This is an archived version of the Handbook. For the current version, please go to or search for this chapter here.

18.4.5  Subgroup analysis

Collecting IPD is also the most practical way to carry out analyses to investigate whether any observed effect of an intervention is consistent across well-defined types of participants, for example whether women gain a smaller or larger benefit from treatment than men. In conventional analyses using aggregate data from publications, it is usually very difficult to extract sufficient compatible data to undertake meaningful subgroup analyses, and especially difficult to characterize individuals by more than one factor at a time. In contrast, IPD permit straightforward categorization of individuals for subgroup analysis (stratified by study) defined by single or multiple factors. The collection of IPD will also allow more complex analyses, such as multilevel modelling, to explore associations between intervention effects and patient characteristics.