the differences and the proportion of PSA samples in each quadrant of the incremental plane. `plot.psa()` with `ref_str` shows the PSA samples themselves on the incremental plane. + From 918da90bcd073f39260b6e395c2e402a4089d99c Mon Sep 17 00:00:00 2001 From: Fernando Alarid-Escudero Date: Wed, 18 Nov 2026 12:39:08 -0600 Subject: [PATCH 12/12] Move check_ref_str() to cers_ref.R (review of #261) Co-Authored-By: Claude --- R/cers_ref.R | 21 +++++++++++++++++++++ R/icers.R | 21 --------------------- man/check_ref_str.Rd | 2 +- 3 files changed, 22 insertions(+), 22 deletions(-) diff --git a/R/cers_ref.R b/R/cers_ref.R index a734571..c26d5e0 100644 --- a/R/cers_ref.R +++ b/R/cers_ref.R @@ -266,6 +266,27 @@ add_cers_ref_uncertainty <- function(df_ref, psa, wtp, nhb) { df_ref } +#' Check a reference strategy name +#' +#' @param ref_str reference strategy name, or \code{NULL}. +#' @param strategies names of the strategies in the analysis. +#' @return \code{ref_str} as a character string, or \code{NULL}. +#' @keywords internal +check_ref_str <- function(ref_str, strategies) { + if (is.null(ref_str)) { + return(NULL) + } + if (length(ref_str) != 1 || is.na(ref_str)) { + stop("`ref_str` must be a single strategy name.", call. = FALSE) + } + ref_str <- as.character(ref_str) + if (!ref_str %in% strategies) { + stop("`ref_str` must be one of the strategies: ", + paste0("\"", strategies, "\"", collapse = ", "), ".", call. = FALSE) + } + ref_str +} + #' Status of each strategy relative to a reference strategy #' #' @param v_inc_cost,v_inc_effect cost and effect of each strategy minus those of the reference. diff --git a/R/icers.R b/R/icers.R index 36392c5..1b849dc 100644 --- a/R/icers.R +++ b/R/icers.R @@ -218,27 +218,6 @@ wtp_caption <- function(wtp, currency, effect_units = "QALYs") { "/", unit) } -#' Check a reference strategy name -#' -#' @param ref_str reference strategy name, or \code{NULL}. -#' @param strategies names of the strategies in the analysis. -#' @return \code{ref_str} as a character string, or \code{NULL}. -#' @keywords internal -check_ref_str <- function(ref_str, strategies) { - if (is.null(ref_str)) { - return(NULL) - } - if (length(ref_str) != 1 || is.na(ref_str)) { - stop("`ref_str` must be a single strategy name.", call. = FALSE) - } - ref_str <- as.character(ref_str) - if (!ref_str %in% strategies) { - stop("`ref_str` must be one of the strategies: ", - paste0("\"", strategies, "\"", collapse = ", "), ".", call. = FALSE) - } - ref_str -} - #' Check the willingness-to-pay and NHB arguments of calculate_icers() #' #' @inheritParams calculate_icers diff --git a/man/check_ref_str.Rd b/man/check_ref_str.Rd index 38e818a..d888f0c 100644 --- a/man/check_ref_str.Rd +++ b/man/check_ref_str.Rd @@ -1,5 +1,5 @@ % Generated by roxygen2: do not edit by hand -% Please edit documentation in R/icers.R +% Please edit documentation in R/cers_ref.R \name{check_ref_str} \alias{check_ref_str} \title{Check a reference strategy name}