Quality of Care for Chronically Ill Children and Medicaid Managed Care in Georgia

Background: States are under pressure to contain costs in Medicaid. These pressures will intensify if Congress enacts recent proposals to block-grant or cap federal funding for the program. One step most states have taken to reduce Medicaid costs is to administer benefits through privately operated Medicaid Managed Care (MMC) insurance plans. To achieve savings, MMC plans employ various cost containment tools, such as prior authorization, preferred drug lists, and restrictive provider networks. These tools impose (1) differential transaction costs across treatment options on clinicians, potentially deterring them from providing preferred treatments; and (2) out-of-pocket costs (e.g., non-covered prescriptions) and travel costs (e.g., referrals to more distant specialists) on patients, who may not adhere to recommended therapies. Consequently, beneficiaries’ quality of care may be reduced. Evidence on the effects of MMC cost containment tools on quality of care is scattershot because information about these tools’ implementation is often proprietary.

Setting: In Georgia, three-quarters of Medicaid beneficiaries are enrolled in MMC plans. Through September 2017, both Medicaid and Peachcare for Kids (Georgia’s Children’s Health Insurance Program) were administered by only three MMC plans, fewer than in most states. Most of Georgia’s 34 federally qualified health center (FQHC) organizations regularly see patients enrolled in each of the three plans.

Data: We are fielding a survey of physicians and nurse practitioners regularly delivering care at FQHCs to children and adolescents, ages 1-18, with asthma or attention deficit hyperactivity disorder (ADHD). The survey will capture information on: difficulties encountered when making referrals and prescribing medications for these chronically ill patients due to MMC plans’ tools; the quality of care respondents perceive their patients receive; how effectively FQHCs manage MMC plans’ administrative requirements; and other FQHC and clinician characteristics. Respondents are asked to reflect on their experiences with MMC patients overall and for each of Georgia’s three MMC plans separately.

The survey, which will be completed by January 2018, will be administered in person (Georgia FQHC convention) and using both online and paper (mail) forms to maximize the response rate. Assuming a 70% response rate (we are following a three-part financial incentive design that has achieved response rates of 75-80% in other clinician samples), we anticipate receiving approximately 85 completed surveys.

Analysis: We will test whether difficulties attributed to MMC plans’ cost containment tools are associated with perceived quality of care for chronically ill child beneficiaries and whether FQHC efforts to manage plan requirements mitigate this relationship. We will first compute simple Pearson correlation coefficients among these factors. We will subsequently estimate generalized ordered logit models of perceived quality as a function of MMC plan tools, FQHC efforts to manage MMC plan requirements, and other factors. We will compute robust (Huber-White) standard errors. We will supplement our interpretation of survey findings through qualitative interviews with select clinicians affiliated with the Georgia Primary Care Association.

Implications: Understanding how MMC beneficiaries’ quality of care is related to plans’ cost containment tools is critical for empowering states to achieve cost savings in their MMC programs without compromising beneficiaries’ outcomes.