Association of Medicaid Privatization With Patient Cancer Outcomes.

Abstract

Increasingly, states outsource administration of Medicaid insurance to privately administered Medicaid managed care organizations. However, on January 1, 2012, Connecticut transitioned from a privately to publicly administered Medicaid system. New Jersey retained a private model. Our objective was to assess rates of early-stage cancer diagnosis and cancer survival in two states with similar sociodemographic characteristics but differing exposures to Medicaid privatization. Using data from the SEER Program between 2007 and 2016, Connecticut and New Jersey Medicaid patients with 10 common solid cancers including breast, lung, colorectal, prostate, kidney, bladder, cervix, uterus, head and neck cancer, and melanoma were included. A difference-in-differences analysis of stage of cancer presentation and cancer survival in Connecticut (intervention) was compared with New Jersey (control). Among 29,328 patients (14,424 patients from Connecticut and 14,904 patients from New Jersey) parallel trends were verified in early cancer diagnosis and survival for both states under privately administered Medicaid (pre-exposure). Connecticut's transition from privately to publicly administered Medicaid was associated with an adjusted 4.0% increase in overall early-stage cancer diagnosis (95% CI, +1.7% to +6.2%) and a 4.7% increase in early-stage cancer diagnosis for cancers with US Preventive Services Taskforce A/B recommendations for cancer screening (95% CI, 1.6% to 7.8%). Public administration of Medicaid was also associated with improved overall survival after cancer diagnosis (hazard ratio, 0.92 [95% CI, 0.85 to 0.99]). No changes were observed in New Jersey. Transition from private to public administration of Medicaid in Connecticut was associated with earlier-stage cancer diagnosis and improved cancer survival.