Abstract The majority of health behaviour research focuses on single risks (smoking, alcohol, diet, exercise) and/or single diseases. However, individuals have multiple lifestyle risks, particularly in deprived populations, and multiple disease outcomes. This project is seeking to inform priority setting for lifestyle interventions, in terms of single or multiple risk factors and individual or area based characteristics, and to demonstrate the potential for using secondary data sets for evaluating health behaviour interventions.
The research applies econometric analysis to linked secondary data sets containing health survey data administratively linked to hospital admissions records. Disease risk (hospital admission) is estimated for smoking-related diseases taking into account other health behaviours, health history, area and individual deprivation and demographics. Predicted risk can be estimated for selected characteristics or combinations of characteristics. The predicted risks of a smoking related disease are shown to increase with area deprivation but the effect is small compared to the effects of smoking status, having a previous history of disease or having other adverse health behaviours.
The outputs from the econometric analysis can also be combined into an economic evaluation model to assess the cost-effectiveness of interventions targeted at different groups.Conclusions Increasingly scarce resources for prevention need to be allocated as efficiently as possible. Modelling approaches which can incorporate multiple risks and multiple diseases are likely to be more useful in informing this process than single risk, single disease models.
Published: 2nd May 2012