Premium
Assessment of cardiovascular risk factors in the elderly: The Framingham heart study
Author(s) -
Larson Martin G.
Publication year - 1995
Publication title -
statistics in medicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.996
H-Index - 183
eISSN - 1097-0258
pISSN - 0277-6715
DOI - 10.1002/sim.4780141604
Subject(s) - medicine , body mass index , hazard ratio , blood pressure , framingham risk score , framingham heart study , diabetes mellitus , confidence interval , proportional hazards model , risk factor , demography , cardiology , disease , endocrinology , sociology
Abstract In elderly Framingham men and women, systolic blood pressure and cigarette smoking status, as well as a subject's age and sex, strongly influenced the risk of developing cardiovascular disease during ten years of follow‐up. Multivariable proportional hazards models were used to assess the roles of several primary risk factors and to examine their secondary effects. The first three factors were noted in both sexes, separately and combined, but the risk function for blood pressure was steeper in men than in women (hazard ratio, HR, 1·53 per 20 mmHg, 95 per cent confidence interval, CI, 1·33 to 1·75 in men; HR = 1·19, 95 per cent CI 1·07 to 1·33 in women). Systolic pressure measured ten years earlier also contributed to CVD risk (HR = 1·16 per 20 mmHg, 95 per cent CI 1·04 to 1·30), even after accounting for current level. Smoking was associated with a 64 per cent elevation in risk, male sex with a 51 per cent increase, and each 5 years increment in age with a 22 per cent increase. Body mass index measured 10 years ago had a modest association, but current body mass index did not. Among diabetic subjects, total serum cholesterol had an asymmetrical U‐shaped risk function, the risk increasing to either side of sex‐specific median values; diabetes per se , however, was not significant in the final model. In non‐diabetic subjects, there was little change in CVD risk up to the median cholesterol values and a modest increase thereafter. None of the risk functions was age dependent.