Changes in population cholesterol levels and coronary heart disease deaths in seven countries

Abstract

Are trends in coronary heart disease deaths based on risk factor changes? To study the relationship between trends in coronary deaths and changes in blood cholesterol in the Seven Countries Study. Sixteen cohorts of men aged 40–59 years from seven countries (U.S.A., Finland, the Netherlands, Italy, Croatia (former Yugoslavia), Serbia (former Yugoslavia), Greece, Japan) were units for the analyses of serum cholesterol measured at entry and after 5 and 10 years, and for mortality over 25 years. In the populations, the ecological relationship of mean serum cholesterol at entry to late coronary heart disease death rates during the 10- to 25-year follow-up was weak, with an R-square of 0·31. Cholesterol measurements made at year 10, and an indicator of cholesterol change during the first 10 years, increased the association (R-square, 0·49). A negative and significant interaction was shown between baseline population cholesterol levels and their 10-year change. As an indicator of acceleration in mortality, cholesterol change over 10 years was also positively correlated (partial R-square 0·44) with the ratio of 25-year to 5-year deaths. In the Seven Countries Study, late coronary heart disease death rates are largely ‘explained’ by changes in blood cholesterol levels during the early phases of the study, mainly due to increases in lower cholesterol levels among some cohorts.