Duke 2008 | ASHE | Equity and Efficiency in Health and Healthcare

Of the 45 million Americans in Medicaid, roughly two-thirds are enrolled in some sort of managed care. Generally, these Medicaid participants are enrolled in managed care organizations (MCOs) that are responsible for providing all of their health care needs. In return, the MCOs receive from the state a monthly payment for each enrollee, regardless of the costs incurred by the MCO in providing their care. A relatively unanswered question is whether the quality of care provided by Medicaid MCOs is influenced by changes in their profits. Given the capitated nature of their revenue, these MCOs may face pressure to respond to poor profits by reducing costs, perhaps by lowering the quality of care that they provide. As such, the well being of Medicaid MCO enrollees may be adversely affected by changes in their MCO's financial condition. While this issue could potentially have a large aggregate impact, to the author's knowledge only two studies have investigated this possibility. Furthermore, both of these papers examined commercial, rather than Medicaid, MCOs.

This paper employs a unique data set to address this question. The data contain measures of both quality of care and profits for eight Medicaid MCOs in a large, diverse state. Eight HEDIS quality of care measures are used to evaluate the degree of preventative care received by enrollees, specifically in the context of child health screenings, female health screenings, and the provision of asthma medications.

These measures are then matched to the level of profits earned by the MCO. The quality and profit measures are measured on a quarterly basis over a three-year period for each MCO in each of the metropolitan areas in which it operates. Additional explanatory variables are also employed to control for demographic factors believed to potentially influence quality of care. A number of econometric complications are addressed in the empirical specification. First, the HEDIS measures are relatively slow-moving, so the model is estimated in first differences. Second, the profit measure is lagged in the specified model. Not only does this reflect the likely lagged response by MCOs to shocks in profits, but lagging the profit variable also mitigates concerns regarding endogeneity due to reverse causality. Third, by differencing the data and including quarter fixed effects, the estimates control for any factors that do not vary over the period for a given MCO in a given metropolitan area and for any state-wide factors in a given quarter. Preliminary regression results indicate that changes in profits earned by an MCO do not influence the level of preventive care that it provides. For each of the eight HEDIS measures, the effects of profits are both economically and statistically insignificant. However, changes in the racial composition of the MCO's enrollees appear to have an effect. Taken together, the results suggest that while incentives included in MCO contracts may help prevent service stinting, policies that address discrepancies across racial lines may be needed to help ensure proper care for all Medicaid participants.