The purpose of the study is to evaluate the effect of a Guided Care nurse on the quality of the health and well-being of the frail elderly. A specially trained registered nurse will work closely with 1-3 primary care physicians to provide the most complex older patients (and their unpaid caregivers) with health care that is comprehensive, coordinated, patient-centered, and proactive. The study will evaluate the effects of Guided Care on:

• older persons' physical and mental health, health services utilization, quality of care, self-efficacy, and satisfaction with care;
• older persons' unpaid caregivers' burden; and
• primary care physicians' satisfaction with their care of chronically ill patients.

Health care for older Americans with chronic conditions is often fragmented and provider-centric. In response, a team of investigators at Johns Hopkins University has translated the scientific principles of seven successful innovations into one patient-centered system of care. Supported by evidence-based guidelines and state-of-the-art information technology, "Guided Care" is undergoing a 12-month pilot test in older primary care patients with complex needs. A specially trained Guided Care nurse (GCN), based in a primary care practice, collaborates with two primary care physicians to provide seven services for 40-60 high-risk patients: comprehensive assessment and care planning; "best practices" for chronic conditions; self-management; healthy lifestyles; coordinating care; educating and supporting unpaid caregivers; and accessing community resources.

The proposed multi-site study will measure the effects of Guided Care on the quality and outcomes of care for high-risk older persons, their unpaid caregivers, and their primary care physicians. The panels of 53 physicians in 7 practices will be screened to identify 1350 high-risk older patients. After about 850 have given informed consent and baseline interviews, clusters of 2-5 physicians at each practice site will be randomized to provide either Guided Care or usual care to their consenting patients. Each physician cluster in the Guided Care group will incorporate a GCN into its practice; the physician clusters in the control group will not.

Interviews and queries of administrative databases will provide evaluative data at baseline and at 12-, 24-, and 32-month follow-up intervals. The primary outcome variables are the participants' physical health and mental health (SF-36 Summary Scales) and health services utilization. Secondary outcome variables include: the quality of care; unpaid caregivers' burden; self-rated health; patient satisfaction; and primary care physicians' satisfaction. Intention-to-treat analyses will have 85% power (range of 70-97%) to detect clinically meaningful differences between the two groups.

The study is designed to facilitate the prompt dissemination of Guided Care, if the results of the trial are favorable. A stakeholders' advisory board, representing consumers, providers, delivery systems, insurers, regulators and policy-makers, will inform the operation and evaluation of the study - and it will facilitate the subsequent dissemination of its tools and technology throughout American health care.

Inclusion Criteria:

• Over 65
• Insured by KPMAG, USFHP/TRICARE, or Medicare FFS
• High likelihood of use of services in the coming year based on predictive modeling using current year's health care expenses

Exclusion Criteria:

• Moving out of area
• Currently assigned to case manager/in case management program
• Cognitive impairment and no legal representative