A randomized controlled trial of intensive care management for disabled Medicaid beneficiaries with high health care costs.

Disabled Medicaid beneficiaries comprise only 5 percent of the Medicaid population, but they account for over half of all Medicaid spending, primarily due to comorbid chronic physical and mental health conditions, with each condition contributing to their health care costs (Kronick et al. 2000; Kronick, Bella, and Gilmer 2009; Kaiser Commission on Medicaid and the Uninsured 2010). These individuals qualify for Medicaid based on disability after demonstrating that they have an impairment that has prevented them from performing “substantial gainful activity” for at least 1 year and after undergoing a financial asset test to indicate they meet specific income requirements (Decker et al. 2013). Disabled Medicaid beneficiaries typically receive health care in multiple disconnected sites, including emergency departments, hospitals, psychiatric inpatient facilities, community clinics, jails, and substance abuse clinics (New Freedom Commission on Mental Health 2003). Extant literature provides little guidance for providing effective services to disabled beneficiaries; moreover, this group is often excluded from clinical trials due to their complex health conditions, disadvantaged status, and erratic help-seeking (Walkup and Yanos 2005). Consequently, health care providers know least about how to care for patients who need their help most (Walkup and Yanos 2005). With the advent of the Patient Protection and Affordable Care Act (PPACA), many states will expand Medicaid eligibility to include up to 16 million currently uninsured individuals (Holahan, Kennedy, and Pelletier 2010). There is some evidence that uninsured Medicaid-eligible persons with chronic health conditions may be more in need of care than the current Medicaid-covered population, as the conditions of the former are more likely undetected or uncontrolled prior to becoming insured (Decker et al. 2013). In addition, up to 30 percent of uninsured Medicaid-eligible individuals are estimated to have behavioral health and/or substance abuse disorders; furthermore, those with such disorders are more likely than those without to have one or more chronic physical conditions (Kaiser Commission on Medicaid and the Uninsured 2011). The PPACA calls for “health homes” to integrate primary and behavioral health care and to focus on care coordination to provide effective care for new beneficiaries with multiple chronic conditions (Boyd et al. 2010). Care coordination models have shown positive results for patients with depression, anxiety, chronic illness, and for select Medicare populations (Katon et al. 2010; Archer et al. 2012; Brown et al. 2012). However, few published studies have examined outcomes among high-risk disabled Medicaid beneficiaries. One exception employed specially trained community health workers in Arkansas to coordinate home and community-based services for Medicaid-eligible adults with physical disabilities, unmet long-term care needs, and high risk for nursing home admission (Felix et al. 2011). Based on a quasi-experimental design, the intervention reduced annual Medicaid spending by 23.8 percent per participant (Felix et al. 2011). The current study adds to the limited evidence by reporting outcomes of the King County Care Partners (KCCP) Program, a community-based, registered nurse (RN)–led, multidisciplinary care management intervention delivered in King County, Washington, to disabled Medicaid beneficiaries with mental health and/or substance abuse problems who were determined to be at risk for high future health care costs (Lessler, Krupski, and Cristofalo 2011). The KCCP intervention was designed to support beneficiaries in addressing their social and health care needs and to enhance care coordination, communication, and integration of services across safety net providers. By connecting participants to community-based health services, we predicted improved access to ambulatory medical and behavioral health services for KCCP participants. We expected improved access to these services could result in reduced hospitalizations and associated costs (Felix et al. 2011; Brown et al. 2012). By addressing social needs through behavioral and social services including housing, we expected reduced criminal activity (Garnick et al. 2014) and homelessness (McCormack et al. 2013). While a recent systematic review of the effectiveness of intensive care management for homeless individuals indicates mixed results, improvements in housing have been identified among those with mental illness (de Vet et al. 2013). Finally, because the intervention was embedded in an agency that authorized long-term care home and community services, and the study population was likely to have unmet needs for these services, we also expected the intervention to be associated with increased use of long-term care.