Page 63 - ESHRE2019

6. Selective reporting
The selection of a subset of the original variables recorded, on the basis of the results, for inclusion in publication of trials
Are all collected outcomes reported?
• Trial registration (beware retrospective registration) • Published study protocol (uncommon)
• Methods stated in paper
Considerations
• Not as relevant in fertility trials with standard gestations of pregnancy • Intervention is unlikely to cause increase in pregnancy rate but not live birth rate, i.e. also an increase in miscarriage rate – though possible
• Studies may use several measures for the same outcome (e.g. quality of life, hot flush frequency) and selectively report the most favourable measure
6. Selective reporting
LOW RISK
• All collected outcomes reported as per trial protocol or trial registration (beware retrospective registration/date first entered)
• All likely/expected outcomes reported (e.g. live birth and relevant adverse events)
UNCLEAR RISK
• Not reporting all expected outcomes and unclear whether these were collected (e.g. no detailed trial registration or protocol available)
HIGH RISK
• Not reporting all outcomes collected as per registration, protocol, methods in paper
Default is unclear
58
PRECONGRESS COURSE 17 I VIENNA, AUSTRIA – 23 JUNE 2019 61