The duration of the association between serum cholesterol and coronary mortality: a 35-year experience

Objective Our purpose was to study the time-related association between a single measurement of total serum cholesterol and the occurrence of coronary deaths in a population sample of middle-aged men. Design This is a longitudinal population study with multiple measurements of serum cholesterol and a 35-year follow-up for coronary mortality in 1622 coronary disease-free men aged 40-59 years belonging to the Italian rural cohorts of the Seven Countries Study. Method A single measurement of serum cholesterol (plus those of five other risk factors as possible confounders) was considered. During 35 years of follow-up, 214 men died from coronary heart disease. Annual partitioned proportional hazards models were solved, one for each independent year of follow-up, to predict the risk of coronary heart disease deaths. Single-year hazard functions were cumulated and smoothed. Similar analyses were made starting from cholesterol measurements taken at years 5, 10 and 20 of follow-up. Results The resulting hazard curves showed regularly increasing risk for coronary events and fitted straight lines with large correlation coefficients, demonstrating a strong association of serum cholesterol with both early and late events. The slopes were steeper for cholesterol measurements taken at younger ages. The slope for cholesterol measurement taken at year 20 (age 60-84 years) was flat. In slopes starting at an early age, two components could be isolated corresponding to different associations with events (stronger for early events when measurements were taken around the age of 50 years and for late events when measurements were taken around the age of 55 or 60 years). Conclusion A single measurement of total serum cholesterol in middle-aged men maintains a relatively monotone relationship with the occurrence of coronary heart disease deaths over 35 years. In the study population, measurements taken in late adulthood had smaller or null association with coronary disease deaths.