Extrapolating Parametric Survival Models in Health Technology Assessment: A Simulation Study

Abstract

Extrapolations of parametric survival models fitted to censored data are routinely used in the assessment of health technologies to estimate mean survival, particularly in diseases that potentially reduce the life expectancy of patients. Akaike’s information criterion (AIC) and Bayesian information criterion (BIC) are commonly used in health technology assessment alongside an assessment of plausibility to determine which statistical model best fits the data and should be used for prediction of long-term treatment effects. We compare fit and estimates of restricted mean survival time (RMST) from 8 parametric models and contrast models preferred in terms of AIC, BIC, and log-likelihood, without considering model plausibility. We assess the methods’ suitability for selecting a parametric model through simulation of data replicating the follow-up of intervention arms for various time-to-event outcomes from 4 clinical trials. Follow-up was replicated through the consideration of recruitment duration and minimum and maximum follow-up times. Ten thousand simulations of each scenario were performed. We demonstrate that the different methods can result in disagreement over the best model and that it is inappropriate to base model selection solely on goodness-of-fit statistics without consideration of hazard behavior and plausibility of extrapolations. We show that typical trial follow-up can be unsuitable for extrapolation, resulting in unreliable estimation of multiple parameter models, and infer that selecting survival models based only on goodness-of-fit statistics is unsuitable due to the high level of uncertainty in a cost-effectiveness analysis. This article demonstrates the potential problems of overreliance on goodness-of-fit statistics when selecting a model for extrapolation. When follow-up is more mature, BIC appears superior to the other selection methods, selecting models with the most accurate and least biased estimates of RMST.

Biased Survival Predictions When Appraising Health Technologies in Heterogeneous Populations

“…We considered three distinct approaches to obtaining a LY estimate. First, we chose the single best-fitting models for each arm independently according to Akaike information criterion (AIC) and Bayesian information criterion (BIC), despite there being limitations with this approach [4]. This means different parametric models could be chosen for each arm, which is only encouraged by the NICE technical support document (TSD)-14 when justified by "clinical expert judgement, biological plausibility, and robust statistical analysis" [5].…”

“…Fig 4. Violin plot of difference in life-year estimation for each method for variations of scenario 2.…”

Using the Angle Index to Assess COVID-19 Damage at the Early Stage and Throughout 2020 in Comparison of Mathematical Models

“…Many researchers [10][11][12][13][14][15][16][17][18][19] have proposed the use of mathematical models to predict the number of COVID-19 cases, while others have investigated the IP days during the COVID-19 pandemic [25][26][27][28][29]. None of these studies, however, used the IP days to compare the ImpactCOVID, nor did they apply the angle index to inspect the ImpactCOVID in countries/regions.…”

“…Given that using less constrained parameters makes the model a better fit for the data [29,30], one constrained term set at the middle point (i.e., P(x2, y2)) of the observations in the QE model [19] yields lower MA accordingly. In contrast, the unconstrained QE model in this study (i.e., using the Solver add-in tool in Microsoft Excel) can yield higher MA than the constrained one.…”

Using the Absolute Advantage Coefficient(AAC) to Measure the Strength of Damage Hit by COVID-19 in India on a Growth-Share Matrix

“…As with all forms of Web-based technology, advances in health communication technology are occurring every moment [47]. The real-time mobile online dashboard for COVID-19 provided to the public is the fourth feature.…”