Study Links Medicaid Cuts to More Late-Stage Breast Cancer Diagnoses

A study published in Cancer finds an association between Medicaid disenrollment in Tennessee and higher rates of late-stage breast cancer diagnoses and delays to treatment.

As the article explains, Tennessee’s financial pressures led the state to tighten Medicaid eligibility requirements in 2005, resulting in the termination of coverage for almost 170,000 beneficiaries. Although this was the largest disenrollment in Medicaid’s history, little research has been done to assess the potential health outcomes of those who lost coverage.

Researchers set out to investigate the link between Tennessee’s Medicaid disenrollment and the timely diagnosis and treatment of breast cancer, since stage at diagnosis and time to treatment are strong predictors of mortality outcomes. Having health insurance coverage is associated with earlier breast cancer diagnosis and better outcomes, potentially because it allows women to receive mammograms at recommended intervals.

Using data from the Tennessee Cancer Registry, the researchers identified over 19,000 women between 21 and 64 years who had been diagnosed with breast cancer in either the 3 years before (2002 to 2004) or 3 years after (2006 to 2008) the Medicaid disenrollment. They classified the 13,534 women included in the study sample according to their disease stage at diagnosis (in situ/localized vs regional/distant) and their delays in time to surgery or treatment (more than 60 days or more than 90 days after diagnosis).

Statistical analyses compared outcomes before and after the disenrollment for women living in low-income or high-income zip codes while controlling for sociodemographic characteristics. A group of non-Tennessean women diagnosed with breast cancer served as a control group.

Overall, Medicaid disenrollment in Tennessee was associated with a 3.8-percentage point increase in late-stage diagnosis for women in that state compared with the control group. The percentage of cancers diagnosed at a late stage (regional or distant) in Tennessee increased after the disenrollment among both zip code income levels, but the increase was greater for women in the low-income areas. Following disenrollment, both groups of women became more likely to experience delays in treatment, but the rate of change was higher for women in high-income zip codes.

Compared with the changes among women in high-income zip codes, Medicaid disenrollment was associated with an increase in late-stage diagnoses of 3.3 percentage points for women living in low-income zip codes, meaning that after disenrollment, the odds of these women having a late stage of disease at diagnosis increased 12.7%. For women in low-income zip codes, disenrollment was associated with decreases of 1.9 and 1.4 percentage points in having delays of greater than 60 days to surgery and greater than 90 days to treatment, respectively, compared with those in high-income zip codes.

The researchers noted that the women in low-income areas were more likely to depend on Medicaid for coverage of preventive care like mammograms or visits to discuss potential symptoms, which could explain why they were more likely to be diagnosed with a later stage of cancer after disenrollment. Women in high-income areas likely had more regular screenings, leading to earlier detection of cancer, but upon being diagnosed they may have been able to seek a second opinion or visit a surgeon to discuss breast reconstruction, which would have prolonged their time to treatment.

They wrote that Tennessee’s experience disenrolling residents from Medicaid “provides several lessons concerning access to and use of preventive care services and subsequent delays in the diagnosis and treatment of breast cancer,” which will be important for states considering rolling back Medicaid expansion.

An editorial published simultaneously in Cancer discusses the implications of these findings, pointing out that breast cancer cases caught at a late stage are both deadlier for women and costlier to Medicaid. The authors suggested that further research using patient-level data will be necessary to help policymakers weigh the cost savings from Medicaid cuts and the potential consequences for low-income women.