Three diets were tested double-blind in the First Study in all five open centers and at Faribault-two experimental (groups B and C) and one control (group D). Plans for both experimental diets involved reductions in saturated fat (< 9% of calories) and cholesterol intake (350-450 mg./day), and increases in polyunsaturated fat intake (15%+, and 18-20% of calories). The two diet plans differed in per cent of total calories from total fat (30% versus 40%) and in ratio of polyunsaturated to saturated fat (1.5 versus 2.0). The control diet (D) was planned to be similar to the usual American diet, i.e., 40 per cent of calories from total fat, 16-18 per cent of calories from saturated fat, no more than 7 per cent of calories from polyunsaturated fat, a P/S ratio of 0.4, and an average daily cholesterol intake of 650-750 mg.
At Faribault, another double-blind experimental diet was included (diet E), providing 40 per cent of calories from total fat, low in saturated fat and cholesterol, very high in polyunsaturated fat, with a P/S ratio of 4.4.
Another experimental Diet (X) was included at the Twin Cities Center, not double-blind with respect to diet assignment, but double-blind with regard to serum lipid responses. It resembled diet B in composition, but utilized diet instruction alone, and purchase of foods exclusively in the open market, without special D-H foods.
To implement the double-blind design for groups B, C, and D at the open centers, Food Centers were established in each city and stocked with special D-H foods for distribution to participants' families on order (with payment handled through periodic billing). Chops, roasts, and steaks were all lean cuts, the same for all diets. The bulk of dietary fat was to come from "filled" meats and dairy products, pastries, mayonnaise, and salad dressings. "Filled" foods were those special products to which specified fats were added. These foods were fabricated by industry so that three varieties (B, C, and D) of each item were made available, essentially identical in appearance and taste, but with different contents (amount and type) of fat and cholesterol.
All participants were given identical dietary instruction aimed at reducing the intake of saturated fat and cholesterol, and were asked to purchase most fat-containing foods from the Food Centers. Food Center managers were aware of each man's diet assignment, and the codes identifying B, C, and D products, and therefore assumed responsibility for the delivery of proper foods. Adherence to diet instructions and use of the special D-H foods as prescribed assured achievement of assigned diet composition.
Diet plans were based on conventional foods required for adequate nutrition. Foods of similar fat content were listed in five groups (D-H exchange lists) and arranged in a logical order for diet instruction. The first group consisted of fat-containing protein foods in which saturated fat was either relatively low, as in poultry, or could be partly removed, as from meat. The second group contained low fat, high carbohydrate foods-breads, cereals, starchy vegetables, and sweets. The third group had fat-containing products, three varieties of each (B, C, and D), in which the kind and amount of fat were adjusted by the manufacturers of special fat-modified D-H foods. This group contained both high protein foods, including dairy and meat products (subgroup A), and high carbohydrate foods (subgroup B). The fourth group included low fat vegetables and fruits. The fifth group was made up of visible fats and oils, also fabricated with three compositions for diet groups B, C, and D. Most fat-containing foods-i.e., those of groups 1, 3, and 5-were to be purchased from D-H Food Centers. Other allowable foods were to be purchased in the open market.
Based upon the participant's estimated caloric need, a one day structured diet plan was developed, consisting of a definite number of servings from the five food groups. To permit flexibility, the man and his wife were instructed concerning exchanges among foods in a given group. Exchanges were also possible between foods in groups 3 and 5.
Four seven-day Food Records were kept by the participants, one during the baseline period and three during the experimental period of the First Study. To permit calculation of nutrient composition from the Food Records, tables were prepared listing the nutrient content of both D-H and open market foods. A Guidebook and Manual For Food Record Coding was developed, together with review procedures, to assure standardized work-up of Food Records. Computer processing and analysis of data were done by Central Staff. Inter-center differences and coding errors in Processing Records were minimal.
An individual diet was prepared for each man, based on a standard plan for estimated calorie level, and on his individual habits, as revealed by the Food Record and other nutritional information collected during the base-line period. The participant and his wife received detailed diet instruction, both verbal and written, including a special illustrated brochure, The Diet-Heart Study Guide, a daily meal plan, exchange lists, recipes, menus, advice on eating out, a list of foods to avoid, etc.
During the 56 weeks of the First Study-4 weeks on mixed diet, 52 weeks on assigned diet-the participant was interviewed by his nutritionist at each of 10 follow-up visits. At each session, the nutritionist made a subjective evaluation of adherence, using a four point rating scale-excellent, good, fair, or poor. In addition, at some visits semi-objective estimates of adherence were made, based on a set of questions concerning kinds and amounts of fat-containing foods eaten during the previous week, and based on information in the 7-day Food Record.
In the Extended Study, the Baltimore, Chicago, and Twin Cities men who had been on diets B, C, and D were all placed on a hypocholesterolemic diet, BC, made up of a combination of D-H foods from diets B and C of the First Study. Total fat was planned to be 30-40 per cent of calories, cholesterol 350-450 mg./day, saturated fat was to be reduced, polyunsaturated fat increased compared with baseline levels, the P/S ratio to be 1.5-2.0. Half the participants continued using D-H meats, the other half were instructed in the purchase and proper preparation of selected lean meats from the open market-to assess relative effectiveness of these two sources. The Twin Cities X group continued its same diet. At Oakland and Boston, men were switched on a double-blind basis to two new diets, F and G, with use of D-H meats only. Diet F was planned to contain 40 per cent of calories from total fat, with 350-450 mg./day of cholesterol, a low saturated fat and high polyunsaturated fat intake, and a P/S ratio of 3.0, i.e. higher than that of groups B, C, or BC. Group G was planned as intermediate between C and D of the First Study, with 40 per cent of calories from total fat, 350-450 mg./day of cholesterol, a modest reduction in saturated fat and a modest increase in polyunsaturated fat compared with baseline levels, a P/S ratio of 1.0.
In the Extended Study, the Faribault subjects continued with diets B, C, D, and E, but less costly meats from the general market were utilized, rather than D-H meats.
In the Second Study, the Baltimore, Chicago, and Twin Cities men were assigned by stratified randomization to double-blind groups BC and D, with use of either D-H or specially selected and trimmed open market lean meats, and either structured or unstructured diets, i.e. a total of 8 groups. Assessment was made of the unstructured diet to simplify the work of nutritionist, participant, and wife. Only meat, fish, poultry, and eggs (group 1 foods) were prescribed. Fat-controlled D-H foods and low-fat foods from the open market were eaten as desired.
The Twin Cities had a Group Y, resembling its Group X of the First and Extended Study, and a Group Z. The latter was an open control group, receiving no dietary instruction, but reporting to the Research Center periodically for evaluation of serum cholesterol, weight, nutrient intake, etc.-to assess whether "exposure" to the Study induced measurable change in these variables.
At the Oakland and Boston Centers, Second Study men were assigned to diets F, G, and D, with use of specially selected and trimmed open market meats.
At Faribault, the Second Study explored the special problem of the possible interaction between dietary cholesterol and type of dietary fat in the regulation of serum cholesterol level.
- © 1968 American Heart Association, Inc.