Do consumers use less health care and buy less health insurance when prices increase? If so, market-based reforms can be used to slow the growth of health costs and expand access by making health insurance less costly. If not, we cannot rely on the price system to accomplish these objectives.
Michael A. Morrisey reviews what we do and do not know about how consumers respond to price changes in a new book recently published by the NFIB Research Foundation, Price Sensitivity in Health Care: Implications for Health Care Policy. In his book, Morrisey discusses why health care consumers in some markets are more responsive to prices than in others, and the implications of these findings for health policy.
| || || |
William J. Dennis Jr., NFIB Research Foundation
| || |
Michael A. Morrisey, University of Alabama at Birmingham
| || |
James A. Baumgardner, Congressional Budget Office
| || || |
Paul B. Ginsburg, Center for Studying Health System Change
| || |
Robert B. Helms, AEI
| || || |
Do consumers use less health care and buy less health insurance when prices increase? If so, market-based reforms can be used to slow the growth of health costs and expand access by making health insurance less costly. If not, we cannot rely on the price system to accomplish these objectives. Michael A. Morrisey reviews what we do and do not know about how consumers respond to price changes in a new book recently published by the NFIB Research Foundation, Price Sensitivity in Health Care: Implications for Health Care Policy. At a September 15 AEI book forum, Morrisey discussed why health care consumers in some markets are more responsive to prices than in others, and the implications of these findings for health policy.
Michael A. Morrisey
University of Alabama at Birmingham
In health care markets, price sensitivity is a measure of how responsive consumers are to changes in price of health services or health insurance. When price sensitivity is alluded to in policy discussions, debate over moral hazard inevitably follows suit. Moral hazard is the idea that insurance engenders wasteful consumption of medical services by sheltering the true cost of health care. While this is a generally accepted premise, the extent to which demand fluctuates when price decreases is still a moot topic. However, with respect to health services, a wealth of valuable research has been performed.
The RAND health insurance experiment was conducted thirty years ago but remains the most relevant study to date. To eliminate the bias associated with those who seek out coverage, the study randomly assigned health insurance to people regardless of their need for health services. In a consistent framework, it addressed a range of services, including physician visits, hospital care, mental health services, and prescription drugs. More importantly, subsequent research has largely confirmed the RAND findings, suggesting that policy solutions can be derived from the estimates even today.
The RAND study is most easily summarized by looking at what was discovered in connection with physician visits. When there was no out-of-pocket obligation, consumers made an average of 4.55 visits to a doctor throughout the year. If the coinsurance rate increased to 25 percent, physician visits declined to 3.33 per year. This trend held up when examining the use of other medical health services and interaction with the health care system in general. In essence, as consumers shoulder more of the bill, their health expenditures correspondingly decrease.
In terms of prescription drugs, the 1970 RAND analysis is slightly dated, since innovative drugs have greatly changed the dynamics of health care in recent years. Nonetheless, the study revealed that those who received free drugs filled 23 percent more prescriptions than those who paid 25 percent of the price. More current research, which has focused on second- and third-tier drugs plans, has found that copays reduce drug use at each tier. This effect was particularly evident at the non-preferred brands tier (third-tier), where consumers are required to pay more out of pocket. In addition, there was greater price sensitivity for drugs with available substitutes and for drugs taken intermittently to treat symptoms as opposed to drugs taken for ongoing care. Policymakers should bear in mind that prescription drug use goes hand in hand with physician visits. Implementation of the Medicare drug benefit will not only heighten prescription drug use, but will also lead to more patient-physician interaction. The result is increased cost pressure on an already unsustainable program.
In theory, price sensitivity would vary with income level, or more accurately, with opportunity cost of time. If low income individuals were faced with a high coinsurance rate, they would end up paying a relatively large portion of the whole price (money price plus time price). Consequently, low income consumers would be more price responsive than high opportunity cost of time or high income consumers. According to the RAND study, the introduction of a 25 percent coinsurance rate caused low income individuals to use 13 percent fewer health services, but resulted in only a 6 percent drop among high income individuals. The policy implication is that a small copay can go a long way with low opportunity cost of time consumers.
Conclusions about health insurance and price sensitivity are based not on one overarching study, like that from RAND, but on a body of research performed over the last decade. A number of studies have demonstrated that insured employees are remarkably sensitive to the out-of-pocket size of the premium when choosing which plan to take. This implies that employee premium contributions can be successfully used to motivate employee choice of plans.
Economic theory suggests that workers pay for health insurance in the form of lower wages and benefits. Jay Bhattacharya and M. Kate Bundorf found that obese workers are paid less but have more generous health coverage, suggesting that indeed the wage-benefit tradeoff exists. Further empirical evidence was provided by Dana Goldman, et al., who investigated the effect of a 10 percent premium increase on employees with a flexible cafeteria-style compensation plan. Employees responded to the increase by either opting for less generous health insurance benefits, lower take-home wages, or reductions in other benefits such as disability coverage or pension plans. If employers are agents for their workers and reflect employee preferences, one should expect that employers are making similar adjustments in compensation. Mandating employer-sponsored health coverage at the federal or state level would force workers to sacrifice wages or benefits for health insurance.
