The Effect of Patient Cost Sharing on Utilization, Health, and Risk Protection
By Hitoshi Shigeoka
National Bureau of Economic Research, Working Paper 19726, December 2013
Abstract
This paper exploits a sharp reduction in patient cost sharing at age 70 in Japan, using a regression discontinuity (RD) design to examine its effect on utilization, health, and financial risk arising from out-of-pocket expenditures. Due to the national policy, cost sharing is 60-80 percent lower at age 70 than at age 69. I find that both outpatient and inpatient care are price sensitive among the elderly. While I find little impact on mortality and other health outcomes, the results show that reduced cost sharing is associated with lower out-of-pocket expenditures, especially at the right tail of the distribution.
1. Introduction
I reach three conclusions. First, I find that reduced cost sharing at age 70 discontinuously increases health care utilization. The corresponding elasticity is modest, at around -0.2 for both outpatient visits and inpatient admissions. Examining patterns of utilization in more detail, I also find that lower patient cost sharing is associated with increases in the number of patients presenting both serious and nonserious diagnoses. For example, I find large increases in outpatient visits for diagnoses that are defined as Ambulatory Care Sensitive Conditions (ACSCs), for which proper and early treatment reduces subsequent avoidable admissions.
Second, in terms of benefits, I do not find that lower patient cost sharing improves any of the health measures I examine, such as mortality and self-reported physical and mental health. Since health is a stock, it may take some time for the most observable health effects to be realized. Therefore, it is challenging to address it using the RD approach unless the causes of death are acute. Nonetheless, I do not find any change even in acute cause-specific mortality. The lack of differences in health in spite of utilization changes implies that patient cost sharing can reduce health care utilization without adversely affecting health, at least in the short run.
Finally, I do find that lower cost sharing at 70 yields reductions in out-of-pocket expenditure, especially at the right tail of the distribution, because the reduction in price at age 70 overwhelms offsetting increases in utilization. This finding suggests that patients with high medical spending benefit substantially from financial protection against risk due to lower cost sharing.
6. Discussion
6.3 Cost—benefit Analysis
Finally, I conduct a simple cost—benefit analysis associated with the change in the price of health care services at age 70. Since I needed to make a number of assumptions, the results from this exercise are mostly speculative. The social cost is the combination of the deadweight loss of program financing and the moral hazard, while the benefit is risk protection against unexpected out-of-pocket medical spending. My estimates suggest that the welfare gain of risk protection from lower patient cost sharing is comparable to the total social cost, indicating that the welfare gain from risk protection may fully cover the total social cost in this setting. One limitation of this welfare analysis is that it does not incorporate welfare gains from health improvements. While I do not find any short-term reduction in mortality or improvement in any self-reported health measures, it is possible that preventive care induced by lower cost sharing at age 70 may prevent severe future health events, thus improving health in the long run. It is infeasible to estimate long-run effects in this framework, because individuals eventually age into treatment.
http://www.nber.org/papers/w19726?utm_campaign=ntw&utm_medium=email&utm_source=ntw
Comment:
By Don McCanne, M.D.
Although this paper is challenging to read because of its technical nature, nevertheless it is an important contribution to our understanding of patient cost sharing for health care services and what that means for the individual and society.
Cost sharing has permeated health care financing in the United States, and it is intensifying. Why? It is commonly thought to be one of the most important tools to control health care spending. By requiring the patient to pay a significant portion of the health care that they actually utilize, it is thought that patients will select only the health care that they really need and will decline care that is of little value. But is the care that they would decline of so little value that it is not worth what would be spent on it? Let’s look at that.
A unique feature of health insurance in Japan provides some insight. When a patient turns 70, cost sharing is reduced 60 to 80 percent. In this paper, Hitoshi Shigeoka of Simon Fraser University in British Columbia does confirm what other studies have shown – reducing cost sharing does modestly increase utilization of outpatient and inpatient services. As with other studies, he did not find that lowering cost sharing improved, on a short time basis, either mortality or self-reported physical or mental health. However, he points out that conclusions cannot be drawn for long-term outcomes, which is also true of other studies such as the RAND HIE and the Oregon Medicaid lottery. A study of limited duration that is not powered to find adverse outcomes should not be used to conclude that there are no adverse outcomes of forgoing care due to cost sharing, yet that is exactly what the policy community is doing when they say that such forgone care does not adversely effect health.
Shigeoka notes that reducing cost sharing resulted in “large increases in outpatient visits for diagnoses that are defined as Ambulatory Care Sensitive Conditions (ACSCs), for which proper and early treatment reduces subsequent avoidable admissions.” This is really important. Even though these studies are not powered to detect adverse outcomes, he has shown that cost sharing does reduce use of services for conditions that are serious enough to result in hospitalizations and other adverse outcomes should the patients not receive appropriate earlier interventions. Bad conditions which cause bad outcomes for which earlier intervention has been shown to be effective are conditions which should be managed regardless of whether or not there exists a study which is powered enough to be conclusive that there is benefit.
Another important point that he makes is that the increased spending because of lower cost sharing should be compared to the social cost of receiving that care. The social cost includes the deadweight loss of program financing (a loss of economic efficiency – less than optimal use of those funds) and moral hazard (accepting greater risk without bearing the consequences – obtaining more health care when insurance provides protection against the costs). Shigeoka estimates that “the welfare gain of risk protection from lower patient cost sharing is comparable to the total social cost, indicating that the welfare gain from risk protection may fully cover the total social cost in this setting.”
It would be a tradeoff except that there are other gains that favor reduced cost sharing. Although he did not demonstrate short-term health gain from lower cost sharing, he did show that improved management of Ambulatory Care Sensitive Conditions does occur which should certainly produce long-term health gains. Another benefit that is never measured is the reassurance that patients receive when they are informed that their presenting complaints do not represent serious disorders, and are further satisfied when they receive medical interventions that can relieve their symptoms. These interventions may never show up in a list of measured processes and outcomes, but they are still fundamental benefits of health care.
Another problem is that these cost sharing studies tend to evaluate patients’ decisions on whether or not to access care. If there were zero cost sharing, these clinical scenarios would still represent a relatively small part of our total health care utilization. The 80 percent of health care that is utilized by the 20 percent of individuals who have more serious problems is care that is quite insensitive to cost sharing, having already exceeded the deductibles. Yet the policy community tends to extrapolate the percentage of savings from forgone care in these largely outpatient situations to our entire health care spending, resulting in preposterous estimates of potential savings.
Further, he reminds us of one on the most fundamental principles of all: “Patients with high medical spending benefit substantially from financial protection against risk due to lower cost sharing.” And isn’t this really about the patient?
]]>