Cumulative subgroup analysis to reduce waste in clinical research for individualised medicine

Abstract

BackgroundAlthough subgroup analyses in clinical trials may provide evidence for individualised medicine, their conduct and interpretation remain controversial.MethodsSubgroup effect can be defined as the difference in treatment effect across patient subgroups. Cumulative subgroup analysis refers to a series of repeated pooling of subgroup effects after adding data from each of related trials chronologically, to investigate the accumulating evidence for subgroup effects. We illustrated the clinical relevance of cumulative subgroup analysis in two case studies using data from published individual patient data (IPD) meta-analyses. Computer simulations were also conducted to examine the statistical properties of cumulative subgroup analysis.ResultsIn case study 1, an IPD meta-analysis of 10 randomised trials (RCTs) on beta blockers for heart failure reported significant interaction of treatment effects with baseline rhythm. Cumulative subgroup analysis could have detected the subgroup effect 15 years earlier, with five fewer trials and 71% less patients, than the IPD meta-analysis which first reported it. Case study 2 involved an IPD meta-analysis of 11 RCTs on treatments for pulmonary arterial hypertension that reported significant subgroup effect by aetiology. Cumulative subgroup analysis could have detected the subgroup effect 6 years earlier, with three fewer trials and 40% less patients than the IPD meta-analysis. Computer simulations have indicated that cumulative subgroup analysis increases the statistical power and is not associated with inflated false positives.ConclusionsTo reduce waste of research data, subgroup analyses in clinical trials should be more widely conducted and adequately reported so that cumulative subgroup analyses could be timely performed to inform clinical practice and further research.Electronic supplementary materialThe online version of this article (doi:10.1186/s12916-016-0744-x) contains supplementary material, which is available to authorized users.

Justification and reporting of subgroup analyses were lacking or inadequate in randomized controlled trials

“…in line with items used in a previous study (Supplementary Appendix 1) [15]. The reporting of subgroup analyses in a trial publication was judged to be sufficient for cumulative subgroup analyses [18] if the trial report provided estimated subgroup effects with standard errors, confidence intervals, or interaction P values [19]. A subgroup effect was considered to be as claimed if the trial authors explicitly stated in the abstract or discussion/ conclusion that the effect of an intervention was different between subgroups or that a clear benefit or harm was seen in one or more subgroups [20].…”

“…As attention has been focused mainly on the problem of inflated false positives, efforts to avoid inappropriate subgroup analyses may have resulted in a recent reduction in the proportion of trial protocols with planned subgroup analyses as shown in the present study. If fewer or no subgroup analyses are conducted in clinical trials, it will be difficult or impossible to examine the consistency in results of a subgroup analysis across different trials, and valuable data from clinical research will be wasted [18]. Therefore, methodological guidance on subgroup analyses in clinical research should emphasize the importance of sufficient reporting of results of all prespecified or post hoc subgroup analyses conducted, rather than insist on only a small number of planned subgroup analyses.…”

“…Particularly, the present study summarized empirical evidence about how specified subgroup analyses were justified or what rationales were used for subgroup analyses in a large number of RCTs. Furthermore, the present study is the first to examine whether results of subgroup analyses were sufficiently reported to facilitate the checking of consistency in subgroup effects reported and the conduct of cumulative subgroup analyses [18]. Similar to cumulative metaanalysis [29], cumulative subgroup analysis refers to a series of repeated meta-analyses of subgroup effects after adding data from each new trial chronologically.…”

“…For a single study, the sample size may be too small to reveal significant subgroup effects. However, if multiple small studies on the same topic conduct the same subgroup analyses, cumulative subgroup analyses may show how consistent are results of subgroup analyses across individual trials and provide a more valid overall estimate of subgroup effects [18].…”

Checklist for clinical applicability of subgroup analysis

“…Observation of consistency greatly supports reliability of the observed difference, and the opposite is true when inconsistency is observed. A meta‐analysis of subgroup results obtained in similar studies can provide additional information …”