How much disease-risk is due to old age and established risk factors?

Abstract

As improved healthcare leads to older populations, individuals will increasingly experience multiple diseases, possibly concurrently (multimorbidity). This article explores whether age and established risk factors are sufficient to explain the incidence rates of multiple, possibly coexisting diseases. By accounting for the limited age-range in UK Biobank data, previous work demonstrated that a Weibull model could accurately describe the incidence of ∼60% of the most common primary hospital diagnoses of diseases. These are used here to predict the age-dependent incidence of diseases with adjustment for established risk factors. A “Poisson binomial” model is combined with these to predict the total number of occurrences of each disease in the UK Biobank cohort that would be expected without pre-existing (prior) disease. For 123 diseases in men and 99 diseases in women, the total observed new cases of each disease (including those from individuals with pre-existing diseases and multimorbidity), were found to be approximately 1.5 times greater than that predicted for individuals without prior disease, and could not be explained by natural statistical variation. The multiple of 1.5 was sufficiently consistent across different diseases to prevent its use for classification of disease types, but there were differences for sub-groups such as smokers with high body mass index, and for some classes of disease (as defined by the International Classification of Diseases version 10). The results suggest that empirical modelling might allow reliable predictions of primary causes of hospital admissions, helping to facilitate the planning of future healthcare needs.