Between 1989 and 1999, Stockport Primary Care Trust in the North West of England implemented a population-wide screening program for coronary heart disease (CHD), whereby residents aged between 35 and 60 years attended screening via their general practices (n=84,646).
Factor analysis was performed using data collected on blood pressure, body mass index (BMI), cholesterol, smoking, and alcohol intake. Factor scores were developed for each component and entered into a logistic regression model in order to explore the links with hospital episodes and mortality due to CHD for the screened participants. Results were stratified by social class (affluent/deprived) and gender, and odds ratios with 95% confidence intervals are reported.
Risk factors clustered into two distinct components for all groups: the first comprised of blood pressure, BMI and cholesterol; and the second comprised alcohol and smoking. These clusters were conceptually labelled as ‘physiological' and ‘lifestyle'. In terms of mortality, the association of the physiological factor was strongest in deprived women (OR = 2.1; 1.6 to 2.7), but the association of the lifestyle factor was strongest in affluent women (OR = 1.7; 1.3 to 2.0). In terms of hospital episodes, the association of the physiological factor was again strongest in deprived women (OR = 1.7; 1.4 to 2.0), but the lifestyle factor had the same predictive effect for both groups (OR = 1.2).
These results suggest that traditional measures of risk assessment which assume that all population groups have the same vulnerability are unlikely to adequately identify risk.
Learning Objectives: At the conclusion of the session, the participant (learner) in this session will be able to
Recognize how risk factors for coronary heart disease may cluster depending on social group.
Evaluate the association that these risk factor clusters have with outcome (e.g. morbidity and mortality rates).
Understand the implications for risk assessment
CORE (COnnecting REpositories)
CORE (COnnecting REpositories)
