Supportive Housing and Its Relationship With Diabetes Diagnosis and Management Among Homeless Persons in New York City

• Published in: American journal of epidemiology 2019-06, Vol.188 (6), p.1120-1129
• Main Authors: Lim, Sungwoo, Miller-Archie, Sara A, Singh, Tejinder P, Wu, Winfred Y, Walters, Sarah C, Gould, L Hannah
• Format: Article
• Language: English
• Subjects: Adults; Aged; Comorbidity; Confidence intervals; Continental Population Groups; Diabetes; Diabetes mellitus; Diabetes Mellitus - diagnosis; Diabetes Mellitus - epidemiology; Diabetes Mellitus - therapy; Female; Glycated Hemoglobin A; Government programs; Health risks; Health Status; Hemoglobin; Homeless people; Homeless Persons - statistics & numerical data; Homelessness; Housing; Humans; Lipids; Lipids - blood; Male; Management services; Medical diagnosis; Mental Disorders - epidemiology; Middle Aged; New York City - epidemiology; Patient Acceptance of Health Care - statistics & numerical data; Placement; Public Housing - statistics & numerical data; Risk management; Statistical analysis
• ISSN: 0002-9262; 1476-6256
• DOI: 10.1093/aje/kwz057

Supportive housing addresses a fundamental survival need among homeless persons, which can lead to reduced risk of diabetes mellitus and improved diabetes care. We tested the association between supportive housing and diabetes outcomes among homeless adults who were eligible for New York City's supportive housing program in 2007-2012. We used multiple administrative data sources, identifying 7,525 Medicaid-eligible adults. The outcomes included receiving medical evaluation and management services, hemoglobin A1C and lipid testing (n = 1,489 persons with baseline diabetes), and incidence of new diabetes diagnoses (n = 6,036 persons without baseline diabetes) in the 2 years postbaseline. Differences in these outcomes by placement were estimated using inverse-probability-of-treatment weighting. Placed persons were more likely to receive evaluation and management services (relative risk (RR) = 1.03, 95% confidence interval (CI): 1.01, 1.04) than unplaced persons. For those with baseline diabetes, placed persons were more likely to receive hemoglobin A1C tests (RR = 1.10, 95% CI: 1.02, 1.19) and lipid tests (RR = 1.09, 95% CI: 1.02, 1.17). For those without baseline diabetes, placement was also associated with lower risk of new diabetes diagnoses (RR = 0.87, 95% CI: 0.76, 0.99). These findings show that benefits of supportive housing may be extended to diabetes care and prevention.