We propose a randomized, controlled trial to test the effectiveness of a structured treatment program (Assist Model) designed to improve primary care management of hypercholesterolemia in community health centers. The Model emphasizes provider-directed diet treatment facilitating referral to a nutritionist and drug treatment as necessary. The randomized trial will include 40 providers and 400 patients equally divided into intervention and control groups. All providers will use a standardized screening protocol, but control group providers will receive no assistance with diet or drug therapy. We have designed and are currently pilot testing the provider-directed diet treatment component of the Assist Model in three practice settings which serve southern, low income patients. A validated Dietary Risk Assessment is the foundation of our Assist Model intervention. The assessment is quickly and easily administered, scored, and interpreted by a non-nutritionist; identifies major sources of saturated fat and cholesterol in the diet of low income, southern individuals; and identifies misconceptions and attitudes which may serve as barriers to dietary change. Results of the assessment are given to the provider in a format which easily identifies problematic and beneficial eating behaviors, also serving as a flow sheet to monitor progress. Targeted diet counseling is conducted by providers in three 5-minute sessions at separate office visits. Color coded educational counseling is conducted by providers in three 5-minute sessions at separate office visits. Color coded educational materials link specific behavior change recommendations with the major dietary problems identified, thus minimizing the amount of nutrition knowledge required of the provider. The dietary assessment and educational materials are culturally specific for a Southern patient population with minimal reading skills. Behavior change theory is used to guide the intervention, with a focus on individual tailoring, environmental shaping, gradual change, reinforcement, and social support. The Assist Model will provide a list of qualified nutritionists in each clinic area to facilitate referral by the provider. If diet treatment alone fails to reduce cholesterol to target levels, the Model offers a straight forward, stepped care approach to drug treatment and provides simple, illustrated hand-outs for patients to enhance compliance. To evaluate the feasibility of the Assist Model in the broader community, we propose a cost effectiveness analysis within the randomized trial, and an assessment of the Model's exportability to other routine practice settings. The latter will be accomplished by training clinic providers and staff to use the Assist Model materials and then determining their willingness and ability to continue the program with minimal outside support. as background for the effectiveness and feasibility assessment of our Assist Model, we propose a baseline and three year follow-up survey of provider and patient patterns of practice regarding application of the National Cholesterol Education Program Adult Treatment Panel guidelines. We will survey all 190 community health center providers in the state and conduct a more detailed survey of six clinics using chart review and patient interviews.
