Methodological review showed correct absolute effect size estimates for time-to-event outcomes in less than one-third of cancer-related systematic reviews

To evaluate in how many cancer-related Cochrane reviews hazard ratio (HR)-based absolute effects in summary of findings (SoF) tables have been correctly calculated and reported. We identified all Cochrane cancer intervention reviews that reported an HR for at least one outcome and provided a SoF tab...

Full description

Saved in:
Bibliographic Details
Published in:Journal of clinical epidemiology 2019-04, Vol.108, p.1-9
Main Authors: Skoetz, Nicole, Goldkuhle, Marius, Weigl, Aaron, Dwan, Kerry, Labonté, Valérie, Dahm, Philipp, Meerpohl, Joerg J., Djulbegovic, Benjamin, van Dalen, Elvira C.
Format: Article
Language:English
Subjects:
Absolute effects
Cancer
Confidence intervals
Data Display
Data Interpretation, Statistical
Disease-Free Survival
Epidemiology
Estimates
Hazard ratio
Humans
Labeling
Mathematical analysis
Methodological
Mortality
Neoplasms - therapy
Peer review
Proportional Hazards Models
Research Report
Review
Reviews
Software
Studies
Summary of findings
Survival
Systematic review
Systematic Reviews as Topic
Tables
Tables (data)
Time-to-event
Treatment Outcome
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:To evaluate in how many cancer-related Cochrane reviews hazard ratio (HR)-based absolute effects in summary of findings (SoF) tables have been correctly calculated and reported. We identified all Cochrane cancer intervention reviews that reported an HR for at least one outcome and provided a SoF table, published between January 2011 and December 2017 in the Cochrane Database of Systematic Reviews. In 28 reviews (29%) of 96 included Cochrane reviews, absolute effects in the SoF tables were calculated in a correct manner. In 23 reviews (24%), absolute effects had been correctly calculated, but there was no explanation given why authors calculated event-free survival (e.g., overall survival) throughout the review but reported number of events in SoF tables (e.g., death). Twelve reviews (13%) provided incorrect absolute effects. For seven reviews (7%), it was unclear if absolute effects were correctly calculated. In 26 (27%) reviews, no absolute effects based on the given HR were calculated. In less than one-third of cancer-related Cochrane reviews, absolute effect size estimates were correctly calculated and reported. There is a need for guidance on how to calculate and report absolute effect estimates based on HR data.
ISSN:0895-4356
1878-5921
DOI:10.1016/j.jclinepi.2018.12.006