Background In European countries Cardiovascular Disease (CVD) risk is higher in winter; this is mainly attributed to cold weather and outdoor temperature falls. These associations could result from the effects of temperature on CVD risk factors, especially in older people in whom CVD risks are high. We have therefore investigated associations between outdoor temperature and established and novel CVD risk factors in older adults from two large Northern European studies.

Methods Data were based on two prospective studies: the British Regional Heart Study (BRHS), a population-based study of older men from 24 British towns examined in 1998–2000, and the PROspective Study of Pravastatin in the Elderly at Risk (PROSPER), a trial of pravastatin vs placebo which took place in 1997–9 and included three coordinating centres (Glasgow, Scotland; Cork, Ireland; and Leiden, The Netherlands). Established (e.g. Systolic Blood Pressure [SBP] and lipids), and novel CVD risk factors (vitamins, inflammatory, and haemostatic factors) were analysed. Outdoor mean temperature data were provided by the national meteorological offices. Associations between outdoor temperature on the day of the examination (exposure variable) and the CVD risk factors (outcomes) were studied. Next, results from the BRHS and PROSPER were pooled using a random effects model, for each of the outcomes separately.

Results 4252 men (77% response rate) aged 60–79 years from the BRHS, and 5804 men and women aged 70–82 from the PROSPER (24% response rate to the clinical trial) participated. Pooled differences in established CVD risk factors for a 1 standard deviation (5 °C) decrease in mean temperature were 0.8% (95% Confidence Intervals (CI) 0.2; 1.4%) for SBP, 0.8% (95% CI 0.4; 1.2%) for total cholesterol), 0.5% (95% CI 0.1; 1.1%), for HDL-cholesterol and 2.1% (95% CI 1.5; 2.7%) for LDL-cholesterol. Pooled differences in novel CVD risk factors for 5 °C decrease in mean temperature were 3.3% (95% CI 1.0; 5.6%) for C-reactive protein, 2.7% for Interleukin-6 (95% CI 1.1; 4.3%), 0.7% (95% CI 0.2; 1.3%) for Fibrinogen, and 0.4% for Plasma viscosity (95% CI 0.3; 0.5%). Conversely, the pooled difference in Vitamin D for 5 °C decrease in mean temperature was −11.2% (95% CI −20; −1%).

Conclusion Variations in outdoor temperatures were associated with variations of CVD risk factors. Markers of inflammation and Vitamin D showed the strongest associations, which may reflect a physiological response of the elderly to typical elements of the cold season, such as low temperatures.