A combination of 4 health behaviours was associated with increasing risk of stroke
Does the combined effect of 4 health behaviours (smoking, physical activity, alcohol use, and fruit and vegetable intake) affect the risk of stroke in community-dwelling people?
cohort study with a mean 11.5-year follow-up (Norfolk component of the European Prospective Investigation of Cancer [EPIC-Norfolk]).
general practices in the UK.
20 040 people who were 40–79 years of age (mean age 58 y, 55% women) with no known myocardial infarction or stroke.
smoking, physical activity, alcohol consumption, and fruit and vegetable intake. Participants were assigned 1 point each for being a non-smoker, being physically active, consuming alcohol moderately (1–14 units/wk), and consuming ⩾5 servings of vegetables and fruit per day (blood concentration of vitamin C ⩾50 μmol/l).
incident stroke (hospital record linkage and death certificate data). Results were adjusted for age, sex, body mass index, systolic blood pressure, cholesterol concentration, aspirin use, diabetes, and social class.
599 strokes (168 fatal) occurred during follow-up. Risk of stroke increased with every point decrease in combined health behaviour score (table).
A combination of 4 health behaviours (smoking, low physical activity, alcohol consumption >2 standard drinks/d, and low fruit and vegetable intake) was associated with increased risk of stroke.
Myint PK, Luben RN, Wareham NJ, et al. Combined effect of health behaviours and risk of first ever stroke in 20 040 men and women over 11 years’ follow-up in Norfolk cohort of European Prospective Investigation of Cancer (EPIC Norfolk): prospective population study. BMJ 2009;338:b349.
Clinical impact ratings: Family/general practice 5/7; Health promotion 6/7; Public/community health 6/7
Source of funding Cancer Research UK; Medical Research Council; Stroke Association; British Heart Foundation; Research Into Ageing; Academy of Medical Sciences; Wellcome Trust.
Lifestyle behaviours of smoking, physical activity, alcohol intake, and diet have been associated with the incidence of stroke. Research examining the combined impact of these lifestyle behaviours on risk of stroke is limited. The study by Myint et al is one of a few studies that have quantified the combined effect of these behaviours on stroke risk. They determined that the presence of these 4 behaviours resulted in a 2-fold difference in the incidence of stroke.
Strengths of this study are the prospective longitudinal design that is appropriate for the research question, large sample, use of trained nurses to conduct health examinations, and adjustment for several potential confounders. Plasma vitamin C concentrations give an objective measurement of fruit and vegetable intake when compared with self-reports of dietary intake, which can be affected by social desirability bias. Background information on the validity of plasma vitamin C concentrations as surrogate markers of fruit and vegetable intake would have been helpful.
Caution is needed when generalising the findings to ethnically diverse populations and populations with comorbid conditions. Because not all eligible participants attended the health check, there was a volunteer bias towards healthier respondents (but this bias would only dilute the strength of associations rather than overestimate the effect). The exclusion of about 9000 respondents because of missing values resulted in the loss of some analysis. Although the investigators did not find a difference between included and excluded participants, differences in other unmeasured characteristics might have influenced the outcome.
The findings of this study are relevant to nurses who work in neurology and coronary care follow-up clinics and rehabilitation, community health nurses whose clientele include adults and older adults, and public health nurses involved in health education, prevention, and promotion. Advanced practice nurses working with adults and older adults in either primary healthcare or acute-care settings would also find the results useful. Although the relative effects of the different factors have not been assessed, patients now have a range of choices to start modifying their risk. Furthermore, the findings suggest that behaviour change need not stop at one success.