Association of depression and gastrointestinal diseases: a three-stage study.

Background: Depression as a mental illness is commonly observed to co-occur with various somatic diseases, such as gastrointestinal diseases. However, previous studies have primarily focused on the risk of mental disorders following physical illnesses. Our study took depression as a risk factor, attempting to explore its relationship with gastrointestinal diseases.

Methods: A total of 457,940 participants (aged 37-73 years) in the UK Biobank were included. The Cox proportional hazards model was used to assess the relationship between depression and gastrointestinal diseases. Mendelian randomization assessed the causal link between depression and gastrointestinal disorders, and seven machine learning algorithms (including LightGBM, XGBoost, and Random Forest) were trained in the total population to develop predictive models for incident gastrointestinal diseases, with model performance evaluated using the area under the receiver operating characteristic curve (AUC).

Results: During a median follow-up period of 13.7 years, 9563 esophagitis events, 36,420 gastroesophageal reflux disease events, 5469 gastric ulcer events, 3096 duodenal ulcer events, 37,225 gastritis and duodenitis events, and 9153 dyspepsia events were recorded. After adjusting for covariates, depression was associated with increased risk of all six diseases. Two-sample MR analysis supported a causal association. Machine learning models demonstrated good discrimination, with the highest predictive accuracy observed for duodenal ulcer (AUC = 0.76) and gastric ulcer (AUC = 0.75).

Conclusions: Addressing depression as a modifiable risk factor may reduce gastrointestinal disease risk, especially in disadvantaged populations, by integrating mental health care into primary care and using predictive models for early intervention.