Dietary behaviour is an important modifiable factor in cardiovascular disease (CVD) prevention. The study aimed to identify dietary patterns (DPs) and explore their association with CVD incidence and risk markers. A follow-up of 1838 middle-aged men, aged 47-67 years recruited into the Caerphilly Prospective Cohort Study at phase 2 (1984-1988) was undertaken. Principal component analysis identified three DPs at baseline, which explained 24.8% of the total variance of food intake. DP1, characterised by higher intakes of white bread, butter, lard, chips and sugar-sweetened beverages and lower intake of wholegrain bread, was associated with higher CVD (HR 1.35: 95% CI: 1.10, 1.67) and stroke (HR 1.77; 95% CI: 1.18, 2.63) incidence. DP3, characterised by higher intakes of sweet puddings and biscuits, wholegrain breakfast cereals and dairy (excluding cheese and butter) and lower alcohol intake, was associated with lower CVD (HR 0.76; 95% CI: 0.62, 0.93), coronary heart disease (HR: 0.68; 95% CI: 0.52, 0.90) and stroke (HR: 0.68; 95% CI: 0.47, 0.99) incidence and a beneficial CVD profile at baseline, while DP1 with an unfavourable profile, showed no clear associations after 12 years follow-up. Dietary pattern 2 (DP2), characterised by higher intake of pulses, fish, poultry, processed/red meat, rice, pasta and vegetables, was not associated with the aforementioned outcomes. These data may provide insight for development of public health initiatives focussing on feasible changes in dietary habits.
Keywords: Caerphilly Prospective Study (CaPS); cardiovascular incidence; cardiovascular risk markers; dietary patterns; principal component analysis.