A population survey of cardiovascular disease in elderly people: design, methods and prevalence results.

A cross-sectional survey of an age- and sex-stratified random sample of the elderly population living in Southampton was undertaken with the object of measuring the frequency of cardiovascular disease by questionnaire and examination, and assessing cardiac anatomy and physiological function by noninvasive methods. The response rate was 64% and 259 men and women aged between 65 and 95 years were interviewed and examined and had a twelve-lead ECG and chest radiograph. Forty per cent of men and 47% of women reported a diagnosis of one or more cardiovascular diseases, of which high blood pressure 33%, coronary heart disease 14% (angina pectoris 11% and myocardial infarction 8%), peripheral arterial disease of the lower limbs 7%, and cerebrovascular disease 6% occurred most frequently. With the exception of high blood pressure, which women reported more frequently than men (40% vs 27%), the prevalence of these diagnoses by sex was similar. The self-administered WHO questionnaires gave point prevalence estimates for angina pectoris of 13%, possible myocardial infarction 7% and intermittent claudication 5%, which were similar to reported prevalences of these diseases, although disagreement in the classification of individuals for each disease was common. When comparing the WHO chest pain questionnaire with the doctors' independent diagnosis of angina pectoris in this population, the sensitivity of the WHO questionnaire was 79%, with a specificity of 98% and a predictive value of 88%. For each sex the sensitivity of the WHO chest pain questionnaire was similar but specificity and predictive value were both lower for women. Thirty-four per cent of men and 41% of women reported taking cardiovascular drugs, of which diuretics and beta-blockers were the most commonly prescribed.