CONTROLLING HEALTH CARE SPENDING: MORE PATIENT “SKIN IN THE GAME?” David Orentlicher1 I. INTRODUCTION................................................................ 348 II. THE HIGH COSTS OF HEALTH CARE ................................ 350 III. IMPROVING THE RETURN ON OUR HEALTH CARE DOLLAR ....................................................... 352 A. Increasing Patient “Skin in the Game” .................... 352 1. Higher Cost-Sharing for Lower-Value Care .......... 355 2. “Reference pricing” ................................................. 356 3. “Scaled Cost-Sharing” ............................................ 356 B. Limits of Cost Containment ...................................... 357 IV. EMPLOYER WELLNESS PROGRAMS ................................. 357 V. CONCLUSION ................................................................. 362 I. INTRODUCTION While health care cost inflation slowed during the past few years, it has started to pick up again,2 and policy makers have good cause for concern about future increases in health care spending. Moreover, even if future increases moderate, policy makers rightly worry about the already high levels of U.S. spending. The need for effective cost containment strategies in health care persists, even though the Affordable Care Act appears to have had some success at containing health care costs. Health care spending reforms can focus on physician and hospital practices or on patient behavior, and popular reform 1 Samuel R. Rosen Professor and Co-Director, Hall Center for Law and Health, Indiana University Robert H. McKinney School of Law. MD, JD, Harvard University. I am grateful for the excellent editing of the Indiana Health Law Review. 2 Drew Altman, New Evidence Health Spending Is Growing Faster Again, WALL ST. J. (June 11, 2015, 3:14 PM), http://blogs.wsj.com/ washwire/2015/06/11/new-evidence-health-spending-is-growing-faster- again/ [perma.cc/D2YC-YBNQ]. http://dx.doi.org/10.18060/3911.0019 2016 349 CONTROLLING HEALTH CARE SPENDING: MORE PATIENT “SKIN IN THE GAME?” proposals include both approaches. For example, rather than paying physicians and hospitals in terms of the quantity of care that they provide and encouraging the provision of too much care, private insurers and government programs are turning more and more to forms of reimbursement that are based on the quality of care delivered. Insurers often adjust physicians’ compensation based on whether they screen their patients for cancer or high cholesterol, administer recommended immunizations, or achieve good control of blood sugar levels for their patients with diabetes.3 The Affordable Care Act addresses patient behavior by requiring insurers to cover important kinds of preventive care for free.4 That way, people will not be discouraged for financial reasons from seeking early care that can keep them healthier and avoid the need for hospitalizations and other expensive treatments. In this article, I consider an increasingly common strategy that insurers use to influence patient behavior—giving people more “skin in the game.” When medical treatment can be obtained at very low cost, people may be too quick to seek it when they feel sick, visiting their physicians when they would do just as well by staying home. Hence, insurers have raised deductibles5 and co-payments6 and shifted the costs of care to patients in other ways7 in the hope that people will 3 Aparna Higgins, German Veselovskiy & Lauren McKown, Provider Performance Measures in Private and Public Programs: Achieving Meaningful Alignment with Flexibility to Innovate, 32 HEALTH AFF. 1453, 1456-57 (2013). 4 See Patient Protection and Affordable Care Act, Pub. L. No. 111- 148, 124 Stat. 119 (2010) (codified as amended at 42 U.S.C. § 18001 et. seq. (2010)). 5 A deductible refers to the costs of care that the patient pays before insurance kicks in. If the deductible is $500, the patient pays the first $500 in health care costs for the year. 6 A co-payment refers to the patient’s share of costs when care is provided. For example, a visit to the doctor’s office may come with a co- pay of $25, with the insurance company picking up the remainder of the physician’s fees for the visit. Co-payments are similar to co-insurance, under which patients pay a percentage of the costs of care, say twenty percent of the costs of a hospitalization. 7 Insurers also shift more costs to patients by raising the annual cap on the patient’s share of their health care costs from deductibles, co- payments, and co-insurance (the cap on total out-of-pocket spending), as http://dx.doi.org/10.1377/hlthaff.2013.0007 350 INDIANA HEALTH LAW REVIEW Vol. 13:2 become more conscious of the costs of their care. Although concerns about patients seeking too much care are important, common strategies for giving patients more skin in the game have been poorly conceived. There is room for skin-in-the- game strategies to contain high health care spending, but only when they are properly designed. II. THE HIGH COSTS OF HEALTH CARE Health care spending in the United States is approaching 18% of Gross Domestic Product (“GDP”), a level well above other economically-advanced democracies.8 Countries such as Canada, Germany, Switzerland, and Japan spend only about 10 to 11% of GDP on health care.9 And current U.S. spending is very high when compared with past U.S. expenditures. In 1980, health care spending accounted for only 9% of GDP.10 To some extent, higher spending makes sense. The United States is a rich country and therefore can afford to spend more on health care than many other countries. It is probably better for a country to spend its plentiful resources on health care than on yachts or tickets to professional football games. But do Americans get enough bang for their extra health care bucks? Concerns about health care spending are focused not only on the amount of spending but also on the fact that the United States does not appear to get sufficient benefit for all of its extra spending. On many health status metrics, the United States lags other countries. For example, life expectancy in the United States trails that of a wide range of countries, not only including Canada, Germany, Switzerland, well as by providing less coverage for care received from physicians or hospitals that are not in the insurance company’s network (“out-of- network” care). 8 OECD, HEALTH AT A GLANCE 2013: OECD INDICATORS 157 (OECD Publishing, 2013), available at http://www.oecd.org/els/health-systems/ Health-at-a-Glance-2013.pdf [https://perma.cc/KAF2-GPE7]. 9 Id. 10 Snapshots: Health Care Spending in the United States & Selected OECD Countries, KAISER FAMILY FOUND. (April 12, 2011), http://kff.org/ health-costs/issue-brief/snapshots-health-care-spending-in-the-united- states-selected-oecd-countries/ [https://perma.cc/A4KL-GTXA]. 2016 351 CONTROLLING HEALTH CARE SPENDING: MORE PATIENT “SKIN IN THE GAME?” and Japan, but also Italy, Spain, Greece, and the United Kingdom.11 Of course, many factors other than health care affect life expectancy and other measures of health. People in Italy, Spain, and Greece may live longer because they consume a Mediterranean diet.12 Perhaps our higher health care spending helps narrow the gap between the United States and other countries even if it does not eliminate the gap. Indeed, some data suggest that Americans do get value for their health care dollar. For example, five-year breast cancer survival rates are higher in the United States than in Canada, Germany, Japan, and the United Kingdom.13 Similarly, five-year colon cancer survival rates are higher in the United States than in Canada, Germany, and the United Kingdom, though lower than in Japan.14 And empirical data indicate that greater spending on cancer care contributes to the higher survival rates. In a study that considered the benefits and costs of cancer care in the United States and Europe, researchers found that the survival gains from the extra spending on cancer in the United States exceeded the costs of the care.15 In another study, researchers found that reductions in deaths from cancer were greatest in countries where cancer care spending rose the most between 1995 and 2007.16 But other data indicate that we spend our health care dollars inefficiently. For example, asthma hospitalization rates are much higher in the United States than in Canada, 11 OECD, supra note 8, at 25. Infant mortality rates also are better in many other countries, including Japan, Portugal, Spain, Greece, France, Poland, and the United Kingdom. Id. at 37. 12 Francesco Sofi et al., Accruing Evidence on Benefits of Adherence to the Mediterranean Diet on Health: An Updated Systematic Review and Meta-Analysis, 92 AM. J. CLINICAL NUTRITION 1189 (2010). 13 OECD, supra note 8, at 127. 14 Id. at 129. 15 Tomas Philipson et al., An Analysis of Whether Higher Health Care Spending in the United States Versus Europe Is 'Worth It' in the Case Of Cancer, 31 HEALTH AFF. 667, 670-71 (2012) (assuming that an extra year of life has an economic value of $150,000 and comparing the economic value from the increased life expectancy to the costs of care). 16 Warren Stevens et al., Cancer Mortality Reductions were Greatest Among Countries Where Cancer Care Spending Rose the Most, 1995−2007, 34 HEALTH AFF. 562 (2015). http://dx.doi.org/10.3945/ajcn.2010.29673 http://dx.doi.org/10.1377/hlthaff.2011.1298 http://dx.doi.org/10.1377/hlthaff.2014.0634 352 INDIANA HEALTH LAW REVIEW Vol. 13:2 France, Japan, and the United Kingdom, and hospitalization rates for diabetes are much higher than in Canada, Spain, Italy, and the United Kingdom.17 If health care did more in the United States to maintain the health of people with asthma or diabetes, hospitalization rates would look more like those in other countries. And a study that estimated the efficiency of health care systems by comparing health care spending with health status of a country’s residents found that the United States trailed a wide range of countries, from Canada, France, Germany, Italy, and the United Kingdom to Mexico, Colombia, Venezuela, and China.18 There also are domestic data suggesting that much health care spending is wasted. U.S. patients treated in high-cost communities are no healthier than patients treated in low- cost communities.19 Indeed, patients actually might fare better in lower-spending areas.20 III. IMPROVING THE RETURN ON OUR HEALTH CARE DOLLAR There are many ways to improve the efficiency of health care spending. If fee-for-service reimbursement encourages physicians to perform too many surgical procedures, it makes sense to rely more on salary-based compensation. Or a percentage of physicians’ compensation could be based on the extent to which they meet quality-related targets for the health care they provide. For example, physicians would be paid more if more of their patients receive an annual influenza vaccine. A. Increasing Patient “Skin in the Game” Should we also try to improve the efficiency of health care spending by giving patients more “skin in the game?” If patients had to pay a higher percentage of their health care 17 OECD, supra note 8, at 109. 18 David B. Evans et al., Comparative Efficiency of National Health Systems: Cross National Econometric Analysis, 323 BMJ 307, 308-09 (2001). 19 Elliott S. Fisher, Julie P. Bynum & Jonathan S. Skinner, Slowing the Growth of Health Care Costs—Lessons from Regional Variation, 360 NEW ENG. J. MED. 849, 850 (2009). 20 Id. http://dx.doi.org/10.1136/bmj.323.7308.307 http://dx.doi.org/10.1056/nejmp0809794 2016 353 CONTROLLING HEALTH CARE SPENDING: MORE PATIENT “SKIN IN THE GAME?” costs, would people be more likely to refrain from seeking care when they really do not need a doctor’s attention? More importantly, would people take better care of themselves if they had to pay more for their medical treatments? Perhaps, Americans would be healthier, and health costs would be lower, if people were more sensitive to the costs of the care that they receive. By its very nature, health care insurance dulls patient sensitivity to the costs of care. Assume, for example, that a particular treatment costs $100 and provides a value to the patient worth only $75. If the patient were paying the full cost of care, the treatment would be declined. But if insurance covers most of the costs of the care, so the patient would face a co-payment of only $25, the patient would likely choose the care. Getting $75 of value for $25 is a good deal.21 As long as we have health care insurance, patients will not be fully sensitive to the costs of their health care. But cost sensitivity is not an all-or-nothing phenomenon. Even if we cannot make patients fully sensitive to the costs of their care,22 we have to decide on the level of sensitivity. If health care coverage is too generous, people may seek too much care, wasting health care resources. If health care coverage is not generous enough, people may not seek enough care, to the detriment of their health. Many employers, insurers, and analysts think that patients have been insufficiently sensitive to the costs of their care.23 Hence, in recent years, we have seen marked increases in the size of deductibles and co-payments to make patients more sensitive to health care costs.24 Indeed, among employee health care plans, the average deductible for individual coverage more than doubled between 2006 and 21 See David Orentlicher, Cost Containment and the Patient Protection and Affordable Care Act, 6 FLA. INT’L U. L. REV. 67, 71 (2010). 22 Nor would we want them to be. An important reason for having health care coverage is to ensure that people can have good access to health care even when they have limited financial resources. 23 See, e.g., JAMES W. HENDERSON, HEALTH ECONOMICS AND POLICY 8 (6th ed. 2014) (“Fully-insured patients have no incentive to limit their utilization [of health care].”). 24 Higher deductibles and co-payments also offer a way to limit increases in health care insurance premiums. 354 INDIANA HEALTH LAW REVIEW Vol. 13:2 2014, and the percentage of individual plans with deductibles of $1,000 or more nearly quadrupled.25 Is this a good trend? If the goal is simply containing costs, then giving patients more skin in the game may be useful. Raising the patient’s share of health care costs through deductibles, co-payments and other out-of-pocket costs reduces patient demand for care. In the RAND Health Insurance Experiment, in which participants were randomly assigned to health care plans with different levels of cost-sharing, researchers found that higher cost-sharing led to fewer physician visits, fewer prescriptions, and fewer hospitalizations.26 But the reductions in financial costs may come with increases in non-financial costs. In particular, when patients refrain from seeking care because of the costs of care, their health may suffer. Several studies indicate that when patients reduce their demand for care because of costs, they many not distinguish between needed and unneeded care. In the RAND study, for example, there was no adverse impact on health for the average person.27 However, for poor individuals with medical problems, those with free