Despite lower fracture rates, homeless patients have worse outcomes after geriatric ground-level falls: an analysis of National Trauma Data Bank.

OBJECTIVES

To evaluate differences in clinical outcomes between housed and homeless older adults hospitalized for ground-level falls.

DESIGN

Retrospective cohort.

SETTING

A nationwide study in the United States using the National Trauma Data Bank (NTDB).

PATIENTS/

PARTICIPANTS

The study identified 1136 homeless and 325,267 housed patients. Patients aged ≥50 years hospitalized for ground-level falls were identified using ICD-10 external cause codes. Housing status was defined through home residence variable.

INTERVENTION: None.

MAIN OUTCOME MEASUREMENTS: The primary outcomes were fracture rate and length of stay (LOS) ≥7 days; secondary outcomes included ICU admission, complications, injury pattern, in-hospital mortality, and discharge location.

RESULTS

Homeless patients were significantly younger (mean age 65.3 vs 74.6 years, P < 0.0001) and had lower fracture rates (51.2% vs 59.3%, P < 0.0001). Rates of femoral neck, femoral shaft/distal end, humeral, and distal radial fracture were significantly lower in the homeless population. Despite lower fracture rates, homeless individuals had worse outcomes: longer mean LOS (8.82 vs 6.63 days, P < 0.0001), higher rates of ICU admission (25.3% vs 19.8%, P < 0.0001), and complications (11.5% vs 6.3%, P < 0.0001). Homelessness was independently associated with prolonged LOS (odds ratio 1.55, 95% confidence intervals 1.37-1.76, P < 0.0001).

CONCLUSIONS

Homeless patients had lower fracture rates, potentially reflecting differences in age. Despite sustaining fewer fractures, homelessness was independently associated with significantly worse outcomes, even after controlling for age, comorbidities, and injury severity. These findings highlight significant disparities driven by homelessness and underscore the need for integrated care models addressing both medical and social determinants of health.

LEVEL OF EVIDENCE: Level III.