First justify combining results
Compare populations, interventions, outcome definitions and follow-up. Explain why pooling is appropriate. A single combined number is not always a useful representation of the available studies.
Understand the model
Study estimates and their precision contribute according to the chosen model. Heterogeneity may reflect context, methods or chance. A heterogeneity statistic does not by itself explain why studies differ.
Examine robustness
Consider sensitivity analyses, study bias and unpublished evidence. A narrow confidence interval obtained from biased studies does not make the answer reliable. The synthesis must retain those limitations.
Interpretation limits
Not every systematic review needs, or can support, a meta-analysis.
Original references
Continue exploring
Research questions
Select tools by method, population and purpose. A thematic connection does not validate use in every case.
Educational content for framing research. Consult the original document, study population and methodology before interpreting a conclusion. Methodology