Housing Status and Cancer Screening in US Veterans.

BACKGROUND: Cancer is a leading cause of death in people experiencing homelessness, who are more commonly diagnosed with late-stage disease and have poorer survival after diagnosis than housed.

OBJECTIVE: To characterize the incidence and timeliness of colorectal and breast cancer screening in a national sample of Veterans.

DESIGN: Retrospective, matched cohort study from 2011 to 2021 in a national sample of Veterans receiving care from the Veterans Health Administration (VA).

PARTICIPANTS: Each Veteran experiencing homelessness was matched to three housed Veterans with the same age, gender, clinic location, and month and year of outpatient clinic appointment.

EXPOSURE: We classified Veterans as homeless if they had any homeless indicator at the matched clinic visit or in the 12 months prior and all others as housed.

MAIN MEASURES: Our primary outcomes were being up to date on screening, receiving a biopsy following a positive screen and timeliness of biopsy. We assessed the association between housing status and our outcomes using conditional Poisson regression models with generalized estimating equations, adjusting for race, ethnicity, marital status, Charlson Comorbidity Index, smoking status, and mental health comorbidities.

KEY RESULTS: Our sample included 2,580,640 Veterans, with 1,935,480 housed and 645,160 experiencing homelessness. Patients experiencing homelessness had a 16% lower adjusted incidence rate ratio (aIRR) of being up to date with colorectal cancer screening when compared to housed (aIRR 0.84, 95%CI 0.83-0.84; p<0.001) and a 13% lower aIRR for breast cancer (aIRR 0.87, 95%CI 0.86-0.88; p<0.001). Following a positive stool-based test, patients experiencing homelessness had a 12% lower aIRR of undergoing diagnostic colonoscopy compared to housed (aIRR 0.88, 95%CI 0.84-0.92; p<0.001). Time to biopsy was similar between groups for both cancer types.

CONCLUSIONS: Veterans experiencing homelessness were less commonly screened for cancer than a matched housed cohort. However, screening rates in this group were higher than in non-Veteran homeless populations. The VA system may offer insights into providing preventative care for this population.