Body mass index and mortality in Seventh-day Adventist men. A critique and re-analysis

The aim of this work was to determine if the relationship between weight-adjusted-for-height (expressed as body mass index or BMI) using the BMI-at-entry and age-at-entry as opposed to BMI at entry and age-at-event (i.e. death, loss to follow-up, or end of the study) would alter the results previously reported from a population of Seventh-day Adventist men. The subjects were 8828 non-smoking, non-drinking Seventh-day Adventist men, ages 30-89 and older on entry, mean follow-up 15 years (maximum 26 years). The BMI and age reported by subjects when they were enrolled into the study were used to calculate the relationship between BMI and mortality. Mortality rates in each of five BMI quintiles were computed by dividing the number of deaths in each quintile by the number of person years of follow-up in the quintile. Rate ratios were computed by dividing each mortality rate by the rate in the reference quintile. The mortality rate ratios were then adjusted for the age difference between each quintile and the reference quintile. Calculations based upon age-at-enrollment rather than 'age-at-event' (as used in the original paper) demonstrate no increase in mortality until a BMI of 27.5 kg/m2 or greater is reached rather than a progressive increase in mortality with increasing BMI.