As this study involves the summary and analysis of other studies, it does not involve medical ethics approval or patient-informed consent.

This study was conducted in accordance with the PRISMA guidelines23, and to systematically assess the relationship between OSA and GID, the study was searched in several databases, including PubMed, Web of science, Willey Library, Cochrane Library, and Scopus. The search strategy included keywords and medical subject terms related to OSA and various GID. The search language was English, and the search period was from the establishment of each database to January 2024.

The search terms included “Obstructive Sleep Apnea”, “OSA”, “Gastrointestinal Diseases”, “Gastroesophageal Reflux Disease (GERD)”, “Irritable Bowel Syndrome (IBS)”. “Inflammatory Bowel Disease (IBD)”, “Crohn’s Disease”, “Ulcerative Colitis “, “Gastritis”, “Peptic Ulcer Disease”, “Gastroenteritis “, “Colorectal Cancer”. The search was performed using a combination of subject terms and free words with matching truncation, using PubMed as an example: (“Obstructive Sleep Apnea” [MeSH Terms] OR “OSA” [Title/Abstract]) AND (“Gastrointestinal Diseases” [MeSH Terms] OR “Gastroesophageal Reflux Disease (GERD)”[Title/Abstract] OR “Irritable Bowel Syndrome (IBS)”[Title/Abstract] OR “Inflammatory Bowel Disease (IBD)”[Title/Abstract] OR “Crohn’s Disease”[Title/Abstract] OR “Ulcerative Colitis”[Title/Abstract] OR “Gastritis”[Title/Abstract] OR “Peptic Ulcer Disease”[Title/Abstract] OR “Gastroenteritis”[Title/Abstract] OR “Colorectal Cancer”[Title/Abstract]).

Inclusion criteria included (1) studies with a clear diagnosis of OSA, and the basis for the diagnosis should include standard sleep monitoring data or recognised clinical diagnostic guidelines; (2) studies assessed the association between OSA and GID (such as GERD, IBS, IBD, Crohn’s Disease, Ulcerative Colitis); (3) studies provided quantitative data, such as risk ratios (RR), odds ratios (OR), or correlation data; (4) the publication was a peer-reviewed full-text article; and (5) the article was written in English.

Exclusion criteria included: (1) interventional studies, case reports, commentaries, expert opinions, conference abstracts, review articles, or non-peer-reviewed publications; (2) studies that did not provide sufficient data to estimate risk ratios or odds ratios; (3) studies in which the diagnostic criteria for OSA or GID were unclear or did not meet widely accepted clinical guidelines; (4) studies that were repetitively published or had duplicated data; (5) studies that were not original, such as secondary data based on published data.

The literature screening process for this study followed a rigorous process, wherein the titles and abstracts of retrieved literature were initially examined by two intended researchers. The preliminary screening aimed to exclude irrelevant studies, specifically those not addressing OSA or a specific GID. Subsequently, a detailed full-text review was undertaken for literature that seems to the inclusion criteria, to determine whether the inclusion criteria were fully met. Discrepancies in opinion were resolved through discussion or, if necessary, by consulting third-party experts.

The Newcastle-Ottawa Scale (NOS), a standardised tool for appraising the quality of observational studies, particularly cohort and case-control studies, was used in this study for literature quality evaluation. The NOS encompasses three main domains: selectivity (choice of the study population), comparability (comparisons between study groups), and outcome (assessment of the study’s results.) The NOS scale reflects the methodological quality of each study by providing it with an overall score of up to 9 points. In this study, each study meeting the inclusion criteria was independently assessed by two researchers using the NOS scale. Disagreements during the scoring process were addressed through discussion, and third-party experts were consulted as needed.

In this study, the data extraction process was carried out independently by two researchers to ensure the accuracy and completeness of the data. A comprehensive data extraction form was designed to gather key information from each study, including (1) basic information about the study, such as authors, year of publication, and type of study design; (2) characteristics of the study population, including sample size, age range, and gender ratio; (3) primary and secondary outcome indicators, including all relevant clinical outcomes and measurements; (4) study results, such as the odds ratio (OR), correlation coefficient, and so on, as well as their confidence intervals and statistical significance levels. Upon completion of each data extraction, investigators cross-verified the data to eliminate potential errors or biases. Any disagreements identified during the data extraction process were resolved through discussion, and third-party expert consultation was sought as necessary.

The Meta-analysis was conducted using the meta-package in R software (R 4.2.2). Firstly, we will extract the corrected Odds Ratio (OR) of the outcome metrics and their 95% Confidence Interval (CI) from each included study. Subsequently, these ORs and 95% CIs were converted to logORs and their standard errors (Standard Error, SE) for meta-analysis. For the heterogeneity test, I² statistic was used to assess the heterogeneity between studies, when I² ≤ 50%, indicating small heterogeneity, a fixed-effect model (FEM) was employed for meta-analysis. If I² > 50%, indicating a high degree of heterogeneity, in which case a random-effects model was used to analyse the data. To assess the quality of the studies, the Newcastle-Ottawa Scale (NOS) was used for cohort studies. In addition, for studies numbering 10 or more, funnel plots, Begg’s test, and Egger’s test were utilized to assess possible publication bias. A statistically significant difference of P < 0.05 was used as the criterion for all tests.