Healthcare costs associated with homelessness in Houston, TX

Background: Because of poor living conditions, deprivation and barriers to health care, homeless populations have higher rates of illness and mortality. However, very little information is available about healthcare costs for the homeless due to the difficulty of studying homeless populations and information systems often do not specifically identify housing status. This study calculates the hospital healthcare costs for homeless and other non-homeless low income patients at the Harris County Hospital District. Methods: We obtained data on 532 admissions for 310 homeless patients and 4836 admissions for 4238 domiciled patients treated by the Harris County Hospital District over a one year period. All patients were over 18 years of age. Descriptive statistics compared the demographic characteristics of homeless vs domiciled patients. The economic analysis was done at two levels: admission and patient (given some patients had more than one admission). Mean comparisons with t tests and non parametric tests were used to compare costs in homeless and domiciled patients. Further analyzes will include log linear multivariate regressions that control for age, gender, race/ethnicity and diagnoses. We will also analyze the frequency and costs of preventable hospitalizations in the homeless and domiciled patients. Results: Compared to domiciled patients, homeless patients were more likely to be male (77% vs 41% ), African American (55% vs 26%), White (29% vs 13%) and older (mean age 48.39 vs 42.70); they were less likely to be Hispanic (15% vs 56%). Average costs per admission are similar for homeless and domiciled patients ($28,387 vs $27,593; p value =0.59), despite homeless patients having longer LOS (7.36 vs 5.88). Homeless patients have more admissions (1.43 vs 1.15) and therefore have higher total hospital costs ($40,494 vs $31,973) as well as higher total LOS (10.43 vs 6.86) during the one year time period. Conclusions: These preliminary findings suggest that homeless patients have higher overall hospital costs, longer LOS per admission, and more hospitalizations than domiciled patients.