This meta-analysis was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA)[13]. Ethics approval is not required when human subjects were not involved.
Search strategy
In this meta-analysis, a comprehensive search was performed on PubMed, Embase, web of science, Scopus, Cochrane trial register, China National Knowledge Infrastructure (CNKI) and Wanfang Databases from inception to October, 2020. Article was restricted to human studies. language limitation was not applied. We used a combined search strategy for PCOS and asthma. For PCOS, the following terms were used:"PCOS", "polycystic ovary syndrome", "polycystic ovary",”Stein Leventhal Syndrome”. For asthma, the following terms were used: "asthma", "wheezing". All Potential studies were browsed by two independent reviewers, and elementary selection was conducted. If there was a controversy on inclusion of one article, two additional authors (J.J. and L.D.) will view the article and made the decision. All studies concerning the difference of the prevalence of asthma between PCOS patients and controls were included. We also browsed the references of included articles for more potential articles.
Inclusion criteria
Observational studies, such as cross-sectional studies, case-control studies and cohort studies comparing the prevalence of asthma in PCOS patients with controls were included in the meta-analysis. Necessary information must be available in studies to calculate the odds ratio [OR], otherwise a risk ratio (OR, relative risk [RR], or hazard ratio[RR]) should be directly provided.
Exclusion criteria
All abstracts, review articles, meta-analyses, letters, case reports and expert opinions were excluded. Studies without complete data or lack control group were also removed. For articles that based on same population source, only the latest and most detailed report was included.
Data extraction and quality assessment
Two independent reviewers(S.H.H. and W.X.X.) extracted following data from article texts and tables: Title, author names, year of publication, study design, geographic region, characteristics of PCOS and matched control population, diagnostic criteria for PCOS and asthma and the event number of PCOS and matched control population. Adjusted risk ratio such as adjusted odds ratio [aOR], adjusted relative risk[aRR] was also collected if given.
The quality of the Included studies were assessed by two investigators (S.H.H. and W.X.X.), using the Newcastle-Ottawa Scale (NOS) separately. If they have different opinions, two other reviewers (L.D.and J.J) will resolve the dissagreement. Only high quality (obtaining ≥70% of the highest score) or moderate quality (obtaining 40-70% of the highest score) articles will be included.
Statistical analysis
The odds ratio (OR) and 95% confidence intervals (CIs) were used to measure the difference of asthma prevalence between PCOS patients and matched control group. I2 test, Galbraith plot and funnel plot analyses were used to detect heterogeneity. Random-effects model was applied due to significantly heterogeneity was found.We also performed subgroup analyses on geographic region, study design and diagnostic criteria for PCOS. Begg’s test and Egger’s test were applied to detect potential publication bias. Statistical analyses were conducted using Stata software version 11.0.