Association of Coded Housing Instability and Hospitalization in the US

Abstract

ImportanceHousing instability and other social determinants of health are increasingly being documented by clinicians. The most common reasons for hospitalization among patients with coded housing instability, however, are not well understood.ObjectiveTo compare the most common reasons for hospitalization among patients with and without coded housing instability.Design, Setting, and ParticipantsThis cross-sectional, retrospective study identified hospitalizations of patients between age 18 and 99 years using the 2017 to 2019 National Inpatient Sample. Data were analyzed from May to September 2022.ExposuresHousing instability was operationalized using 5 International Classification of Diseases, 10th Revision, Social Determinants of Health Z-Codes addressing problems related to housing: homelessness; inadequate housing; discord with neighbors, lodgers, and landlords; residential institution problems; and other related problems.Main Outcomes and MeasuresThe primary outcome of interest was reason for inpatient admission. Bivariate comparisons of patient characteristics, primary diagnoses, length of stay, and hospitalization costs among patients with and without coded housing instability were performed.ResultsAmong the 87 348 604 hospitalizations analyzed, the mean (SD) age was 58 (20) years and patients were more likely to be women (50 174 117 [57.4%]) and White (58 763 014 [67.3%]). Housing instability was coded for 945 090 hospitalizations. Hospitalized patients with housing instability, compared with those without instability, were more likely to be men (668 255 patients with coded instability [70.7%] vs 36 506 229 patients without [42.3%]; P < .001), younger (mean [SD] age 45.5 [14.0] vs 58.4 [20.2] years), Black (235 355 patients [24.9%] vs 12 929 158 patients [15.0%]), Medicaid beneficiaries (521 555 patients [55.2%] vs 15 541 175 patients [18.0%]), uninsured (117 375 patients [12.4%] vs 3 476 841 patients [4.0%]), and discharged against medical advice (28 890 patients [8.4%] vs 451 855 patients [1.6%]). The most common reason for hospitalization among patients with coded housing instability was mental, behavioral, and neurodevelopmental disorders (475 575 patients [50.3%]), which cost a total of $3.5 billion. Other common reasons included injury (69 270 patients [7.3%]) and circulatory system diseases (64 700 patients [6.8%]). Coded housing instability was also significantly associated with longer mean (SD) hospital stays (6.7 [.06] vs 4.8 [.01] days) and a cost of $9.3 billion. Hospitalized patients with housing instability had 18.6 times greater odds of having a primary diagnosis of mental, behavioral, and neurodevelopmental disorders (475 575 patients [50.3%] vs 4 470 675 patients [5.2%]; odds ratio, 18.56; 95% CI, 17.86 to 19.29).Conclusions and RelevanceIn this cross-sectional study, hospitalizations among patients with coded housing instability had higher admission rates for mental, behavioral, and neurodevelopmental disorders, longer stays, and increased costs. Findings suggest that efforts to improve housing instability, mental and behavioral health, and inpatient hospital utilization across multiple sectors may find areas for synergistic collaboration.

Sociodemographic and Clinical Predictors of Prolonged Length of Corneal Ulcer Hospitalizations

“…Housing insecurity has been previously implicated in higher rates of being lost to follow-up in patients with diabetic retinopathy and higher risk for treatment barriers such as medication noncompliance and delayed time to resolution in patients with infectious keratitis . In a previous large database study, the authors investigated the impact of housing insecurity on all hospitalizations in the US and found a trend toward longer LOS, although these results did not reach significance.…”

Association of Housing Status with Types of Operations and Post-Operative Healthcare Utilization

“…Z codes are markedly underutilized with studies consistently showing that under 1% of claims contain Z codes related to housing issues and likely are more specific rather than sensitive for housing concerns. [29][30][31] Screening questions, though also imperfect measures, have been shown to have better area under the curve scores than Z codes for housing. 32 More complete documentation of housing status, including housing instability, may facilitate a more nuanced analysis of drivers of disparities.…”

“…Other literature has used Z codes to classify patients as unhoused. Z codes are markedly underutilized with studies consistently showing that under 1% of claims contain Z codes related to housing issues and likely are more specific rather than sensitive for housing concerns 29–31 . Screening questions, though also imperfect measures, have been shown to have better area under the curve scores than Z codes for housing 32 .…”

Morbidity and Length of Stay After Injury Among People Experiencing Homelessness in North America

“…[6][7][8][9][10] Traumatic injury is the second leading reason for hospitalization among people experiencing homelessness. 11 A previous study 9 demonstrated that injured people experiencing homelessness are more likely to be admitted to the hospital than housed individuals. However, little is known about their subsequent hospital course and the incidence of morbidity, surgical procedures, and intensive care unit (ICU) admissions.…”

“…Traumatic injury is the second leading reason for hospitalization among people experiencing homelessness . A previous study demonstrated that injured people experiencing homelessness are more likely to be admitted to the hospital than housed individuals.…”