Subgroup analyses in clinical research: too tempting?

European Journal of Endocrinology, vol. 189, pp. E1–E3

Abstract

In many biomedical studies, subgroup analyses are performed to identify subgroups of patients in whom a treatment is most effective, or a risk factor has the largest effect. While both are referred to as subgroup analysis, it is important to distinguish between the estimation of effects within subgroups and the comparison of effects across subgroups. Both are discussed, and we outline the implications regarding sample size and statistical methods for estimation of effects. Also, the risk of false-positive findings-which potentially increases with subgroup analysis-is discussed, as well as the distinction between effect modification and interaction.

Authors 2

1. Rolf H. H. Groenwold corresponding Aachen Department of Biomedical Data Sciences Department of Clinical Epidemiology

   Leiden University Medical Center

   Affiliation as printed

   Department of Biomedical Data Sciences, Leiden University Medical Center , Leiden, 2333 ZA , the Netherlands

   Department of Clinical Epidemiology, Leiden University Medical Center , Leiden, 2333 ZA , the Netherlands

2. Olaf M. Dekkers Aachen Department of Clinical Epidemiology Department of Endocrinology, Leiden University Medical Center

   Leiden University Medical Center

   Affiliation as printed

   Department of Clinical Epidemiology, Leiden University Medical Center , Leiden, 2333 ZA , the Netherlands

   Department of Endocrinology, Leiden University Medical Center , Leiden, 2333 ZA , the Netherlands