Description and the concentration of high-cost elderly patients

Evidence from administrative claims data

Abstract

Objectives: Older people tend to use healthcare services more frequently. However, the quantity of medical resource use and the concentration of medical cost among elderly people remains unclear in a super-ageing society, such as Japan. In this study, we described the demographics, annual medical costs and the concentration of high-cost elderly patients using individual-level administrative claims data and assessed their local fiscal impact.

Methods: An analysis of patients aged more than 65 years old was conducted using claims data from the two major insurance schemes, Employees' Health Insurance and National Health Insurance, in two prefectures. Patient demographics, the number of diseases (based on unique ICD-10 chapters) and the proportion of the top 1% to 5% of annual spenders were described, and spending inequalities were calculated using the Gini coefficient. The resource use among the top 1% was assessed based on the cost categories of admission, out-patient clinics, pharmaceuticals and rehabilitation. Medical treatment was also examined in detail according to disease, medication and procedures for the top 100 cases.

Results: The total annual medical cost was 634.9bn yen for the study patients (n= 879,245). Women comprised 58.1% of the population studied, while the proportions by age group were 47.5%, 36.9% and 15.6% for those aged 65–74, 75–84 and 85+ years, respectively. The top 1% of spenders accounted for 12.4% (78.4bn yen), the top 5% for 37.9% (240.6bn yen) and the top 10% for 54.2% (343.9bn yen) of total spending. In addition, 78.5% of the population experienced no admission to hospital. The mean number of diseases per person was 7.5; this number tended to increase with increasing spending category (i.e. top 5% = 10.6 and the top 1% = 11.1). The Gini coefficient for the local population overall was 0.663, and this value tended to decrease with increasing age category (0.710, 0.679 and 0.639 for those aged 65–69, 70–74 and 75+ years, respectively). A detailed analysis of medical treatments showed that medical devices and drugs related to cardiovascular and hematological diseases were major drivers of high annual spending, while surgical operations for multiple diseases and repetitive hemodialysis were also seen to be important drivers.

Conclusion: A large proportion of annual medical spending was concentrated in a relatively small section of the elderly population. The diseases and treatments that drive these costs may provide clues to enable better and more efficient allocation of resources for medical provision both at policy level and at professional autonomy level in the future.