Statistical Analysis
Meta-analysis was conducted using
Stata14.0 (StataCorp, College Station, TX, USA). Heterogeneity was
assessed by χ² analysis. Data synthesis was conducted using a
random-effects model if I² > 50%, and using a
fixed-effects model otherwise. RR was converted to OR for data
synthesis. Subgroup analysis was performed with stratification according
to geographical location and quality grade. Meta-regression analysis was
conducted using year of publication, region, incidence of asthma in the
control group, sample size and quality score as explanatory covariates.
Sensitivity analysis was conducted with the method of reducing one piece
of literature at a time to evaluate the impact of a single study on this
meta-analysis. Publication bias was examined using the funnel plot and
Begg’s bias test if 10 or more studies were included. P<0.05 (2-sided) was considered to be statistically
significant.