An important question that may arise when studying the effects of medicines is whether such effects differ between subgroups of patients (effect modification). To answer this question, one can stratify the study population, e.g. by gender, and compare the effects in these subgroups. In CONSORT 2010 Explanation and Elaboration: Updated guidelines for reporting parallel group randomised trials (J Clin Epidemiol 2010;63(8):e1-37) and Interaction revisited: the difference between two estimates (BMJ 2003;326:219), it is recommended to perform a formal statistical test to assess if there are statistically significant differences between subgroups for these effects. The study report should explain which method was used to examine these differences and specify which subgroup analyses were predefined in the study protocol and which ones were performed while analysing the data (Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration. Epidemiology 2007;18:805-35).
Effect modification can be measured in two ways: on an additive scale (based on risk differences [RD]), or on a multiplicative scale (based on relative risks [RR]). From the perspective of public health and clinical decision making, the additive scale is usually considered most appropriate. The standard measure for interaction on the additive scale is the relative excess risk due to interaction (RERI), as explained in the textbook Modern Epidemiology (K. Rothman, S. Greenland, T. Lash. 3rd Edition, Lippincott Williams & Wilkins, 2008). Other measures of interaction include the attributable proportion (A) and the synergy index (S). With sufficient sample size, most interaction tests perform similarly with regard to type 1 error rates and power according to Exploring interaction effects in small samples increases rates of false-positive and false-negative findings: results from a systematic review and simulation study (J Clin Epidemiol 2014; 67(7):821-9). In small samples (<250), the Breslow-Day and Tarone test performed best for interactions on the odds-ratio scale, whereas Likelihood Ratio and RERI-based tests performed better on RD scale. When exposure prevents outcome, in small samples the RERI-based test is relatively underpowered compared to other tests. Possible solutions include choosing an alternative interaction test, or recoding exposure categories taking the category with the lowest risk as reference.
Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration (Epidemiology 2007;18:805-35) and Recommendations for presenting analyses of effect modification and interaction (Int J Epidemiol 2012;41:514-20) recommend that effect modification should be reported as follows:
It should be kept in mind that past drug use should be considered as a potential effect modifier in studies assessing the risk of occurrence of events associated with recent drug use. This is shown in Evidence of the depletion of susceptibles effect in non-experimental pharmacoepidemiologic research (J Clin Epidemiol 1994;47(7):731-7) in the context of a hospital-based case-control study on NSAIDs and the risk of upper gastrointestinal bleeding.
|Annex 1.||Guidance on conducting systematic revies and meta-analyses of completed comparative pharmacoepidemiological studies of safety outcomes|