Adverse Social Determinants of Health and Urinary Incontinence Among Older Adult Women in Home Health Care

Objective: To leverage secondary data extracted using a natural language processing algorithm to examine relationships between adverse social determinants of health and urinary incontinence (UI) among older adult women in home health care. Design: Cross-sectional study. Setting and Participants: We analyzed home health care episode data from 2015 to 2017 for women ≥65 years old with UI identified based on federally mandated assessments. Methods: Adverse social determinants of health were identified a priori (food insecurity; low education/literacy; problems with the social environment, physical environment, economic circumstances, and health care access) and extracted from clinical notes using natural language processing. We used multivariable logistic regression models to estimate the relationship of adverse social determinants of health with UI. Results: In total, 51,422 home health care episodes reflecting 38,311 patients (mean age 82 years) were included. Episodes with documentation about each adverse social determinant of health had significantly higher odds of UI compared with those without such documentation, notably problems with physical environment [adjusted odds ratio (aOR) 1.38, 95% CI 1.31–1.45], problems with economic circumstances (aOR 1.37, 95% CI 1.28–1.47), and food insecurity (aOR 1.22, 95% CI 1.10–1.35). Conclusions and Implications: Adverse social determinants of health, including problems with the physical and social environment, economic circumstances, health care access, low education/literacy, and food insecurity, were each associated with increased odds of UI. These factors may be key targets for interventions aimed at reducing disparities and improving UI and related outcomes for older adult women.