Partitioned Survival and State Transition Models for Healthcare Decision Making in Oncology: Where Are We Now?

•Partitioned survival models (PSMs) are routinely used to inform reimbursement decisions for oncology drugs.•PSMs differ from state transition models, as they do not include a structural link between intermediate clinical endpoints (eg, disease progression) and survival. Survival extrapolations have been shown to differ markedly between the 2 modeling approaches. Although PSMs have some practical advantages, their structure limits the extent to which sensitivity analyses can be used to explore clinical uncertainties in the extrapolation period.•Decision makers should more carefully scrutinize survival predictions from PSMs and other modeling approaches to assess whether model predictions are credible. This should include an assessment of life-years accrued in individual health states, and should take account of trial data on individual clinical events, external data, and expert opinion. Owing to the challenges in robustly estimating transition probabilities, for the time being we recommend state transition models are used alongside PSMs to support the assessment of clinical uncertainties in the extrapolation period, for example, by allowing exploration of different survival outcomes following disease progression.