The effects of patient cost-sharing on health expenditure and health among older people: Heterogeneity across income groups

Abstract

Despite rapidly rising health expenditure associated with population aging, empirical evidence on the effects of cost-sharing on older people is still limited. This study estimated the effects of cost-sharing on the utilization of healthcare and health among older people, the most intensive users of healthcare. We employed a regression discontinuity design by exploiting a drastic reduction in the coinsurance rate from 30 to 10% at age 70 in Japan. We used large administrative claims data as well as income information at the individual level provided by a municipality. Using the claims data with 1,420,252 person-month observations for health expenditure, we found that reduced cost-sharing modestly increased outpatient expenditure, with an implied price elasticity of - 0.07. When examining the effects of reduced cost-sharing by income, we found that the price elasticities for outpatient expenditure were almost zero, - 0.08, and - 0.11 for lower-, middle-, and higher-income individuals, respectively, suggesting that lower-income individuals do not have more elastic demand for outpatient care compared with other income groups. Using large-scale mail survey data with 3404 observations for self-reported health, we found that the cost-sharing reduction significantly improved self-reported health only among lower-income individuals, but drawing clear conclusions about health outcomes is difficult because of a lack of strong graphical evidence to support health improvement. Our results suggest that varying cost-sharing by income for older people (i.e., smaller cost-sharing for lower-income individuals and larger cost-sharing for higher-income individuals) may reduce health expenditure without compromising health.

Keywords: Cost-sharing; Health expenditure; Income inequality; Older people; Regression discontinuity design.

Fig. 1

Outpatient expenditure by age (per person-month in JPY). Note: Dots represent the average health expenditure on outpatient care by age in months. The vertical dotted lines indicate the age threshold of 70 years. The coinsurance rate was 30% before age 70 and 10% after age 70. Dark lines are from fitting a linear function of age in month, separately for before and after age 70, excluding two months before and after the age threshold

Fig. 2

Inpatient expenditure by age (per person-month in JPY). Note: Dots represent the average health expenditure on inpatient care by age in months. The vertical dotted lines indicate the age threshold of 70 years. The coinsurance rate was 30% before age 70 and 10% after age 70. Dark lines are from fitting a linear function of age in month, separately for before and after age 70, excluding 2 months before and after the age threshold

Fig. 3

Proportion of excellent/good health status by age (per person per two months). Note: Dots represent the proportion of excellent or good health status by age. The vertical dotted lines indicate the age threshold of 70 years. The coinsurance rate was 30% before age 70 and 10% after age 70. Dark lines are from fitting a linear function of age in months, separately for before and after age 70