Data extraction
Data extraction was performed at a meta-analysis and individual study
level. At a meta-analysis level, we extracted information on first
author, year of publication, exposure, outcome, window of exposure
(preconception, pregnancy, delivery or childhood), number of studies,
effect size and 95% CI, level of adjustment and model of analysis
(fixed effects or random effects). At an individual study level, we
extracted information on first author, year of publication,
epidemiological design, number of cases and total sample size in
case-control studies or total population in cohort studies, maximally
adjusted effect size and 95% CI.
The data extraction database included study quality indicators, in
particular elements from the 11-point AMSTAR measurement tool to assess
the methodological quality of systematic reviews. AMSTAR is considered a
reliable and valid tool for quality assessment of systematic reviews and
meta-analyses of both interventional and observational research . Based
on the AMSTAR tool, the study quality was categorized into low (0-3
points), moderate (4-7 points), and high (8-11 points). Two
investigators (CT, AK) independently performed the data extraction and
quality assessment; disagreements were resolved by discussion with a
third investigator (MK).