Adjusting for publication bias: modelling the selection process.

UNLABELLED: RATIONALE, AIMS AND BACKGROUND: Systematic review with meta-analysis, a statistical technique for combining results of several studies, is progressively being used to guide decisions in medicine. Publication bias is acknowledged as a threat to the validity of systematic reviews and its existence may lead to inappropriate decisions about patient management or health policy. It is said to occur when the results of research available in the literature are not representative of the totality of all research. The selection mechanism that causes publication bias is complex, yet despite an extensive literature of empirical research identifying risk factors for publication, little work has been done to improve models of selection. Methods METHODS: that adjust combined meta-analytic estimates for publication bias are compared and applied to a systematic review of oral rehydration solution in the treatment of dehydration. Within a weighted distributions framework models of the selection process are considered and developed further. CONCLUSIONS: Weighted distributions offer a flexible approach that allows the potential to modify the selection function to incorporate other factors. Methods that adjust combined estimates should not be used to provide an alternative answer but to consider the robustness of the combined estimate to publication bias.

Pregnancy outcome in severe placental abruption.

OBJECTIVE: To determine the relationship between decision to delivery interval and perinatal outcome in severe placental abruption. DESIGN: A case-control study. SETTING: Large inner city teaching hospital. METHODS: Retrospective case note review of pregnancies terminated following severe placental aburption and fetal bradycardia. One year paediatric follow up by case note review or postal questionnaire. The differences in outcome (death or cerebral palsy) were examined using non-parametric and univariate analysis for the following time periods--times from onset of symptoms to delivery, onset of symptoms to admission, admission to delivery, onset bradycardia to delivery and decision to delivery interval. MAIN OUTCOME MEASURES: Prenatal death or survival with cerebral palsy. RESULTS: Thirty-three women with singleton pregnancies over 28 weeks of gestation, admitted with clinically overt placental abruption, where delivery was effected for fetal bradycardia. Eleven of the pregnancies had a poor outcome (cases), eight infants died and three surviving infants have cerebral palsy. Twenty-two pregnancies had a good outcome (controls): survival with no developmental delay. No statistically significant relationship was found between maternal age, parity, gestation, or birthweight and a poor outcome. A statistically significant relationship between time from decision to delivery was identified (P = 0.02, Mann-Whitney U test). The results of a univariate logistic regression for this variable suggest that the odds ratio of a poor outcome for delivery at 20 minutes compared with 30 minutes is 0.44 (95% CI 0.22-0.86). Fifty-five percent of infants were delivered within 20 minutes of the decision to deliver. Serious maternal morbidity was rare. CONCLUSION: In this small study of severe placental abruption complicated by fetal bradycardia, a decision to delivery interval of 20 minutes or less was associated with substantially reduced neonatal morbidity and mortality.