Since employer-sponsored health insurance is not subject to federal or state income taxes or to Social Security or Medicare payroll taxes, employees and employers have a strong incentive to shift compensation toward untaxed health insurance. As tax rates rise, employers find it advantageous to offer and employees to purchase more health insurance than otherwise needed. To enhance price sensitivity in this market, we should simply replace the tax exclusion of employer-sponsored health insurance with tax cuts or tax credits that apply to all or certain income groups.
Congressional Budget Office
Price Sensitivity in Health Care: Implications for Health Care Policy is a thorough compendium of the empirical evidence on this topic. The bottom line is that people, workers, firms, and patients respond to prices in health care. The book is a valuable source of references and provides a general overview for those who are interested in studying the dynamics of the health care market.
However, to model the market in a more technical sense, one needs to read each study in its entirety to determine exactly how elasticity is defined. The use of the term is quite inconsistent across the studies. Regression specification should be examined to clarify the meaning of elasticity in the particular study. The book also does not expand upon the precise location of the mean of the data. This is important since predictions and forecasting become more uncertain as you move further away from the mean of the data. While relevant, these modeling issues go beyond the scope of the book.
In light of the RAND study, there appears to be a phenomenon of diminishing returns with respect to cost-sharing. When a consumer with no out-of-pocket expense is suddenly faced with a 25 percent coinsurance rate, utilization is greatly reduced. Although, as the coinsurance rate increases to what would be considered a high deductible, the extent of the drop in utilization tapers off. In reality, the majority of beneficiaries already have some form of coinsurance within their plans, thus tweaking the rates would not cause the profound effect seen with RAND’s zero deductible population.
The book contains a paucity of research on the relationship between cost-sharing and health outcomes. When faced with higher coinsurance rates, do consumers cut back on appropriate or inappropriate care? Recent studies find that both occur, but the impact of these decisions on overall health status is still debatable.
While the book focuses on consumer response to price, there are actually two contrasting approaches for dealing with moral hazard. The first strategy is to face consumers with more out-of-pocket costs and allow them, along with their physicians, to decide how much care to receive. The other side of the spectrum involves physician--rather than patient--perception of price. Patients can obtain care through firms, such as Health Maintenance Organizations, which use financial or nonfinancial means to compel doctors and medical groups to efficiently assess the cost of health care. If consumers were fully informed about the expected costs and benefits of medical care, and their decisions were not distorted by the tax code, which approach would best replicate consumer decisions?
Paul B. Ginsburg
Center for Studying Health System Change
The importance of the RAND Health Insurance Experiment (HIE) is that it convinced many skeptical policymakers that price sensitivity does indeed exist in the health care market. Prior and subsequent research has produced results consistent with those of the RAND-HIE. However, today the issue is not the magnitude of the sensitivity, but rather whether consumers are making good or bad decisions when faced with these incentives. As a result, research now tends to focus on the quality of patient decisions. When perusing both current and past literature, it is important to differentiate between the types of patient cost-sharing and the types of incentives, as some are far better at empowering consumers to make decisions in their best interest.
The RAND-HIE is slightly outmoded, since health spending is now a much higher percentage of gross domestic product (GDP). National health expenditures were 7 percent of GDP in the 1970s and are projected to be 15.6 percent of GDP in 2005. A deductible that has the same relationship to income today will have a smaller percentage effect on medical care spending and pose a more significant financial burden on patients. Cost-sharing may be a less powerful tool today than it was in the past, since overall costs have grown exponentially over the past two decades.
While cost-sharing can control spending, many consumers would prefer to opt for a more restrictive plan that requires less patient cost-sharing. Our household surveys have asked hypothetical questions concerning the willingness of consumers to sacrifice provider choice for cost savings. Past surveys concluded that a slight majority would be inclined to make this tradeoff. However, the 2005 survey showed a substantial increase across all income and demographic groups in the proportion of people willing to forgo provider choice for cost savings.
A key driver of health care spending trends is advancements in medical technology. Will greater price sensitivity feed back into research and development decisions made by medical equipment and pharmaceutical industries that create these advancements? Will increased cost-sharing impede the development of innovative medical technologies?
Regarding market level changes, will there be a stronger behavioral response to increased patient cost-sharing when everyone has a consumer-driven plan? Recently, media and literature have been devoted to educating consumers on how to make economical choices in the health care market. If the medical system evolves to be more cognizant of cost, cost-sharing will have a greater impact on reducing health care spending.
AEI research assistant Elizabeth DuPre and AEI intern Lauren Dewey prepared this summary.