Counting the time lived, the time left or illness? Age, proximity to death, morbidity and prescribing expenditures

•Prescription medication expenditures have significantly increased in the last two decades.•Proximity to death is a more important driver of prescription expenditure than age.•But proximity to death may itself be a proxy for morbidity, disability or senescence.•Accounting for chronic conditions reduced the observed proximity to death effect.•Projections of prescription expenditure need to consider mortality and morbidity.

The objective is to understand what really drives prescription expenditure at the end of life in order to inform future expenditure projections and service planning. To achieve this objective an empirical analysis of public medication expenditure on the older population (individuals ≥ 70 years of age) in Ireland (n = 231,780) was undertaken. A two part model is used to analysis the individual effects of age, proximity to death (PTD) and morbidity using individual patient-level data from administrative pharmacy records for 2006-2009 covering the population of community medication users. Decedents (n = 14,084) consistently use more medications and incur larger expenditures than similar survivors, especially in the last 6 months of life. The data show a positive and statistically significant impact of PTD on prescribing expenditures with minimal effect for age alone even accounting for patient morbidities. Nevertheless improved measures of morbidity are required to fully test the hypothesis that age and PTD are proxies for morbidity. The evidence presented refutes age as a driver of prescription expenditure and highlights the importance of accounting for mortality in future expenditure projections.