Five theses concerning the clinical consequences of pathology and prognostic factors

We discuss possible justifications to split study populations from a biomet rical point of view. The existence of prognostic differences between subgro ups are neither a sufficient nor a necessary reason to justify a splitting decision. There are essentially two separate types of relevant arguments to justify a split of patient study populations: a) Different toxicity/benefi t trade-offs concerning the acceptability of a particularly aggressive trea tment, b) Evidence for strong treatment by subgroup interactions. i.e. Trea tment differences differ markedly by biologically defined subgroups. The la tter is what the research ideal of biologically specific treatment asks for . Subgroup analysis is notoriously difficult. Formal statistical analysis m ust be complemented by specific evidence from basic sciences. Meta-subgroup analyses may be an option if a biologically specific hypothesis on which t reatment component interacts with what biological feature allows to operati onally identify all those randomised trials in which the effect should be p resent.