A multinomial model for comorbidity in England of long-standing cardiovascular disease, diabetes and obesity.

Comorbidity has been found to be significantly related to increased levels of mortality, decreased functional status and quality of life, increasing dependence on health services and an increased risk of mental and social problems. Previous research into comorbidity has mainly focused on identifying the most common groupings of illnesses found among elderly healthcare users. In contrast, this paper pools data from the Health Survey for England from 2008 to 2012 to form a representative sample of individuals in private households in England to explore the risk of comorbidity among the general population; and to take account of not only the demographic but also the socioeconomic and area-level determinants of comorbidity. Using a multinomial logistic model, this research confirms that age and gender are significant predictors of cardiovascular disease, diabetes and obesity, whether examined singly or in any comorbidity combination. Across the seven possible disease combinations, the odds ratios are lowest for those individuals with a high income (6 of 7), home-owning (5 of 7), degree educated (7 of 7) and living in the least deprived area (6 of 7), when controlling for demographic and smoking characteristics. The important influence of socioeconomic factors associated with comorbidity risk indicates that healthcare policy needs to move from a focus on age profiles to take better account of individual and local area socioeconomic circumstances.