HEIGHT AND INCIDENCE OF CARDIOVASCULAR-DISEASE IN MALE PHYSICIANS

Background. An inverse association between height and risk of coronary heart disease (CHD) has been reported in several case-control and coh ort studies, but the reasons for the association remain uncertain. We evaluated this association among 22 071 male physicians, a population homogeneous for high educational attainment and socioeconomic status i n adulthood. Methods and Results. The study population was comprised o f participants in the Physicians' Health Study, a randomized, double-b lind, placebo-controlled trial of low-dose aspirin and beta-carotene i n the primary prevention of cardiovascular disease and cancer among US male physicians, aged 40 to 84 years, in 1982. Participants were clas sified into five height categories at study entry, from shortest to ta llest, and were followed an average of 60.2 months to determine the in cidence of myocardial infarction (MI), stroke, and death from cardiova scular disease. Men in the tallest (greater-than-or-equal-to 73 in. or 185.4 cm) compared with the shortest (less-than-or-equal-to 67 in. or 170.2 cm) height category had a 35% lower risk of MI (relative risk, 0.65; 95% confidence interval, 0.44 to 0.99; P=.04), after adjusting f or known cardiovascular risk factors. Further, a marginally significan t inverse trend (P trend=.05) across the height categories was observe d. Although the relationship was not strictly linear, for every inch o f added height, there was an approximate 2% to 3% decline in risk of M I. In contrast, men in the tallest compared with the shortest height c ategory had only small and nonsignificant decreases in risk of stroke and cardiovascular death. While no significant trend in risks of these end points across the height categories was observed, the numbers of events for these end points were far less than for MI, and thus the co nfidence intervals were wide. Conclusions. These data indicate that he ight is inversely associated with subsequent risk of MI. At this time, a few mechanisms are plausible, but none are convincing. Other epidem iological and basic research efforts are needed to explore a variety o f physiological correlates of height that may be responsible for media ting the height-MI association. In the meantime, while height is not m odifiable, it is easy to measure and may be useful to evaluate CHD dis ease risk profiles and target lifestyle interventions.