Search

Background Extrapolation of time-to-event data can be a critical component of cost-effectiveness analysis. Objectives To contrast the value of external data on treatment effects as a selection aid in model fitting to the clinical data or for the direct extrapolation of survival. Methods We assume the existence of external summary data on both treatment and control and consider two scenarios: availability of external individual patient data (IPD) on the control only and an absence of external IPD. We describe how the summary data can be used to extrapolate survival or to assess the plausibility of extrapolations of the clinical data. We assess the merit of either approach using a comparison of cemented and cementless total hip replacement as a case study. Merit is judged by comparing incremental net benefit (INB) obtained in scenarios with incomplete IPD with that derived from modeling external IPD on both treatment and control. Results Measures of fit with the external summary data did not identify survival model specifications that best estimated INB. Addition of external IPD for the control only did not improve estimates of INB. Extrapolation of survival using the external summary data comparing treatment and control improved estimates of INB. Conclusions Our case study indicates that summary data comparing treatment and control are more valuable than IPD limited to the control when extrapolating event rates for cost-effectiveness analysis. These data are best exploited in direct extrapolation of event rates rather than as an aid to select extrapolations on the basis of the clinical data.

Objectives This article proposes an integrated approach to the development, validation, and evaluation of new risk prediction models illustrated with the Fungal Infection Risk Evaluation study, which developed risk models to identify non-neutropenic, critically ill adult patients at high risk of invasive fungal disease (IFD). Methods Our decision-analytical model compared alternative strategies for preventing IFD at up to three clinical decision time points (critical care admission, after 24 hours, and end of day 3), followed with antifungal prophylaxis for those judged "high" risk versus "no formal risk assessment." We developed prognostic models to predict the risk of IFD before critical care unit discharge, with data from 35,455 admissions to 70 UK adult, critical care units, and validated the models externally. The decision model was populated with positive predictive values and negative predictive values from the best-fitting risk models. We projected lifetime cost-effectiveness and expected value of partial perfect information for groups of parameters. Results The risk prediction models performed well in internal and external validation. Risk assessment and prophylaxis at the end of day 3 was the most cost-effective strategy at the 2% and 1% risk threshold. Risk assessment at each time point was the most cost-effective strategy at a 0.5% risk threshold. Expected values of partial perfect information were high for positive predictive values or negative predictive values (£11 million–£13 million) and quality-adjusted life-years (£11 million). Conclusions It is cost-effective to formally assess the risk of IFD for non-neutropenic, critically ill adult patients. This integrated approach to developing and evaluating risk models is useful for informing clinical practice and future research investment.