Do we over- or underestimate the overall cost burden of infectious diseases in ageing adults?

Abstract

Background arithmetic average values about disease burden across ageing adults are often used, which assumes homogeneity in group characteristics such as age, sex, disease frequency (incidence rates), and cost distributions. The question arises about how much outcome results such as overall cost obtained under this homogeneity assumption deviate from real-world population data that may manifest non-homogeneous distributions. Methods the method explores the amount of deviation measured between homogeneity versus non-homogeneity for overall infection costs in ageing adults as the outcome measure to assess. Population modelling is used with an extended sensitivity analysis plan (ESAP) that simulates non-homogeneous, age-specific distributional spread for demography, infectious disease, and its severity in people aged > 65 years old over a 1-year period in univariant and multivariant assessments. Costs are adjusted for 3 severity levels with increased difference between them using multiplication factors. Results the assumed full homogenous dataset systematically overestimates up to 10% the overall cost in ageing adults when compared with a group simulated with non-homogeneous distributions for age, infection, severity, and cost, mainly due to the demographic age-composition. Overall cost of a proposed homogeneous condition tends to underestimate the spending of non-homogeneous conditions when the reference case has a partial homogeneous set-up or when the demographic change in the non-homogeneous condition evolves towards age-demographic homogeneity (same number of people with increasing age), a likely evolution in the coming 10 to 20 years. Conclusion assessing the current cost burden of infectious diseases in ageing adults must consider exact age-composition of demography, infection spread with severity levels and their cost differences to avoid unrealistic cost estimates when assuming homogeneous group conditions.