Skip to content

README: the acceptability frontier and expected loss curves (Alarid-Escudero et al., 2019) - #257

Merged
feralaes merged 1 commit into
mainfrom
readme-elc
Nov 14, 2026
Merged

feralaes merged 1 commit into
mainfrom
readme-elc

Conversation

@feralaes

Copy link
Copy Markdown
Collaborator

Base: built on #256 (the log-axis fix for plot.exp_loss()), so the expected loss figure is rendered with readable axes. When #256 merges, GitHub retargets this PR to main.

Fernando asked for the README to feature the expected loss curves paper:

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

Changes

  1. What it does: the PSA bullet now names CEACs together with the CEAF, and expected loss curves (ELCs) as showing the expected foregone benefit, the optimal strategy and the EVPI in one figure, citing the paper.
  2. Example, why the CEAF is needed: a short paragraph under the CEAC figure. Its numbers are computed from cancer_psa_data inline, so they can't drift from the code. At $50,000/QALY, Surgery has the highest probability of being cost-effective (0.369), but Chemo has the highest expected net benefit (0.356) and is on the frontier.
  3. New subsection, "Expected loss curves":
    • what an ELC shows: how likely a strategy is to be the wrong choice, combined with how much is lost when it is;
    • why it's useful: optimal strategy, switch points and EVPI in one figure;
    • the paper's recommendation to report ELCs alongside CEACs, the CEAF and the EVPI;
    • the figure, plot(calc_exp_loss(l_psa, wtp = v_wtp), n_x_ticks = 10);
    • one sentence on what the figure shows: at $50,000/QALY, Chemo's expected loss is $14,954 per patient, which equals the EVPI. A hidden check stops the README from rendering if those two values ever disagree.
  4. Citation: the full reference is listed below citation("dampack").

The second figure is 46 KB. R CMD check --as-cran still reports the installed size as OK, with only the development-version NOTE. The DOI link returns HTTP 200.

…019 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.
Base automatically changed from elc-log-breaks to main November 14, 2026 03:27
@feralaes
feralaes merged commit cc2bcca into main Nov 14, 2026
13 checks passed
@feralaes
feralaes deleted the readme-elc branch November 14, 2026 03:34
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant