Risk of suicidal ideation, suicide attempts, and suicide deaths in persons with sleep apnea: Protocol for a systemic review and meta-analysis

Aim
To estimate the pooled prevalence and incidence of suicidal ideation, attempts, and deaths in people with sleep apnea.
Method
We will identify epidemiological studies reporting the prevalence or incidence rate of suicide in people with sleep apnea. We will search the following databases: PubMed (MEDLINE), Scopus, Cochrane Library, OVID (HEALTH STAR), OVID (MEDLINE) and Joana Briggs Institute EBF Database. No age, geographical location, study-design or language limits will be applied. This protocol was developed in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocols (PRISMA-P) guidelines. Two reviewers (YY and YP) will independently screen citations, abstracts and will identify full-text articles for inclusion, extract data, and appraise the quality and bias of included studies. Discrepancies will be resolved by consulting with a third researcher (MC). Study quality will be assessed by the Newcastle-Ottawa Scale. The primary outcomes will be the overall prevalence or incidence of suicidal ideation, attempts and completion and the risk of suicide in people with sleep apnea. For pooling of the studies, we will use a random-effects model with a logit transformation. The DerSimonian and Laird (DL) random-effects method will be used to estimate the pooled inter-study variance. We will assess the between-study heterogeneity using I2 statistics, and Cochrane’s Q statistic (significance level < 0.05). If the I2 is high (>75%), we will perform subgroup meta-analyses and conduct a meta-regression analysis to explore sources of study heterogeneity using study level median age, study-level proportions of race, gender, depression and quality scores. We will report effect estimates as suicide risk per 1000 individuals. Egger’s test and funnel plots will be used to assess publication bias, and adjusted estimates using trim and fill methods will be reported if publication bias is suspected.