Probabilistic sensitivity analysis
Set up, run, and review this analysis in Lifescio.
Supported analysis
Each replication draws transition rows from Dirichlet distributions and positive costs from gamma distributions, then evaluates the cohort model. Higher Dirichlet concentration means less transition uncertainty. Cost coefficient of variation controls relative cost uncertainty.
Supported scope: Utilities are fixed. Rows, costs and strategy parameters are sampled independently. Zero transition probabilities remain structural zeros. This does not capture correlated evidence, structural uncertainty or arbitrary user-specified distributions.
Prepare data and configure the analysis
First specify both state-transition strategies, annual rewards, initial proportions, cycle timing, and horizon. Transition Dirichlet concentration controls uncertainty around each transition row: larger values concentrate draws more tightly around the entered probabilities. A zero cell stays zero; the tool does not add unobserved transitions automatically.
Cost coefficient of variation is cost standard deviation divided by mean and controls gamma sampling. Simulation replications counts parameter draws. Utilities are fixed. Transition rows, strategy parameters, and state costs are sampled independently across their respective groups; shared evidence is not automatically correlated. Keep these limitations with the run.
Input reference (14 fields)
Enter your data and prespecified choices. Ranges include both limits. For table fields, use cell editing, Paste data, or Import CSV / TSV. Review whether the first imported row contains headings and map the source columns before applying the import. Empty numeric cells are not treated as zero. Open file is for an exported Lifescio analysis JSON file, not a spreadsheet.
| Input | Format | Allowed range | Guidance |
|---|---|---|---|
| Health states | Text | — | Names correspond to transition rows and columns. Add one name or comma-separated names in the State manager. It creates matching matrices for both strategies. |
| Comparator transitions | Table | — | Each row must sum to 1. Probabilities apply to one model cycle. Rows and columns follow the order in the State manager. Fill every cell, including zeros, before saving or running. |
| Intervention transitions | Table | — | Use the same state order as the comparator. Rows and columns follow the order in the State manager. Fill every cell, including zeros, before saving or running. |
| Annual state rewards | Table | — | Costs per year and utilities; the engine applies cycle length. Columns in order: Comparator cost; Comparator utility; Intervention cost; Intervention utility. |
| Initial state proportions | Text | — | One proportion per state, summing to 1. |
| Number of cycles | Whole number | 1–500 | — |
| Cycle length (years) | Number | 0.01–10 | — |
| Annual cost discount rate | Number | 0–0.2 | — |
| Annual QALY discount rate | Number | 0–0.2 | — |
| State reward timing | Selection | — | Choices: trapezoid, start. |
| Willingness to pay per QALY | Number | 0–10000000 | — |
| Simulation replications | Whole number | 100–20000 | — |
| Transition Dirichlet concentration | Number | 1–100000 | — |
| Cost coefficient of variation | Number | 0.001–2 | — |
Run and review
- Give the analysis a descriptive name and review every example value. Record units, data preparation, and planning assumptions in Reproducibility & notes. For simulations, record the seed and replication or sampling settings.
- Choose Run analysis. Resolve input errors before interpreting output. In the connected app, a run uses a saved input revision; local preview runs need the page to stay open.
- Review Diagnostics for method limitations, warnings, and available numerical checks. A completed run means computation finished; it does not establish that the assumptions fit your study.
- Use the Result table selector to inspect each output. Editing inputs does not recalculate existing results: an earlier-input notice means you need to run again to obtain matching results.
Understand the outputs
Strategy outcomes, Incremental comparison, and occupancy tables describe the fixed-input base case. They are not averages over the probabilistic draws. For uncertainty results, select PSA draws or Acceptability curve.
The cost-effectiveness plane plots incremental health against incremental cost. The acceptability table includes a Monte Carlo SE for each displayed threshold. The curve spans zero to twice the entered willingness-to-pay threshold, with a minimum upper range of one. Distinguish its numerical simulation error from the parameter uncertainty represented by the spread of points.
Troubleshooting and common mistakes
If the run exceeds its operation budget, reduce draws, cycles, or states thoughtfully; the budget depends on all three. Increasing concentration to speed convergence changes the uncertainty assumptions and is not a computational fix.
If uncertainty looks unrealistically narrow, inspect fixed utilities, structural zeros, concentration, and cost variation. If the decision changes across seeds, increase replications and review Monte Carlo error. Correlated evidence or different uncertainty distributions require a model beyond the currently supported setup.
After correcting inputs, run again and check that the results belong to the new input snapshot. Increasing simulation draws improves numerical precision only; it does not repair invalid data or an unsuitable model.
Save, compare, and export
In the connected app, Save analysis retains the inputs in your account. Run history lets you inspect earlier runs, each with the input revision it used. In local preview, Save analysis downloads an input file instead; download results before leaving the page.
Use Result + inputs for a JSON archive containing the completed result and its inputs, Download CSV for the selected result table, PDF report for a report, and a chart's SVG button for its figure. Table displays round to three decimals; CSV and JSON exports retain calculation precision. Use those values when assessing a result near a decision boundary.
To reopen an archive, select the matching analysis method and choose Open file. Imported inputs start a new draft; a result archive can also display its retained result. It does not attach itself to an existing account analysis or recompute the result. Save or run the imported draft when you want a new account record.