A Review of Published Analyses of Case-Cohort Studies and Recommendations for Future Reporting

Abstract

The case-cohort study design combines the advantages of a cohort study with the efficiency of a nested case-control study. However, unlike more standard observational study designs, there are currently no guidelines for reporting results from case-cohort studies. Our aim was to review recent practice in reporting these studies, and develop recommendations for the future. By searching papers published in 24 major medical and epidemiological journals between January 2010 and March 2013 using PubMed, Scopus and Web of Knowledge, we identified 32 papers reporting case-cohort studies. The median subcohort sampling fraction was 4.1% (interquartile range 3.7% to 9.1%). The papers varied in their approaches to describing the numbers of individuals in the original cohort and the subcohort, presenting descriptive data, and in the level of detail provided about the statistical methods used, so it was not always possible to be sure that appropriate analyses had been conducted. Based on the findings of our review, we make recommendations about reporting of the study design, subcohort definition, numbers of participants, descriptive information and statistical methods, which could be used alongside existing STROBE guidelines for reporting observational studies.

Circulating Omentin as a Novel Biomarker for Colorectal Cancer Risk: Data from the EPIC–Potsdam Cohort Study

“…S1). We considered recent recommendations for case-cohort study reporting in data analysis and representation (30).…”

cchs: An R Package for Stratified Case-Cohort Studies

“…In case-cohort studies, stratified selection of the subcohort seems to be common (Sharp et al, 2014), so there are likely to be plenty of investigations where cchs may be useful. The way in which cchs manipulates the data makes it faster and more computationally efficient than the previously published SAS and S-Plus code.…”

“…The subcohort is about half as big as the set of cases. In case-cohort studies in general, the sampling fraction is typically in the area of 5% but sometimes larger (Sharp et al, 2014), and the subcohort is usually larger than the set of cases (Juraschek et al, 2013;Lamb et al, 2013;Jones et al, 2015) but occasionally smaller (Huxley et al, 2013).…”

A method making fewer assumptions gave the most reliable estimates of exposure–outcome associations in stratified case–cohort studies

“…If multiple imputation was used with Model V, it would need to be done in each stratum separately, using Rubin's rules, before the meta-analysis [20] . We have only considered methods based on Cox regression, with a nonparametric baseline hazard, since these seem to be used almost exclusively in practice [21] , but parametric survival models for stratified case–cohort studies could be developed. These might have particular relevance for risk prediction [22] , whereas the focus of this article has been on estimating the risk association of a particular exposure.…”