The Association Between Federal Housing Assistance and Cancer Stage at Diagnosis in Older Adults

A new large observational study suggests that federal housing assistance may be tied to earlier detection of certain cancers among older Americans. In an analysis of Medicare beneficiaries diagnosed between 2007 and 2019, researchers found that those receiving Department of Housing and Urban Development (HUD) assistance were modestly less likely to be diagnosed at distant (late) stage for breast, colorectal, and non–small cell lung cancers — though not for prostate cancer. 

Investigators linked data from the SEER (Surveillance, Epidemiology, and End Results) cancer registries, Medicare claims, and HUD records to assemble a cohort of 52,532 individuals aged 66 to 95 years who were receiving HUD assistance at diagnosis. They matched each assisted individual to similar individuals not receiving assistance (controls) using propensity scoring to balance demographic, insurance, geography, and neighborhood-level factors. After matching, comparisons focused on the likelihood of being diagnosed at localized, regional, or distant stage (or in situ, for breast cancer).

Among women with breast cancer, those with housing assistance had lower odds of being diagnosed with regional disease (adjusted odds ratio [aOR] 0.86, 95% CI 0.81–0.93) or distant disease (aOR 0.85, 95% CI 0.82–0.90) relative to localized disease, compared to unassisted counterparts. For colorectal cancer, assisted individuals had lower odds of distant-stage diagnosis (aOR 0.90, 95% CI 0.83–0.98). The association was strongest for non–small cell lung cancer, where housing assistance was associated with a 17 % lower odds of distant-stage diagnosis (aOR 0.83, 95% CI 0.79–0.86). There was no statistically significant association between housing assistance and stage at diagnosis for prostate cancer.

In secondary analyses, the associations with earlier-stage diagnosis held across subgroups defined by race, ethnicity, Medicare plan type, neighborhood socioeconomic status, and calendar period. The strength of the association varied by type of housing assistance: for example, both multifamily housing subsidies and Housing Choice vouchers were linked to lower odds of distant-stage breast cancer. Sensitivity checks adjusting for comorbidity or using alternative staging definitions did not materially change the results.

To approximate population-level impacts, the authors estimate that broader access to housing assistance among low-income older renters might avert about 378 cases of regional or distant-stage breast, colorectal, or lung cancer, with first‑year cancer care cost savings around $15 million. They note, however, that those financial gains are modest relative to the overall federal investment in housing assistance and likely underestimate long-term benefits.

The authors acknowledge limitations, including possible underidentification of HUD assistance, unmeasured confounding (for example, individual income or health behaviors), and restricted generalizability beyond SEER regions and older adults. They caution that the observational nature of the study precludes causal inference, but suggest the findings support further investigation into housing security as a social determinant of cancer outcomes.

If confirmed in other settings, these associations imply that policies improving housing affordability and stability may help reduce delays in cancer diagnosis and contribute to narrowing disparities in cancer outcomes — especially among older adults with constrained resources.