Modelling survival data to account for model uncertainty: a single model or model averaging?

Thamrin, Sri Astuti; McGree, James; Mengersen, Kerrie

doi:10.1186/2193-1801-2-665

Cited by 7 publications

(11 citation statements)

References 52 publications

(87 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Further research is required to understand the conditions under which evidence are rich enough to justify a more complex model. As an alternative to model selection, model averaging could be performed [ 25 ]. The case-study demonstrated that extrapolations were poor when the available follow-up did not include all the turning-points in the hazard function.…”

Section: Discussionmentioning

confidence: 99%

Comparing current and emerging practice models for the extrapolation of survival data: a simulation study and case-study

Kearns

Stevenson

Triantafyllopoulos

et al. 2021

BMC Med Res Methodol

View full text Add to dashboard Cite

Background Estimates of future survival can be a key evidence source when deciding if a medical treatment should be funded. Current practice is to use standard parametric models for generating extrapolations. Several emerging, more flexible, survival models are available which can provide improved within-sample fit. This study aimed to assess if these emerging practice models also provided improved extrapolations. Methods Both a simulation study and a case-study were used to assess the goodness of fit of five classes of survival model. These were: current practice models, Royston Parmar models (RPMs), Fractional polynomials (FPs), Generalised additive models (GAMs), and Dynamic survival models (DSMs). The simulation study used a mixture-Weibull model as the data-generating mechanism with varying lengths of follow-up and sample sizes. The case-study was long-term follow-up of a prostate cancer trial. For both studies, models were fit to an early data-cut of the data, and extrapolations compared to the known long-term follow-up. Results The emerging practice models provided better within-sample fit than current practice models. For data-rich simulation scenarios (large sample sizes or long follow-up), the GAMs and DSMs provided improved extrapolations compared with current practice. Extrapolations from FPs were always very poor whilst those from RPMs were similar to current practice. With short follow-up all the models struggled to provide useful extrapolations. In the case-study all the models provided very similar estimates, but extrapolations were all poor as no model was able to capture a turning-point during the extrapolated period. Conclusions Good within-sample fit does not guarantee good extrapolation performance. Both GAMs and DSMs may be considered as candidate extrapolation models in addition to current practice. Further research into when these flexible models are most useful, and the role of external evidence to improve extrapolations is required.

show abstract

Section: Discussionmentioning

confidence: 99%

Comparing current and emerging practice models for the extrapolation of survival data: a simulation study and case-study

Kearns

Stevenson

Triantafyllopoulos

et al. 2021

BMC Med Res Methodol

View full text Add to dashboard Cite

show abstract

“…This choice would consider the specifics of the extrapolation problem, such as the plausibility of extrapolations, the richness of the available data, and the qualitative differences in extrapolations arising from different models. As an alternative to model selection, model averaging could be performed (23). The casestudy demonstrated that extrapolations were poor when the available follow-up did not include all the turning-points in the hazard function.…”

Section: Discussionmentioning

confidence: 99%

Comparing Current and Emerging Practice Models For the Extrapolation of Survival Data: A Simulation Study and Case-Study

Kearns

Stevenson

Triantafyllopoulos

et al. 2021

Preprint

View full text Add to dashboard Cite

BackgroundEstimates of future survival can be a key evidence source when deciding if a medical treatment should be funded. Current practice is to use standard parametric models for generating extrapolations. Several emerging, more flexible, survival models are available which can provide improved within-sample fit. This study aimed to assess if these emerging practice models also provided improved extrapolations.MethodsBoth a simulation study and a case-study were used to assess the goodness of fit of five classes of survival model. These were: current practice models, Royston Parmar models (RPMs), Fractional polynomials (FPs), Generalised additive models (GAMs), and Dynamic survival models (DSMs). The simulation study used a mixture-Weibull model as the data-generating mechanism with varying lengths of follow-up and sample sizes. The case-study was long-term follow-up of a prostate cancer trial. For both studies, models were fit to an early data-cut of the data, and extrapolations compared to the known long-term follow-up.ResultsThe emerging practice models provided better within-sample fit than current practice models. For data-rich simulation scenarios (large sample sizes or long follow-up), the GAMs and DSMs provided improved extrapolations compared with current practice. Extrapolations from FPs were always very poor whilst those from RPMs were similar to current practice. With short follow-up all the models struggled to provide useful extrapolations. In the case-study all the models provided very similar estimates, but extrapolations were all poor as no model was able to capture a turning-point during the extrapolated period. ConclusionsGood within-sample fit does not guarantee good extrapolation performance. Both GAMs and DSMs may be considered as candidate extrapolation models in addition to current practice. Further research into when these flexible models are most useful, and the role of external evidence to improve extrapolations is required.

show abstract

“…These measures provide information about the goodness of fit between observed data and model estimates across the available trial follow-up but give no indication about its suitability for extrapolation. 4,6 Besides, when the sample size is sufficiently large, one of the alternative models usually emerges with the highest probability given the data, 17 and the BMA1 would be reduced to an analysis of the “best” model. Thus, one model can easily dominate the others on the basis of criteria that may have little value in extrapolated space.…”

Section: Methodsmentioning

confidence: 99%

“…11,12 BMA has been previously proposed in cost-effectiveness models with cross-sectional data 13–15 or for covariate selection in a Markov model. 16 Thamrin and others 17 suggested that a BMA approach would be appropriate for predicting survival when the sample size was reduced.…”

mentioning

confidence: 99%

Bayesian Solutions for Handling Uncertainty in Survival Extrapolation

2016

View full text Add to dashboard Cite

show abstract

Modelling survival data to account for model uncertainty: a single model or model averaging?

Cited by 7 publications

References 52 publications

Comparing current and emerging practice models for the extrapolation of survival data: a simulation study and case-study

Comparing current and emerging practice models for the extrapolation of survival data: a simulation study and case-study

Comparing Current and Emerging Practice Models For the Extrapolation of Survival Data: A Simulation Study and Case-Study

Bayesian Solutions for Handling Uncertainty in Survival Extrapolation

Contact Info

Product

Resources

About