David E. Kloecker, MPhil; Melanie J. Davies, MD; Kamlesh Khunti, MD, PhD; Francesco Zaccardi, MD, PhD
Disclaimer: The views expressed in this publication are those of the authors and not necessarily those of the National Health Service, National Institute for Health Research (NIHR), or Department of Health.
Financial Support: From NIHR Applied Research Collaborations East Midlands and NIHR Leicester Biomedical Research Centre. The funding bodies had no role in the design, conduct, or analysis of the study or in the decision to submit the manuscript for publication.
Disclosures: Disclosures can be viewed at www.acponline.org/authors/icmje/ConflictOfInterestForms.do?msNum=M19-3286.
Corresponding Author: David E. Kloecker, MPhil, Leicester Real World Evidence Unit, Leicester Diabetes Centre, Leicester General Hospital, Gwendolen Road, Leicester LE5 4PW, United Kingdom; e-mail, email@example.com or firstname.lastname@example.org.
Current Author Addresses: Mr. Kloecker: Leicester Real World Evidence Unit, Leicester Diabetes Centre, Leicester General Hospital, Gwendolen Road, Leicester LE5 4PW, United Kingdom.
Drs. Davies, Khunti, and Zaccardi: Leicester Diabetes Centre, Leicester General Hospital, Gwendolen Road, Leicester LE5 4PW, Leicestershire, United Kingdom.
Author Contributions: Conception and design: F. Zaccardi.
Analysis and interpretation of the data: D.E. Kloecker, F. Zaccardi.
Drafting of the article: D.E. Kloecker, F. Zaccardi.
Critical revision for important intellectual content: D.E. Kloecker, M.J. Davies, K. Khunti, F. Zaccardi.
Final approval of the article: D.E. Kloecker, M.J. Davies, K. Khunti, F. Zaccardi.
Provision of study materials or patients: D.E. Kloecker.
Statistical expertise: D.E. Kloecker, F. Zaccardi.
Administrative, technical, or logistic support: D.E. Kloecker.
Collection and assembly of data: D.E. Kloecker.
The restricted mean survival time (RMST) has been advocated as an alternative or a supplement to the hazard ratio for reporting the effect of an intervention in a randomized clinical trial. The RMST difference allows quantification of the postponement of an outcome during a specified (restricted) interval and corresponds to the difference between the areas under the 2 survival curves for the intervention and control groups. This article presents examples of the use of the RMST in a research and a clinical context. First, the authors demonstrate how the RMST difference can answer research questions about the efficacy of different treatments. Estimates are presented for the effects of pharmacologic or strategy-driven glucose-lowering interventions for adults with type 2 diabetes from 36 trials and 9 follow-up studies reporting cardiovascular outcomes and mortality. The authors show how these measures may be used to mitigate uncertainty about the efficacy of intensive glucose control. Second, the authors demonstrate how the RMST difference may be used in the setting of a clinical consultation to guide the decision to start or discontinue a treatment. They then discuss the advantages of the RMST over the absolute risk difference, the number needed to treat, and the median survival time difference. They argue that the RMST difference is both easy to interpret and flexible in its application to different settings. Finally, they highlight the major limitations of the RMST, including difficulties in comparing studies of heterogeneous designs and in inferring the long-term effects of treatments using trials of short duration, and summarize the available statistical software for calculating the RMST.
Kloecker DE, Davies MJ, Khunti K, et al. Uses and Limitations of the Restricted Mean Survival Time: Illustrative Examples From Cardiovascular Outcomes and Mortality Trials in Type 2 Diabetes. Ann Intern Med. 2020;:. [Epub ahead of print 24 March 2020]. doi: https://doi.org/10.7326/M19-3286
Download citation file:
Published: Ann Intern Med. 2020.
Published at www.annals.org on 24 March 2020
Research and Reporting Methods.
Results provided by:
Copyright © 2020 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use