Impact of social determinants of health on fecal incontinence treatment in older women.

BACKGROUND: Many women with fecal incontinence do not seek care despite the availability of effective treatments. Factors influencing care-seeking for fecal incontinence are not well elucidated, and the role of social determinants of health in fecal incontinence treatment utilization is unknown. OBJECTIVE: Our primary aim was to determine the association between social determinants of health and treatment utilization among Medicare beneficiaries with fecal incontinence. We secondarily aimed to determine the baseline treatment utilization rate and to determine factors associated with fecal incontinence treatment utilization. STUDY DESIGN: We conducted a retrospective cohort study of Medicare beneficiaries with fecal incontinence based on 2010 to 2018 claims data from a 5% national sample. Women with fecal incontinence were identified by diagnosis codes, and those receiving treatment were identified by Current Procedural Terminology codes for pelvic floor physical therapy with biofeedback, sacral neuromodulation, anal sphincteroplasty, percutaneous tibial nerve stimulation, and anal procedures. Comorbidity was assessed via the Charlson comorbidity index. Social determinants of health were defined by the Social Vulnerability Index, a census-based score accounting for factors such as socioeconomic status, disability, ethnicity, language, housing type, and transportation by county. Social Vulnerability Index is reported as a percentile rank, with higher percentiles reflecting greater vulnerability. Additional social determinant of health variables analyzed included Medicaid dual eligibility status, per capita income, and proportion of the population below poverty level. The association between social determinants of health and treatment for fecal incontinence was evaluated using Cox proportional hazards models. RESULTS: We identified 33,010 women with a diagnosis of fecal incontinence, of whom 3160 (9.6%) underwent treatment. Treatment modalities included anal procedures (6.5%), sacral neuromodulation (2.4%), percutaneous tibial nerve stimulation (0.9%), anal sphincteroplasty (0.4%), and pelvic floor physical therapy with biofeedback (0.1%). Those who did not undergo treatment were older, more commonly Medicaid dual eligible, had lower per capita incomes, higher poverty rates, and higher Charlson comorbidity index scores (all P<.01, Table 1). Higher Social Vulnerability Index scores (hazard ratio, 0.88; 95% confidence interval, 0.79-0.97), Medicaid dual eligibility (hazard ratio, 0.45; 95% confidence interval, 0.39-0.52), and residence in high-poverty counties (hazard ratio, 0.82; 95% confidence interval, 0.74-0.9) were associated with lower likelihood of treatment, whereas higher income was associated with greater likelihood of treatment (hazard ratio, 1.44; 95% confidence interval, 1.3-1.59). The association between treatment and Medicaid dual eligibility (hazard ratio, 0.91; 95% confidence interval, 0.82-1.01), income (hazard ratio, 1.41; 95% confidence interval, 1.27-1.56), and poverty rate (hazard ratio, 0.86; 95% confidence interval, 0.78-0.95) persisted after accounting for patient characteristics; the association between Social Vulnerability Index and treatment did not. Increasing age (hazard ratio, 0.96; 95% confidence interval, 0.96-0.97), Black race (hazard ratio, 0.82; 95% confidence interval, 0.7-0.97), higher Charlson comorbidity index (hazard ratio, 0.65; 95% confidence interval, 0.06-0.70), depression (hazard ratio, 0.66; 95% confidence interval, 0.53-0.81), immobility (hazard ratio, 0.36; 95% confidence interval, 0.22-0.61), and loose stools (hazard ratio, 0.87; 95% confidence interval, 0.79-0.94) were associated with lower treatment receipt, whereas urinary incontinence (hazard ratio, 1.71; 95% confidence interval, 1.57-1.85) and constipation (hazard ratio, 1.29; 95% confidence interval, 1.19-1.40) were associated with higher likelihood of treatment. CONCLUSION: Treatment utilization among women with fecal incontinence is low even within an insured population. In addition to comorbid conditions, social factors reflecting social disadvantage are associated with lower treatment utilization. Future efforts to increase treatment utilization should target this vulnerable group of women.

Duke Scholars

Author Jonathan Charles Routh Urology