Risk of bias-assessment
Among the eligible studies, 7 (44%) were of high quality. Most studies were at low risk for selection bias (12/16, 75%) and all studies controlled for at least one confounding factor (maternal age, body mass index (BMI) and parity). Given we only considered studies using validated questionnaires for patient self-reporting, by definition we scored all studies at risk of bias for the assessment of outcome. Nine of the included studies for quantitative synthesis (9/16, 56%) were at high risk of bias for outcome/exposure domain, though. Details on quality assessment and risk of bias for individual studies are displayed in Appendix 2B.