From: INES: Interactive tool for construction and extrapolation of partitioned survival models
Parameter | Description | Usual source | Univariate sensitivity analysis |
---|---|---|---|
Probabilities of survival in PFS and OS in intervention and control | Spreadsheet table of probabilities at monthly time points | Numerical data from published Kaplan–Meier curves. These can be extracted using the WebPlotDigitizer tool | No |
Number at risk | Spreadsheet table of number at risk during the follow up of the clinical study | Usually available alongside published Kaplan–Meier curves | No |
Number of events (optional) | The number of events for each treatment group and outcome (deaths, and progression-or-death). This can be left null | Usually available alongside published Kaplan–Meier curves. This data can increase the precision of the simulation of the IPD, but is not essential | No |
Hazard ratio | The hazard ratio for PFS and OS comparing the intervention with the control | Clinical trial reports or literature. The user has the option of either using published values obtained from a clinical trial report or meta-analysis, or calculating the HR based on the data provided to the model | Yes |
Discount rate | A number, usually between 0 and 6%. It must be expressed as the annual rate. The INES tool will automatically convert to monthly equivalent | Official guidelines for benefit–cost analysis in each country | Yes |
Unit costs | The cost or price per unit of each resource included in the costs | Official sources or literature | Yes |
Resource use during the PFS state | The number of units consumed for each of the resources included in the costs, in each treatment group, during PFS. Consumption can vary over time. The user can express the consumption in units per day, week or month | Clinical trial reports or literature | No |
Resource use during the post progression state | The number of units consumed for each of the resources after progression (by initial treatment group). Consumption cannot vary over time. The user can express the consumption in units per day, week or month | Clinical trial reports or literature | No |
Utility weights during the PFS state | Utility weights in each treatment group during the progression-free state, Weights cannot vary over time. The model will calculate life years if all the utility weights are set to 1 | Clinical trial reports or literature | Yes |
Utility weights during the post progression state | Utility weight during post-progression. Weights cannot vary over time or vary by initial treatment. The model will calculate life years if all the utility weights are set to 1 | Clinical trial reports or literature | Yes |