Background In Scotland, as in other countries, cardiovascular disease (CVD) mortality has substantially declined over time. However, this decline may be slowing among younger groups and there are still large inequalities in mortality between socio-economic groups.

Objectives To examine secular changes in the prevalence of known CVD risk factors in the Scottish population according to socio-economic position.

Design Representative data on health and health-related behaviours from the Scottish Health Surveys, collected by stratified and clustered probability sampling.

Participants Data are available from four surveys: 1995, 1998, 2003 and 2008 (6190, 6656, 5497 and 4202 respondents respectively aged 25–64 years).

Main outcome measures Prevalence of cigarette smoking, excess alcohol consumption (men>21, women>14 units/week), hypertension (self-reported), diabetes (self-reported) and obesity (measured BMI ≥30kg/m²).

Methods Prevalences, adjusted for survey weights, and stratified by gender and by gender/age were determined according to the individual's social class (Registrar General's) and educational level (highest qualification), standardised to the European standard population. Time-trends and trends across socio-economic categories were assessed using linear regression. The slope index of inequality (SII) was calculated for each risk factor in every survey.

Results In 1995 the prevalence of smoking among professionals was 14.1% (95% CI 9.5 to 18.8%) compared to unskilled workers 51.0% (46.1 to 55.8%) with SII of 39.0. By 2008 the SII for smoking had decreased slightly to 35.7, with decreased prevalence in all social classes, except in the unskilled. Self-reported hypertension increased moderately within most educational levels; in men the corresponding SII changed little between the first and last survey (8.8 and 8.2 respectively) while for women it increased from 4.2 to 12.1. Diabetes prevalence increased in all educational categories. By 2008, the prevalence was 5.8% (3.9–7.7%) among all those with no qualifications and the SII had increased from 1.0, in 1995, to 3.9; (from 1.8 to 5.1 among women). The prevalence of obesity increased across the surveys with minimal change in the inequality gap. among those with no qualifications the prevalence increased from 23.3% (21.2–25.3%) at baseline to 31.2% (26.5–35.8%) in 2008. In the corresponding years the SII for obesity had increased from 10.7 to 13.0. Difficulties in reporting alcohol consumption trends arise from changes in recording practices between surveys; approaches will be presented.

Conclusions Individuals of lower socio-economic status continue to carry the heaviest burden of CVD risk factors. There has been little, if no reduction in the inequality gap over time; indeed for some factors it may be growing.