Porthole
Using Marginal Structural Models to Estimate the Effect of the Maryland Medicaid Health Home Waiver on Health Care Utilization Among Individuals with Serious Mental Illness (307852)
*Sachini Bandara, Johns Hopkins Bloomberg School of Public HealthColleen Barry, Johns Hopkins Bloomberg School of Public Health
Gail Daumit, Johns Hopkins School of Medicine
Alene Kennedy-Hendricks, Johns Hopkins Bloomberg School of Public Health
Beth McGinty, Johns Hopkins Bloomberg School of Public Health
Elizabeth Stuart, Johns Hopkins Bloomberg School of Public Health
Keywords: marginal structural model, serious mental illness, Medicaid, care coordination
The Maryland Medicaid health home program, established through the Affordable Care Act, integrates primary care services into specialty mental health programs for adults with serious mental illness (SMI). We evaluated the effect of this waiver program on all-cause, somatic, and behavioral health emergency department (ED) and inpatient utilization. To account for time-varying confounding introduced by rolling enrollment into the program, we used a marginal structural model estimated with inverse probability of treatment weighting. We analyzed Maryland Medicaid administrative claims data for 12,203 enrollees with SMI over a 39-month period (October 1, 2012-December 31, 2016); 3,293 individuals were enrolled in a BHH and 8,911 were never enrolled. Health home enrollment was associated with reduced probability of all-cause (23% vs. 25% within 3 months, p<0.01) and somatic ED visits (21% vs. 24%, p<0.01). We discuss study implications, as well as opportunities and challenges of broader use of marginal structural models in health policy.