After the recruitment and consent we chose specific days for fasting blood draws and offered a specific evening at each clinic when group visits would be offered. Twenty-three of the 120 patients could not attend the group visits and blood draw sessions as scheduled, and for personal or scheduling reasons withdrew from the study. The remaining 97 patients (29.9% women) were stratified according to a single entry LDL level and then assigned using an alternating table to create 2 groups with equal LDL levels. The 2 groups were then randomly assigned as experimental and control groups. After this stratification based on LDL levels and random group assignment, we compared the ages, total cholesterol/HDL ratios, hemoglobin (Hb) A1C levels, triglyceride levels, blood pressures, and body mass indexes of the 2 groups and found them to be similar at entry Table 1. During the 1-year study, 4 of the 49 patients in the experimental group and 3 of the 48 patients in the control group dropped out for scheduling or personal reasons before completing the study. Thus, 45 experimental and 45 control subjects completed the trial.
The mean LDL levels for the entire study population decreased from 3.7 mmol per L (142 mg/dL) in January 1997, when the subjects were identified to 3.1 mmol per L (118.5 mg/dL) in September 1997, when they were randomized into groups. This LDL reduction was presumably because of a health maintenance organization–directed campaign to lower LDL levels in this heart care population with lipid-lowering medications. A single medication (simvastatin) accounted for 89% of the lipid-lowering medication used by these subjects. Despite prestudy LDL reductions, the majority of the patients recruited for the D.I.E.T intervention had not yet achieved a 35% LDL reduction.
Intervention
The experimental group met for 14 90-minute group visits over 1 year: weekly for the first month and then monthly. Classes taught by a licensed practical nurse highlighted an antioxidant-rich diet with a maximum of 20% of calories from fat, and encouraged the use of monounsaturated and n-3 fatty acid types of fat in lieu of saturated and polyunsaturated fats. The intervention patients also received a textbook (The 28-Day Antioxidant Diet Program) that included information for shopping lists, menu plans, and food-monitoring sheets. Additional recipes were added at the group’s request and were reviewed during the lectures. Cooking demonstrations were performed. A gradual increase in physical activities, such as walking, was also encouraged. Significant others were strongly encouraged to participate in these classes.
Specific intervention goals were aimed at increasing fruit and vegetable intake to 7 or more servings per day, adding garlic and antioxidant-rich herbs and 1 serving of a legume or soy product daily. The program emphasized choosing low-glycemic carbohydrate sources. Also, the intervention had a goal for each subject to exercise for 30 to 45 minutes 5 to 6 days per week and to reach a target heart rate 60% to 70% of their maximum predicted rate. Within the capitated health system where this intervention occurred, the cost for materials and personnel to continue these types of group visit sessions for CAD patients on an ongoing basis would be $7 (United States) per member per month (PMPM).
Patients in the control group were given written information that included a handout to follow the National Cholesterol Education Program’s Step II-III diet, a handout on choosing dietary fats, information on increasing produce and whole grain intake, and American Heart Association sample meal plans. The control group did not meet in group visits but continued to receive care as usual from their providers.
Members of both groups had their fasting lipids levels and Hb A1C levels drawn and compared at 0 and 12 months. All laboratory results were forwarded to the patients’ primary care physicians, who were strongly encouraged to treat elevated lipid levels with lifestyle-change recommendations and lipid-lowering medications until LDL levels reached a specific target: a 30% to 35% reduction in LDL from their pre-event LDL, or if the pre-event LDL was unavailable, a final LDL less than 2.6 mmol per L (100 mg/dL) or an LDL less than 3.4 mmol per L (130 mg/dL) and with a total cholesterol/HDL ratio less than 4.0.
Data Collection
Blood samples were collected at each clinic, and fasting lipid and Hb A1C samples were obtained. Thirty-day food frequency questionnaires from the Fred Hutchinson Women’s Health Initiative (WHI) were completed by study patients at 0, 3, and 12 months. An additional food intake questionnaire was given to assess both legume intake in servings per week and the type of fat used for cooking and baking.