Extensive evidence implicates diet as one of the key factors in the etiology of atherosclerosis, and suggests that the disease can be prevented by changes in diet, particularly to lower serum cholesterol level. In 1960, an Executive Committee on Diet and Heart Disease was established to explore the matter of mass field trials on the prevention of atherosclerotic coronary heart disease. After two years of deliberations, the Executive Committee concluded that a well controlled mass field trial was needed to test the hypothesis that alteration of the amount and type of fat and the amount of cholesterol in the diet would decrease incidence of first attacks of clinical coronary heart disease in middle-aged American men. It estimated that a study population of up to 100,000 men, with a follow-up period of four or five years, would be required to obtain a statistically reliable test of the hypothesis if the incidence were reduced by 20 per cent. The Committee further concluded that, on the basis of information then available to it, a definitive decision could not be made concerning the feasibility of such a large-scale study. It therefore recommended that the next major endeavor be a smaller feasibility study. Based on this recommendation, a common protocol was developed for a collaborative feasibility study. Feasibility of two types of mass field trial was to be evaluated: 1. among free-living men from the general population, and 2. among long-term residents of mental institutions, with both studies to be conducted on a double-blind basis, through use of special fat-modified foods. The original protocol enumerated eleven topics to be clarified by this National Diet-Heart Study in a one to two year research endeavor: 1. Effectiveness of recruitment of volunteers from several sources; 2. rate of volunteering; 3. effectiveness of three concomitant dietary modifications-moderate reduction of per cent calories derived from total fat and/or saturated fat, increase in per cent calories from polyunsaturated fat and in the polyunsaturated to saturated fat ratio (P/S), and moderate reduction in dietary cholesterol; 4. availability, palatability, distribution, and cost of specially prepared foods; 5. indices of adherence to diet; 6. extent of reduction of serum cholesterol levels; 7. dropout, non-adherence, and non-response rates; 8. feasibility of a double-blind technique; 9. personnel, equipment, and physical facilities necessary; 10. geographic and laboratory variations within a common protocol; 11. experience in the pooling data from a cooperative study. Findings of the first months of the Study revealed several problems requiring further investigation. Therefore, the decision was made to continue beyond the first year, with appropriate modification of the protocol for the second year. These further efforts-designated the Extended Study and the Second Study respectively-aimed to evaluate the following major problems, in addition to those explored during the initial year (First Study): 1. What proportion of the participants in the First Study would volunteer for the Extended Study? What would the non-adherence and dropout rates be in the Extended Study and the Second Study (with use of newly recruited men in the latter)? 2. Could new individuals be recruited for such studies with the explicit understanding that some of them would be randomly assigned to a control group? 3. Could costs be reduced by having participants select lean cuts of meat on the open market? 4. Could participants be satisfactorily instructed to select and trim meats? Would their intake of fat and cholesterol be well controlled? Would they demonstrate adequate blood lipid responses? 5. Could new experimental diets, acceptable to the participants, be devised and evaluated? How would they compare to the diets utilized in the First Study? Could the differences between control and experimental groups be enhanced? 6. Could diet modification programs be made less demanding for participants, nutritionists, and investigators? Would such altered programs be effective? 7. Could any evidence of a seasonal variation in serum cholesterol level be detected, as a possible confounding factor influencing the trend of serum cholesterol level? Facilities for the study of free-living or "open" populations were established in five cities (Baltimore, Boston, Chicago, Minneapolis-St. Paul, Oakland), and of an institional or "closed" population at the Faribault State School and Hospital, Minnesota. A Central Staff was established, and-in co-operation with the Principal Investigators-it procured and distributed the special fat-modified foods; developed forms and a manual of uniform procedures; distributed these materials; administered the double-blind design; maintained quality control; and assumed responsibility for the central collection, tabulation, analysis, and distribution of data. A Central Laboratory analyzed all sera for total cholesterol.