Correlates of Diagnostic Codes for Documenting Housing Instability: A Citywide Cohort Study.

Abstract

ICD-10 Z codes allow for systematic documentation of people experiencing housing instability within healthcare settings, but little is known about Z code usage patterns in people experiencing housing instability. This analysis performed in 2025 identifies the healthcare utilization patterns and clinical and sociodemographic characteristics associated with documented housing instability in a large urban cohort of people experiencing housing instability from 2016 to 2022. A reference standard for housing instability was created using detailed data from a clinic network serving people experiencing housing instability. Linking this with electronic health record data from 6 other health systems throughout the city of Chicago created a retrospective cohort of housing-insecure patients with ≥1 visit at another participating health system. In addition to descriptive statistics, the assignment (yes/no) of Z59 diagnosis codes is analyzed using a modified Poisson model to estimate rate ratios, controlling for relevant patient-level sociodemographic and clinical factors. People experiencing housing instability with Z59 diagnosis codes had more emergency department and inpatient encounters than those without Z59 codes. Multivariable findings indicated that Z59 code recipients were more likely to be male (rate ratio=1.35; 95% CI=1.25, 1.46) and to experience serious medical and mental health comorbidities. Z59 code recipients were more likely to report living in a shelter (rate ratio=1.31; 95% CI=1.21, 1.42), living on the street (rate ratio=1.88; 95% CI=1.61, 2.20), or receiving outreach care in these settings (rate ratio=1.61; 95% CI=1.49, 1.74). Rates of Z59 code assignment varied among participating sites. Although Z59 codes for housing instability are infrequently used in healthcare settings, they may help identify people experiencing housing instability in greatest need of medical and social interventions.