Health related quality of life by age, gender and history of cardiovascular disease: results from the Health Survey for England
R Ara, J Brazier
Introduction: While there is an increasing volume of literature describing the health state utility values (HSUV) individuals with angina, heart attack or stroke, research comparing HSUVs for these conditions within the same study is sparse. This hinders analysts wishing to explore the benefits of interventions in cardiovascular disease (CVD). The objective of this study was to obtain EQ-5D scores to inform health states in CVD economic models using the same source for each health state.
Methods: EQ-5D data (N = 26,679) from individuals aged 16 to 98 years taking part in the Health Survey for England (2003, 2006) was used. Regressions were employed to explore the significance of age, gender and history of CVD on HSUVs taking into account the number and type of cardiovascular condition(s) and time since event. The predictive ability of the model was assessed using errors in predicted values on both the individual and the sub-group levels.
Results: Our results show HSUVs differ by age, gender, CV history, time since CV event and the number of concurrent CV conditions. The regression model is reasonably accurate when predicting mean values for sub-groups stratified by age, CV condition and time since event, with 67% of predicted values within the minimal important difference for the EQ-5D index.
Conclusions: The results provide a consistent basis for analysts to populate the most frequently defined health states (no CVD, heart attack, angina and stroke) in CVD models. Further research is required to validate these results in other datasets.