StatsMiniBlog: Subgroup or sensitivity analyses?


Perhaps becoming a little obscure, but there are some folk in the world who become concerned about undertaking analysis in systematic reviews.

Some of these are described as “subgroup” analysis, others “sensitivity”. What’s the difference, and why?

Simply put – there’s often little difference. But a practical differentiation might go along the lines of

Subgroup analysis (or meta-regression): “How might this [intervention] have a different effect in different groups” e.g. different ages of patient, or different type of quinolone, or different diagnostic classification of disease

Sensitivity analysis: “How might our analysis be affected by something in the assumptions we have made?” e.g. lumping high and low quality studies together, regarding outcomes at 15 and 30 days as equivalent, or using fixed vs. random effects meta-analysis

– Archi

(Visited 145 times, 1 visits today)