Statistical heterogeneity in systematic reviews of clinical trials: a critical appraisal of guidelines and practice

Ref ID 293
First Author J. Higgins
Journal JOURNAL OF HEALTH SERVICES RESEARCH & POLICY
Year Of Publishing 2002
URL https://pubmed.ncbi.nlm.nih.gov/11822262/
Keywords Heterogeneity
Pre-specification
Subgroup
General medical
Problem(s) Inadequate analysis of heterogeneity
Unplanned or unjustified subgroup or sensitivity analyses
Number of systematic reviews included 39
Summary of Findings Plans for addressing heterogeneity between studies in systematic review protocols were seldom followed in the 39 Cochrane reviews examined. Fifteen of the 28 protocols pre-specified potential effect modifiers for use in subgroup analyses or meta-regression. Just two reviews performed only subgroup analyses that had been pre-specified in the protocol. Ten reviews did not undertake any of the planned subgroup analyses or meta-regressions, and in four further reviews at least one planned subgroup analysis was not undertaken. Ten reviews undertook subgroup analyses that were not mentioned in the protocol. Eight of the 11 reviews without protocols undertook subgroup analyses; it was not possible to determine whether these had been prespecified.
Did the article find that the problem(s) led to qualitative changes in interpretation of the results? Not Applicable
Are the methods of the article described in enough detail to replicate the study? Yes