The relation between adult height and haemorrhagic and ischaemic stroke in the Renfrew/Paisley study

Open Access
Publisher Website
Google Scholar
Abstract
The Renfrew/Paisley study, carried out between 1972 and 1976, involved 15 406 (almost 80%) residents aged 45–64 years from the towns of Renfrew and Paisley. Participants were representative of the population of the west of Scotland. Full details are available elsewhere.4 Data collected included height, blood pressure, plasma cholesterol, smoking habit, occupational social class and past medical history. A computerised linkage of hospital admissions in Scotland provided records of main diagnoses of stroke over a 20 year follow up period. After omitting participants with missing data, 15 393 people were included in the analyses. Height quintiles were calculated separately for men and women with height groups formed by taking each fifth as the corresponding fifth for men and women respectively. Strokes were classified as ischaemic or haemorrhagic (see table 1). Cox's proportional hazards models were used to calculate relative rates for height quintiles, and per 10 cm increase in height. There were no significant interactions between height and sex for each type of stroke, so data for men and women were analysed together. Additional adjustment was made for other risk factors for stroke.