Causes of the trends and variations in CHD mortality in different countries.

Rapid changes in national rates, with little evidence of cohort effects, must substantially reflect changes in incidence rates due to socioeconomic and behavioral influences, operating with a rather short incubation period. Every newly-rich society usually experiences its epidemic of coronary heart disease. Rate changes for men and women are correlated, but the regression is asymmetrical: greater falls in women seem to reflect some rather uniform widespread sex-specific change, on top of which is another and highly variable factor common to both sexes. Trends are correlated with diet (especially with the constitution of fats). A correlation with stroke trends suggests a common link with population changes in blood pressure, in which obesity may play a part. Other important influences on population trends and differences remain unidentified or unconfirmed.