English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Decision makers require predictions of costeffectiveness over a time horizon that is sufficient to capture material differences in costs and health outcomes across relevant comparators. The ideal Randomized Controlled Trial (RCT) for costeffectiveness analysis (CEA) would: include all relevant comparators, produce unbiased precise estimates when analyzed as randomized, and measure all endpoints relevant to the decision over a sufficient time horizon. Such trials are rare. To fill this gap, decision analytical models are developed to synthesize the available evidence in an attempt to recreate an ‘‘ideal trial’’. Decision models that include ‘‘time to event’’ endpoints typically need to extrapolate beyond the observed data to account for censoring. The method(s) chosen for extrapolation should minimize bias and maximize precision. However, in practice, it is challenging to assess and balance these criteria. It has been recognized that we are unlikely to obtain the best estimates simply by closer scrutiny of data from individual trials. Rather, we need to select methods that make use of the available evidence in its entirety, including trial and observational data and expert opinion. The methods chosen should also be transparent and facilitate sensitivity analysis. This allows decision makers to judge whether additional evidence is required, and to substitute their own judgements regarding uncertain parameters. This laudable goal raises important research questions regarding the appropriate selection of methods for synthesis and temporal extrapolation. The 8 papers published in this special issue highlight recent progress in methods for extrapolation in CEA. Several of these papers discuss different aspects of evidence synthesis. Guyot and others harness Bayesian multi-parameter evidence synthesis to combine evidence from 4 sources: an RCT, general population database, cancer registry and expert opinion. Negrı́n and others argue in favor of Bayesian model averaging as a principled approach for weighting alternative parametric extrapolations. The methodological review by Jackson and others defines the assumptions made by approaches for incorporating external data. The authors also promote a future research agenda for developing methods to formally eliciting expert opinion for extrapolation methods. Hoogenveen and others present a mathematical modelling approach for predicting the effects of interventions on mortality across multiple disease areas. Lousdal and others advocate an approach for estimating mean survival times when only published estimates of median survival are available. Meacock and others consider some of the issues that may arise when attempting to extrapolate survival in the broader setting of health policy evaluations. The papers by Williams and others discuss the relative merits of extrapolation models with alternative structures, namely partitioned survival models (PSMs) and state transition models (STMs); for example multi-state and Markov models. In the case study by Williams and others, the estimates of the Incremental Cost-Effectiveness Ratios (ICERs) range from £13,000 to £29,000 per QALY, according to precisely which method is chosen. Clearly, the choice of which model structure Received 19 December 2016 from Institute of Health and Wellbeing, University of Glasgow, Glasgow, UK (NH), and Department of Health Services Research and Policy, London School of Hygiene and Tropical Medicine, London, UK (RG). Conflicts of interest: none. Revision accepted for publication 19 December 2016.

Extrapolation of Survival Data in Cost-effectiveness Analyses: The Need for Causal Clarity