Homelessness Tied to Worse Cancer Outcomes Among Veterans

Homelessness was associated with diagnosis of colorectal cancer at later stages and increased rates of all-cause mortality for those with lung and colorectal cancers, according to a retrospective study of veterans.

In a cohort of over 100,000 patients who received care from the Department of Veterans Affairs (VA) health system, those experiencing homelessness were more likely to be diagnosed with stage IV colorectal cancer compared with those who had housing (22% vs 19%; P=0.02), reported Hannah C. Decker, MD, of the University of California San Francisco, and colleagues.

Patients experiencing homelessness also had longer mean lengths of hospital stay for all cancer types studied, which included lung, colorectal, and breast cancers, they noted in JAMA Network Open.

Notably, unhoused patients also had higher rates of all-cause mortality 3 months after diagnosis of both lung (adjusted HR 1.1, 95% CI 1.1-1.2, PP

However, Decker and colleagues also pointed out that these findings differed from those seen in other healthcare settings, which reported 1.7 to 2.3 times the rate of lung cancer mortality and 1.4 to 1.7 times the rate of colorectal cancer mortality in patients experiencing homelessness.

Thus, the findings in this study “suggest that there may be important systems in the VA that could inform policy to improve oncologic outcomes for patients experiencing homelessness,” they wrote.

In a commentary accompanying the study, Maria C. Raven, MD, MPH, also of the University of California San Francisco, said there were several factors that help explain how the VA is able to provide more equal cancer outcomes compared with other health systems in the U.S.

For example, Raven noted that insurance is not an access barrier within the VA, and that the VA has a longstanding, nationally integrated electronic health record that reduces disparities when comparing unhoused with housed veterans.

Raven also pointed out that a VA housing initiative that began in 2009 enabled outreach workers to interact with veterans while they were homeless to help coordinate medical care.

“While the [VA] system is not perfect, based on existing research this study shows it outperforms other systems of care,” she wrote. “Reducing healthcare disparities for people experiencing homelessness is essential, and the [VA] can serve as an example of what it takes to do so.”

Decker and colleagues said that prior research evaluating the impact of homelessness on cancer treatment has focused on screening, but cancer outcomes have been understudied in patients experiencing homelessness.

Using data from the VA to characterize the diagnosis, treatment, surgical outcomes, and mortality in patients with cancer who have experienced homelessness is useful, they noted, since the agency “has a robust data infrastructure and has recorded housing status and addressed homelessness among beneficiaries for over a decade.”

In this study, Decker and colleagues identified all U.S. veterans diagnosed with lung, colorectal, or breast cancer who received VA care from October 2011 through September 2020.

Of the sample of 109,485 veterans, 68% had lung cancer, 26% had colorectal cancer, and 6% had breast cancer. Mean age was 68.5, 92% were men, and 79% were white. Of these veterans, 5% were experiencing homelessness at the time of cancer diagnosis.

They observed that a much larger proportion of veterans who were homeless were Black compared with those who had housing (37% vs 17%), suggesting there could be an interaction between race and housing status that was associated with overall outcomes, and “should be explored in further studies.”

In addition, veterans experiencing homelessness were diagnosed with cancer at younger ages than those with housing, “which may prompt clinicians to have a higher index of suspicion for malignant neoplasm in younger patients with appropriate symptoms who are experiencing homelessness.”

There were no statistically significant differences in stage at diagnosis in lung or breast cancer patients by housing status. While homelessness was associated with increased rates of all-cause mortality for patients with lung and colorectal cancers, there was no such association with breast cancer (HR 1.2, 95% CI 0.8-1.8).