Bayesian Meta-Analysis of Health State Utility Values: A Tutorial with a Practical Application in Heart Failure.

Santos, Joseph Alvin Ramos; Grant, Robert; Di Tanna, Gian Luca (2024). Bayesian Meta-Analysis of Health State Utility Values: A Tutorial with a Practical Application in Heart Failure. PharmacoEconomics, 42(7), pp. 721-735. Springer 10.1007/s40273-024-01387-7

[img]
Preview
Text
s40273-024-01387-7.pdf - Published Version
Available under License Creative Commons: Attribution-Noncommercial (CC-BY-NC).

Download (1MB) | Preview

Researchers incorporate health state utility values as inputs to inform economic models. However, for a particular health state or condition, multiple utility values derived from different studies typically exist and a single study is often insufficient to represent the best available source of utility needed to inform policy decisions. The purpose of this paper is to provide an introductory guidance for conducting Bayesian meta-analysis of health state utility values to generate a single parameter input for economic evaluation, using R. The tutorial is illustrated using data from a systematic review of health state utilities of patients with heart failure, with 21 studies that reported utilities measured using the EuroQol-5D (EQ-5D). Explanations, key considerations and suggested readings are provided for each step of the tutorial, adhering to a clear workflow for conducting Bayesian meta-analysis: (1) setting-up the data; (2) employing methods to impute missing standard deviations; (3) defining the priors; (4) fitting the model; (5) diagnosing model convergence; (6) interpreting the results; and (7) performing sensitivity analyses. The posterior distributions for the pooled effect size (i.e. mean health state utility) and between-study heterogeneity are discussed and interpreted in light of the data, priors and models used. We hope that this tutorial will foster interest in Bayesian methods and their applications in the meta-analysis of utilities.

Item Type:

Journal Article (Original Article)

Division/Institute:

04 Faculty of Medicine > Pre-clinic Human Medicine > Department of Clinical Research (DCR)

UniBE Contributor:

Di Tanna, Gian Luca

Subjects:

600 Technology > 610 Medicine & health

ISSN:

1179-2027

Publisher:

Springer

Language:

English

Submitter:

Pubmed Import

Date Deposited:

22 May 2024 10:24

Last Modified:

16 Jun 2024 00:19

Publisher DOI:

10.1007/s40273-024-01387-7

PubMed ID:

38767714

BORIS DOI:

10.48350/196924

URI:

https://boris.unibe.ch/id/eprint/196924

Actions (login required)

Edit item Edit item
Provide Feedback