Thirteen RCTs (n=6,708) were included in the meta-analysis. Sample size varied from 22 to 3,407. Three RCTs were judged as fair quality and 10 were judged as poor quality.
Pooled analysis:
When the studies were pooled, elective induction of labour was associated with a non-significant difference in perinatal mortality compared with expectant management for post-date pregnancy (relative risk 0.33, 95% CI: 0.10 to 1.09, p=0.07; 11 RCTs). Elective induction of labour was associated with a significant reduction in meconium aspiration syndrome (relative risk 0.43, 95% CI: 0.23 to 0.79, p=0.007; seven RCTs), a significantly lower mean birth weight (weighted mean difference -44.41, 95% CI: -79.37 to -9.45, p=0.01; eight RCTs) and a significant reduction in caesarean section (relative risk 0.87, 95% CI: 0.80 to 0.96; p=0.004; 13 RCTs). There were no statistical differences in other outcomes for the two groups.
Subgroup analyses:
Elective induction of labor in the 41-week group was associated with a significant reduction in meconium aspiration syndrome (relative risk 0.35, 95%CI: 0.16 to 0.75, p=0.007; five RCTs) and a significant reduction in caesarean section (relative risk 0.87, 95% CI: 0.79 to 0.96, p=0.006; nine RCTs). In the 42-week group, elective induction of labour was associated with a significant lower mean birth weight compared with expectant management (weighted mean difference -101.58, 95% CI: -179.01 to -24.15, p=0.01; three RCTs).
No statistically significant heterogeneity was observed in the outcomes. Sensitivity analysis altered the result for the outcome of caesarean delivery significantly. The result of assessing publication bias was unreported.