Trends in insurance coverage and out -of -pocket payments for mental health and substance abuse services: An examination of Medical Expenditure Panel Survey data, 1996–2004

Abstract

The research objectives were to examine trends over time in patient out-of-pocket spending and insurance coverage of outpatient mental health and substance abuse visits, and to also assess other factors that may affect patient expenditures as a percentage of total visit spending. Data from 1996-2004 were extracted from the Medical Expenditure Panel Survey, a publicly available, nationally representative household survey for the U.S. civilian non-institutionalized population. The unit of observation was the office-based visit; there were 521,888 events for 65,749 privately insured individuals (476,545 non-MH/SA visits and 45,343 MH/SA visits). The two outcome variables of interest were: (1) proportion visit expenditures paid by the patient out-of-pocket; and (2) a binary outcome of whether the patient paid 100% of the visit expenditures. Fractional logistic regression, a technique applied when the dependent variable is a proportion, was performed in a generalized linear model to examine the first outcome. Logistic regression was used to analyze models with the second outcome. Results. Over 1996-2004, the percent of MH/SA expenses paid by the patient out-of-pocket (%OOP) declined significantly, as was the case with other types of health care visits. An analysis of the binary outcome of 100% OOP (where the patient paid all of the expenses) vs. < 100% OOP, had similar results: over 1996-2004, there was a significant decrease in the likelihood that MH/SA patients would pay 100% of the visit expense; nevertheless, significantly more MH/SA visits than non-MH/SA encounters continue to be paid in full by the patient. Hence, there remains a consistent gap between the lower percentage of visit costs paid for non-MH/SA and the higher proportion paid for mental health and substance abuse visits. The results concerning payment-in-full for MH/SA visits suggest that insurance may be less of a factor in payment for MH/SA services than with non-MH/SA care. Overall, these analyses indicate that insurance coverage in general is improving, while a variety of factors may contribute to a continuing lag in the extent of insurance payment for MH/SA services. Despite improvements in coverage of MH/SA care, there is not genuine parity in private insurance payment for MH/SA and non-MH/SA care.