survHE
survHE is an R package that helps researchers estimate survival times in health economic evaluations using statistical models. It makes it easier to analyze data from clinical trials and use the results in cost-effectiveness models, replacing less reliable methods like spreadsheets.
At a glance
Use when
Estimating long-term survival from clinical trial data for use in cost-effectiveness models, especially when extrapolation beyond observed data is needed.
Avoid when
Working with non-parametric survival estimates only, or when users lack access to or familiarity with R programming.
Inputs
Survival data from clinical trials (e.g., time-to-event data, censoring indicators), optionally with covariates; model specifications for parametric survival models.
Outputs
Fitted survival models, estimated mean survival times, survival curves, uncertainty quantification (e.g., credible/confidence intervals), model comparison metrics (e.g., AIC, DIC, WAIC).
How it works
survHE is an R package designed for survival analysis in health economic evaluation and cost-effectiveness modeling. It integrates frequentist methods via flexsurv and Bayesian approaches using rstan (Hamiltonian Monte Carlo) or INLA (integrated nested Laplace approximation). The package supports pre-compilation of multiple parametric survival models, enabling flexible and efficient model fitting. It streamlines the transition from trial data to economic modeling by automating complex statistical workflows.
- HTA domains
- Clinical Effectiveness, Costs & Economic Evaluation, Safety
- Assumptions
- Survival times follow a parametric distribution; censoring is non-informative; model specification is appropriate for the data.
- Strengths
- Supports both frequentist and Bayesian frameworks,Pre-compiles a wide range of parametric models,Integrates seamlessly with R for reproducible research,Reduces reliance on error-prone spreadsheet tools,Facilitates direct use of results in economic models
- Limitations
- Requires familiarity with R and statistical modeling,Bayesian methods may require long computation times with rstan,INLA availability depends on model structure,Assumes correct specification of parametric forms
- Also known as
- survHE, Survival Analysis for Health Economic Evaluation
Questions this answers
- › What is the mean survival time based on limited trial data?
- › Which parametric survival model best fits the observed survival data?
- › How can survival estimates be integrated into cost-effectiveness models?
- › What are the uncertainty intervals around survival predictions?
- › How can Bayesian and frequentist survival analyses be performed consistently?
- › How can modelers avoid error-prone spreadsheet-based workflows?
References & sources
Similar by meaning
Beta record. Generated from the primary source via AI extraction and independent audit, pending final human review.