Reproducibility of published model-based cancer drug cost-effectiveness analyses: a study protocol for a cross-sectional analysis.

Abstract

<h4>Introduction</h4>Model-based cost-effectiveness analysis (CEA) of pharmaceuticals informs reimbursement and pricing in many healthcare systems, and it is essential that CEA evidence is valid and reliable. Several studies have reported lacking transparency in CEA studies. In this study protocol, we describe a study that will investigate whether model-based CEA studies of cancer drugs are transparent and informative enough to enable the reproduction of study findings.<h4>Methods and analysis</h4>This study protocol outlines a study where we will identify CEA studies indexed in MEDLINE from 2015 to 2023 based on predefined search terms. We will include English-language CEA studies evaluating pharmaceutical treatments based on decision-analytical modelling methods that report cost-effectiveness results using life-years, quality-adjusted life-years and/or disability-adjusted life-years as health outcome metric(s). Two authors will screen abstracts and full text for inclusion. We anticipate that a maximum of 150 studies will be included after a full-text review. A data extraction template is designed to capture information used to determine reproducibility together with other information that will be analysed as potential determinants of reproducibility in logistic and linear regression analyses.<h4>Ethics and dissemination</h4>This study design has been deemed exempt from ethical approval. All collected data will be made available in an online repository that will host the study protocol and other supplementary data. Results from this proposed study will be published in peer-reviewed journals and at scientific conferences and workshops.