Access to the Public Home Health Care and the Use of Health Care Services among the Elderly: Are They Substitutes or Complements?

The rapid increase in elderly population around the world has led to a dramatic growth in the demand for long-term care. By definition, long-term care differs from general health care services in that the care provision lasts for a long time as long as the care receivers (usually with chronic illness and/or disability) are alive. As a result, long-term care and health care services are often financed and delivered in separated schemes. Few studies pay attention on the interaction between the access to long-term care and the use of health care services because the lack of comprehensive data on both sectors for given individuals.

In this study, we explore the unique data set obtained from Taiwan’s national health insurance program which provides insurance coverage for both general health care services and long-term care services delivered in a home setting. The longitudinal health insurance claims data that links the use of both home health care and general health care services for the same individuals enable us to investigate the relationship between the access to home health care benefits and the use of health care services among the elderly in Taiwan. Specifically, we employ the Heckman’s sample selection model to investigate the determinants of the probability of using home health care among the elderly population and the utilization pattern of home health care, including number of visit per year and cost per visit, conditional on the use of home health care benefits. In our analysis, we will employ regional medical resources as the exclusion restriction to correct the selection bias. Based on the framework of the Heckman’s sample selection model, we also investigate the question on how the utilization pattern of home health care in the current periods affects the use of health care services in the future periods, including both outpatient and inpatients costs per year as well as the probability of inpatient admission and the length of hospital stay.

The hypotheses to be tested in this paper are twofold. First, if there is non-financial barrier to get access to home health care, we hypothesis that the utilization pattern of home health care is not uniform across disabled elderly. That is, disabled elderly with certain characteristics are more likely to use the home health care than their counterparts. Second, we hypothesis that the use of home health care is beneficial to the health outcome and hence reduces the use of health care services in the later periods after controlling for the effect of all other variables. An important implication for our study is that there is an efficiency gain under the integrated system for both long-term care and general health care if the empirical evidence supports both hypotheses.