Unmet social needs and diverticulitis: a phenotyping algorithm and cross-sectional analysis.

Area of Science:

• Medical informatics
• Public health
• Gastroenterology

Background:

• Diverticular disease severity varies significantly.
• Social determinants of health can impact disease presentation and outcomes.
• Accurate phenotyping is crucial for understanding disease burden.

Purpose of the Study:

• To validate a phenotyping algorithm for classifying diverticular disease severity.
• To investigate the association between unmet social needs and diverticular disease severity.

Main Methods:

• Developed and validated a phenotyping algorithm for diverticulosis, mild diverticulitis, and severe/recurrent diverticulitis.
• Applied the algorithm to a cohort from the All of Us Research Program.
• Analyzed the distribution of social barriers (food insecurity, housing instability, care access) across deprivation levels.
• Used logistic regression to assess the association between social needs and disease severity.

Main Results:

• The phenotyping algorithm demonstrated high accuracy with positive predictive values from 0.87 to 0.97 and negative predictive values from 0.97 to 0.99.
• Unmet social needs showed variable distribution across deprivation quintiles.
• An association was found between unmet social needs and increased odds of operative or recurrent inpatient diverticulitis (OR 1.61 [95% CI 1.19-2.17]) compared to diverticulosis.

Conclusions:

• A validated phenotyping algorithm effectively categorizes diverticular disease severity.
• Unmet social needs are significantly associated with more severe presentations of diverticulitis.
• Understanding social barriers within disease-specific cohorts can guide targeted clinical interventions.