Summary and conclusionsTwenty-four factories or other occupational groups, employing 18 210 men aged 40 to 59, were formed into matched pairs. One of each pair was allocated randomly to receive a five to six year programme of medical examinations and intervention to reduce the levels of the main coronary risk factors. Men at factories in the intervention group were given advice on dietary reduction of plasma cholesterol concentrations, stopping or reducing cigarette smoking, regular exercise for the sedentary and reduced energy intake for the overweight, and hypertension was treated. The programme was delivered mainly through existing occupational medical services, helped by a small central staff. Personal consultations were largely confined to men with a high risk of developing coronary heart disease. Changes in risk factors were assessed by regular standardised examinations of random samples of men. The spread of information by general propaganda proved easy, but a change in habits seemed to require personal contact. Small but significant reductions occurred, mainly in the high-risk group, but these were not sustained when pressure was relaxed.

The relationship between fish consumption and coronary heart disease mortality was investigated using data on foodstuff consumption and mortality from 21 countries. A moderate negative association was found which appeared stable over different periods. This association disappeared when the effects of other foodstuffs were controlled for in multiple regression analysis. One feature to emerge was the anomalous position of Japan in consumption of several foodstuffs. Inclusive or exclusion of this country from regression analysis had a powerful effect on which independent variables entered the model. This illustrates the danger of using single countries, for example Japan with a high fish consumption and low CHD mortality, to support causal hypotheses about diet and disease.

No abstract

By studying all coronary heart attacks presenting within defined communities it should be possible to avoid the distortions and omissions inherent in hospital-based case series. In practice the technique presents several problems. Measures of frequency and outcome are very sensitive to the diagnostic criteria used. Data of varying quality are mixed and specific attack rates can be calculated only for items for which the census provides a denominator. Patients presenting to different medical services have different outcomes, but probably less because of treatment than because the severity of the attack affects behaviour in it. Despite these problems, some such intelligence system is of value in any comprehensive strategy for coronary heart disease.

No abstract