The Effect of a Disease Management Intervention on Quality and Outcomes of Dementia Care

Context Dementia is an incurable chronic disease, but assistance to caregivers can reduce the severity of patients' symptoms and delay institutionalization. Because this assistance requires provision of multiple health care and social services, patients and caregivers might benefit from a coordinated system of care. Contribution The investigators randomly assigned patients with dementia and their caregivers to usual care or to a coordinated system of care. In the coordinated system, care managers regularly assessed patient and caregiver pairs and coordinated guideline-recommended provision of services by health care providers and community agencies using computerized information systems. The study found that pairs cared for in the coordinated system received higher-quality health care and more needed assistance than those who received usual care. Cautions The study focused on a relatively homogeneous population of white, well-educated, otherwise healthy, noninstitutionalized patients with health insurance. The findings might not apply to other populations. Implications The quality of care for patients with dementia and their caregivers can be improved with a model of care in which services provided by the health system and community agencies are coordinated by a care manager. The Editors Dementia has enormous health and financial consequences for affected individuals, their family caregivers, and society (15). Although most dementia is currently neither preventable nor reversible, existing practice guidelines reflect evidence from randomized, controlled trials that caregiver assistance and support can delay institutionalization (68), and nonpharmacologic management of problem behaviors and depression can reduce symptom severity and improve patient health (9, 10). Yet, adherence to published dementia care guidelines (1114) is poor (15), including inappropriate use of psychoactive medications for initial management of agitation (16); low referral rates to community agencies (17, 18); and underdetection of elder abuse, for which cognitive impairment is a risk factor (19), and of depression (20), a common comorbid condition (21). There are few controlled trials of improvement interventions for dementia care. An opinion leader intervention improved neurologists' reported adherence to dementia care guidelines (18), and a recent trial using advanced practice nurses within primary care improved care quality (22); other trials reported limited impacts (23, 24). All studies were conducted primarily within or outside of health care delivery systems; to date, trials of comprehensive, guideline-based dementia care interventions that incorporate all chronic care model components, including substantive collaboration between health systems and community agencies (25, 26), have not been reported. Yet, this would seem to be a well-suited approach because of the multifaceted nature of dementia care management and the need for coordinated efforts across health care delivery systems and community agencies to provide recommended medical, behavioral, and social services and support. To address gaps in care quality for persons with dementia, we designed and tested a guideline-based, comprehensive dementia care management intervention in a clinic-level, cluster randomized, controlled trial (27). Intervention components were based on the chronic care model and emphasized linkages with community resources and multiagency coordination (25, 26, 28). Key components included dementia care managers, formal procedures for communication within and between organizations and agencies, Internet-based care management, collaborative care planning with caregivers, caregiver self-management support, ongoing follow-up, and provider education. We hypothesized that patient and caregiver dyads in intervention clinics would receive care in higher accordance with guidelines; receive more community resources and services; have fewer unmet caregiving assistance needs; and have better patient and caregiver health and caregiving quality, social support, and perceived care quality than dyads receiving usual care. Methods The institutional review boards of the University of California, Los Angeles, and all participating health care organizations approved the study procedures. All enrolled patient and caregiver dyads provided written informed consent. Three health care organizations (Kaiser Permanente San Diego, Scripps Clinic, University of California, San Diego, Healthcare) and 3 community agencies (Alzheimer's AssociationSan Diego Chapter, Southern Caregiver Resource Center, Meals on WheelsGreater San Diego) providing services for persons with dementia and their informal caregivers in the San Diego metropolitan area participated. Private group practice, academic group practice, and health maintenance organization practice types were represented by the 3 health care organizations. Study Participants The study included 18 clinics from the participating health care organizations (4 to 8 per organization) (Table 1). The inclusion criterion was that these be primary care clinics. Table 1. Characteristics of Participating Health Care Organizations* Patients with dementia who were age 65 years or older and receiving Medicare were identified by querying health care organization administrative databases for occurrence during the previous year of a dementia diagnosis code (29) at an outpatient visit or hospitalization or a cholinesterase inhibitor prescription. Patient lists were grouped by primary provider and were circulated to those providers for confirmation of dementia diagnosis and identification of additional potential participants from his or her practice (generating <5% of the sample). Patients had to have an informal caregiver (age 18 years). Patients were recruited by mail with English- and Spanish-language letters from their physician and an opt-out postcard, followed by up to 3 mailings of a patient consent or assent form (depending on the physician's designation of the patient's ability to provide consent), a caregiver consent form (containing a proxy consent for the patient if he or she was judged unable to provide consent), and a baseline caregiver survey. Randomization was conducted at the clinic level, based on the clinic-focused design of the care management intervention. Within each health care organization, we paired clinics by patient volume; within each pair, we randomly assigned 1 clinic to the intervention and the other clinic to usual care using a computerized random-number generator operated by a study statistician. Participants were unaware of clinic randomization status at enrollment and at completion of the baseline survey and were not reminded of randomization status at follow-up. Intervention A steering committee that included a physician from each health care organization, a leader from each community agency, a community caregiver, and investigators used a formal method (30) to identify 23 existing dementia guideline recommendations as care goals (11, 12, 14). They also designed a structured assessment, algorithms linking specific care management actions to assessment results, and interorganization care coordination and referral protocols. A key intervention element was health care organization and community agencybased dementia care managers (primarily social workers) who received formal training and used an Internet-based care management software system for care planning and coordination (25, 26, 28). Every enrolled patient and caregiver dyad in the intervention group was assigned 1 health care organization care manager, who contacted them to schedule a structured home assessment. Assessment responses were entered into the software system, generating a preliminary problem list and guides to care-plan actions. The care manager collaborated with the caregiver to prioritize problem areas; teach problem-solving skills; initiate care plan actions; and send an assessment summary, a problem list, and selected recommendations to the patient's primary care physician and other designated providers. A menu of potential care plan actions (for example, referral for respite care services) was documented in a comprehensive care management manual. The care management protocol included ongoing follow-up, usually by telephone, with frequency based on need and a formal in-home reassessment every 6 months to assess the need for major care-plan revisions. The software system had a feature to enable efficient tracking of multiple cases and tasks. Referrals to a particular community agency were guided by flagged problem areas. With patient and caregiver consent, referrals were communicated through the software system to that agency, whose designated care manager subsequently received system access to the assessment, problem list, and care plan. Each dyad could have 1 or more community agency care managers. Care managers from the health care organizations and community agencies received the same formal education and training program, which was conducted jointly, and met monthly to refine care coordination procedures. Care management began within a month after enrollment of the first dyads and was active throughout the study follow-up unless a case was closed, for example, because a patient moved out of the study area and no longer was enrolled in the health care organization. At each intervention clinic, more than 90 minutes of standardized, interactive seminars (in up to 5 sessions) on relevant care issues, including evaluation of acute behavioral changes, depression management, and determination of decision-making capacity (31), were offered to primary care providers. Selected intervention tools and documents with more detailed descriptions can be accessed at www.adc.ucla.edu/access/access.swf. Patients, caregivers, and providers in the usual care group were not offered study interventions. Outcome Measures Primary Outcome The study's primary outcome was a