Inpatient Care and Outcomes Among People With Cancer Experiencing Homelessness
Kanan Shah,Patricia Mae G Santos,Lillian A Boe,Justin M Barnes,Anna Tao,C Jillian Tsai,Fumiko Chino
DOI: https://doi.org/10.1001/jamaoncol.2024.3645
2024-09-05
Abstract:Importance: Cancer is a leading cause of death among people experiencing homelessness (PEH) in the US. Acute care settings are important sources of care for PEH; however, the association of housing status with inpatient care remains understudied, particularly in the context of cancer. Objective: To assess whether housing status is associated with differences in the inpatient care of hospitalized adults with cancer. Design, setting, and participants: This cross-sectional study included hospitalized inpatient adults aged 18 years or older diagnosed with cancer who were identified using data from the 2016 to 2020 National Inpatient Sample. Propensity score matching was used to create a cohort of PEH and housed individuals matched according to age, sex, race and ethnicity, insurance type, cancer diagnosis, number of comorbidities, substance use disorder, severity of illness, year of admission, hospital location, hospital ownership, region, and hospital bed size. Matched pairs were identified using a 1:1 nearest neighbor matching algorithm without replacement, accounting for survey weights. Data were analyzed from August 1, 2022, to April 30, 2024. Exposure: Housing status. Main outcomes and measures: The associations of receipt of invasive procedures, systemic therapy, or radiotherapy during hospitalization (primary outcomes) as well as inpatient death, high cost of stay, and discharge against medical advice (AMA) (secondary outcomes) with housing status. Odds ratios and 95% CIs were estimated with multivariable logistic regression, with adjustment for patient, disease, and hospital characteristics of the matched cohort. Results: The unmatched cohort comprised 13 838 612 individuals (median [IQR] age, 67 [57-76] years; 7 329 473 males [53.0%]) and included 13 793 462 housed individuals (median [IQR] age, 68 [58-77] years) and 45 150 (median [IQR] age, 58 [52-64] years) individuals who were experiencing homelessness after accounting for survey weights. The PEH cohort had a higher prevalence of lung (17.3% vs 14.5%) and upper gastrointestinal (15.2% vs 10.5%) cancers, comorbid substance use disorder (70.2% vs 15.3%), and HIV (5.3% vs 0.5%). Despite having higher rates of moderate or major illness severity (80.1% vs 74.0%) and longer length of stay (≥5 days: 62.2% vs 49.1%), PEH were less likely to receive invasive procedures (adjusted odds ratio [AOR], 0.53; 95% CI, 0.49-0.56), receive systemic therapy (AOR, 0.73; 95% CI, 0.63-0.85), or have a higher-than-median cost of stay (AOR, 0.71; 95% CI, 0.65-0.77). Although PEH had lower rates of inpatient death (AOR, 0.79; 95% CI, 0.68-0.92), they were 4 times more likely to be discharged AMA (AOR, 4.29; 95% CI, 3.63-5.06). Conclusions and relevance: In this nationally representative cross-sectional study of hospitalized adults with cancer, disparities in inpatient care of PEH highlight opportunities to promote equitable cancer care in this socioeconomically vulnerable population.