Page 50 - ESHRE2019
P. 50
Basic effect measures
• Dichotomous
– Relative risk or risk ratio
– Odds ratio
– Absolute risk difference
– Number needed to treat
• Continuous
– Mean difference
– Standardized mean difference
Steps in doing a meta-analysis
1. do a basic literature search
2. define comparisons for your review
3. decide on the outcomes for each comparison
4. select a summary statistic for each comparison
5. decide on fixed or random model
6. write a protocol and register your study
7. do an extended literature search
8. assess the similarity of study results within each comparison
9. consider the reliability of the summaries
48
45
