Association of Medicaid enrollee characteristics and primary care utilization with cancer outcomes for the period spanning Medicaid expansion in New Jersey.

BACKGROUND: Cancer outcomes for Medicaid enrollees may be affected by patients' primary care (PC) utilization and complex Medicaid enrollment dynamics, which have recently changed for many states under the Affordable Care Act.
METHODS: With New Jersey State Cancer Registry and linked Medicaid claims data, a retrospective cohort study was conducted for patients with incident breast, colorectal, or invasive cervical cancer (aged 21-64 years) diagnosed in 2012-2014. Associations of Medicaid enrollment factors and PC utilization with the stage at diagnosis and treatment delays were examined with multivariate logistic regression models.
RESULTS: The study included 19,209 total cancer cases and 3253 linked Medicaid cases. Medicaid cases were more likely to be diagnosed at a late stage and to experience treatment delays in comparison with non-Medicaid cases. In adjusted analyses, Medicaid cases with 1 or more PC visits before the diagnosis had lower odds of a late-stage diagnosis (odds ratio, 0.47; 95% confidence interval, 0.33-0.67) in comparison with Medicaid cases with no outpatient visits. New enrollees (<6 months) and longer term enrollees in fee-for-service (FFS) Medicaid had greater odds of a late-stage diagnosis and treatment delays in comparison with those in Medicaid managed care.
CONCLUSIONS: Medicaid patients with cancer diagnosed just before and in the initial year of eligibility expansion had worse outcomes than non-Medicaid cases. Poor outcomes were especially pronounced among new enrollees, those without outpatient visits before their diagnosis, and FFS enrollees. Targeted strategies to enhance care continuity, including access to PC providers before the diagnosis and a better understanding of pathways to cancer care upon Medicaid enrollment, are needed to improve outcomes in this population.