|
Article Excerpt
About 125 million Americans live with one or more chronic conditions, a number that is expected to increase dramatically over the next several decades (Anderson and Horvath 2002). Among people aged 65 and older, most health care spending (96 percent) is for the treatment and management of chronic disease (Kane 1998). Although those with chronic conditions benefit from comprehensive support and continuity of care (Rothman and Wagner 2003), health care tends to be fragmented across different settings, with variations in eligibility, regulations, management, financing, and reimbursement. Described by the Institute of Medicine (2001) as a "nightmare to navigate," these "siloed" services make it difficult for those with chronic conditions to obtain the care they need.
Although managed care offers the promise of improved coordination and delivery of chronic care (Kane 1998), studies of the experience of older adults suggest disappointing health outcomes (Ware et al. 1996) and consumer satisfaction (Landon et al. 2004; Pourat, Kagawa-Singer, and Wallace 2006). Nevertheless, several promising practices have emerged. For example, the Program of All-Inclusive Care for the Elderly (PACE) that integrates acute and long-term care has demonstrated reduced utilization of hospitals and nursing facilities, high satisfaction, and improved quality of life (Nadash 2004; Friedman et al. 2005). A Kaiser Permanente program in Colorado offering prevention, education, and comprehensive management of acute and primary care improved patient satisfaction and reduced health care utilization (Kane 1998). Project IMPACT, which treated depressed older adults using education, patient support, and a multidisciplinary team significantly reduced depression and improved patient satisfaction (Unutzer et al. 2002). These findings suggest that outcomes improve when managed care offers patient-oriented, flexible chronic care services that promote consumer autonomy and choice in health care decision making (Davis et al. 1995). Interventions not specific to managed care that promote education, choice, and control include web-based access to health and social care information (Shugarman, Nishita, and Wilber 2006), "cash and counseling" type programs that provide eligible consumers greater discretion in purchasing chronic care services (e.g., personal care) with professional decision-making support (Benjamin 2001), and comprehensive geriatric assessment (Kane and Kane 2000). While web-based approaches have yet to be systematically evaluated, cash and counseling programs have improved satisfaction and cost-effectiveness compared with agency directed models of personal care support (Phillips and Schneider 2002) and geriatric assessment has reduced nursing home placement compared with usual care (Phibbs et al. 2006).
The Care Advocate Program (CA Program), the focus of this article, bridged medical and social care delivery systems using telephone-based care management to coordinate health and long-term care services for chronically ill older adults. Part of the Program for Elders in Managed Care Initiative, the CA Program was designed to improve care for managed care members by helping them link to noninsured home- and community-based services (HCBS) and reconnect with health plan services when needed. An evaluation using a randomized control trial found that the intent-to-treat group (ITT) had reductions in utilization of insured medical services, with no change in satisfaction and a decrease in retention (Shannon, Wilber, and Allen 2006) The evaluation also found an unexpected reduction in mortality (Wilber and Shannon 2003). The goal of this analysis is to systematically examine and test this unexpected result.
Attempts to identify mortality risk have revealed its complexity, as multiple factors may be involved including sociodemographic characteristics, presence and severity of disease, mental and physical impairments, health service utilization, and health care service delivery structures (Wolinsky, Johnson, and Stump 1995; Crimmins 2004; Seeman et al. 2004). Researchers have postulated that community-based care management in partnership with medical services could reduce mortality outcomes if goal-focused treatment protocols and targeting strategies were grounded in empirical data on mortality risk (Schore, Brown, and Cheh 1999; Miller and Weissert 2000). However, the effectiveness of these models to affect mortality outcomes is mixed. Several programs in which care management was a central component resulted in reduced mortality (Chatterji et al. 1998; Shapiro and Taylor 2002; Albert et al. 2005). Other studies found that community-based medical interventions with a care management component had improved outcomes such as decreased incidence of heart disease (Luepker et al. 1996), improved care of chronic conditions (Bodenheimer, Wagner, and Grumbach 2002), and greater self-management of chronic diseases (Lorig et al. 2001), with no reductions in mortality. Similarly, studies of medically focused models in managed care and hospital settings linked with social care systems showed improvements in geriatric assessment in emergency rooms (Miller et al. 1996), and reductions in inpatient hospital days (Leveille et al. 1998), nursing home placement (Fischer et al. 2003), and health care costs (Boult et al. 2000), but no reductions in mortality.
CA Program Model
The CA Program offered telephone-based care management to older adults with high health care utilization enrolled in a Medicare managed care health plan. A health care utilization algorithm, described below in the Methods section, was created to identify members who had multiple needs based on health care utilization in the previous year. The CA Program used master's level social work care managers called "care advocates" geographically located in and employed by two community-based social service agencies. These agencies had no prior or ongoing financial relationship with the health plan or the four participating medical groups beyond the externally funded project. Standardized instruments and protocols and monthly coordination meetings were used to ensure uniformity across sites (Wilber et al. 2003; Shannon, Wilber, and Allen 2006). The term "care advocate" was used to denote the role of educator, consultant, and coach (Kunkel, Duffy-Durham, and Scala 2000), and to emphasize consumer choice and a mutual partnership toward identified goals.
To identify immediate and long-term service needs, care advocates completed an 83-item psychosocial and functional assessment with participants, which was used to discuss options and link participants to HCBS not covered by the health plan's insured services. Examples of HCBS referral types were in-home care, nutrition, home safety, transportation, noninsured adaptive equipment, and supportive services. Care advocates also referred participants back to their medical group via the...
|