care had better health measures and lower predicted mortality rates than their counterparts who were discouraged from seeking care by their deductibles or co-payments.28 In another study, which involved emergency department care, researchers again found that increased cost-sharing had an adverse effect on health for the poor.29 Higher-income individuals in high-deductible plans reduced their emergency department visits only for “low-severity” services—services that were not urgent and could be provided at a clinic or doctor’s office at a later date.30 But, low-income persons 25 KAISER FAMILY FOUND. & HEALTH RESEARCH & EDUC. TRUST, EMPLOYER HEALTH BENEFITS: 2014 ANNUAL SURVEY 125-26 (2014), available at http://files.kff.org/attachment/2014-employer-health- benefits-survey-full-report [https://perma.cc/HD8U-Y8GS]. 26 JOSEPH P. NEWHOUSE & THE INS. EXPERIMENT GRP., FREE FOR ALL?: LESSONS FROM THE RAND HEALTH INSURANCE EXPERIMENT 338 (1993). 27 Id. at 338-39. 28 Id. at 339. 29 J. Frank Wharam et al., Low-Socioeconomic-Status Enrollees in High-Deductible Plans Reduced High-Severity Emergency Care, 32 HEALTH AFF. 1398 (2013). 30 Id at 1399. http://dx.doi.org/10.1377/hlthaff.2012.1426 2016 355 CONTROLLING HEALTH CARE SPENDING: MORE PATIENT “SKIN IN THE GAME?” reduced visits for both “low-severity” services and the kinds of “high-severity” services that should be treated urgently in an emergency department.31 Or consider a study that analyzed the impact of a new deductible and co-payments for prescription drugs.32 The increases in out-of-pocket costs led low-income persons to reduce their use of both low-value and high-value drugs, and accompanying the reduction in drug use, there was an increase in “serious adverse events” (hospitalizations, nursing home admissions, and deaths).33 While broad increases in patient cost-sharing seem ill- advised because of their adverse effects on patient health, might more targeted increases be useful? Recall in this regard that in the RAND study, greater cost-sharing for the average person led to a reduction in health care spending with no harm to health.34 A few possibilities for targeted cost-sharing come to mind. 1. Higher Cost-Sharing for Lower-Value Care If the goal of patient skin in the game is to discourage unnecessary care while preserving desirable care, then it makes sense to reserve higher cost-sharing for lower-value care. The Affordable Care Act’s requirement of free preventive care is a good model for this approach.35 We want people to receive effective preventive care—a high value kind of care—so the Affordable Care Act prohibits the imposition of any fees on people when they obtain the care. Similarly, to encourage the use of generic rather than more expensive brand-name versions of the same drug, insurers often require higher co-payments for brand-name drugs. As a general matter, heath care policy should remove obstacles to desired behavior while erecting obstacles to undesired behavior. 31 Id at 1403. 32 Robyn Tamblyn et al., Adverse Events Associated with Prescription Drug Cost-Sharing among Elderly and Poor Persons, 285 JAMA 421, 421 (2001). 33 Id. 34 JOSEPH P. NEWHOUSE & THE INS. EXPERIMENT GRP., supra note 26. 35 See Patient Protection and Affordable Care Act, Pub. L. No. 111- 148, 124 Stat. 119 (2010) (codified as amended at 42 U.S.C. § 18001 et. seq. (2010)). http://dx.doi.org/10.1001/jama.285.4.421 356 INDIANA HEALTH LAW REVIEW Vol. 13:2 2. “Reference pricing” The high cost problem is not only a problem of patients receiving unnecessary care; it also is a problem of patients receiving necessary care at excessive prices. Hip replacement surgery might cost $40,000 at one hospital and $80,000 at another hospital with no difference in quality (or possibly lower quality at the higher price). Accordingly, some insurers will reimburse for surgical procedures only at a fixed “reference” price that reflects the fees charged by low-cost, high-quality physicians and hospitals.36 If a patient chooses a more expensive provider of care, the patient is responsible for the difference between the provider’s fees and the reference price. Data on reference-pricing indicate that it leads patients to switch to lower-cost providers.37 It also causes higher-cost providers to lower their fees.38 3. “Scaled Cost-Sharing” The degree to which patients are sensitive to the costs of their care depends on their income and wealth.39 A deductible of $1,000 represents 5% of income for a family earning $20,000, but only 0.5% of income for a family earning $200,000. Or when annual caps on out-of-pocket spending are set at $6,000, they represent 30% of income for a family earning $20,000 but only 3% of income for a family earning $200,000. Hence, standard policies for out-of-pocket costs will likely have a bigger impact on the care-seeking behavior of lower income persons. And as suggested by the previously- discussed studies on the health effects of cost-sharing, lower- income persons may be overly discouraged from seeking care by standard cost-sharing policies. Accordingly, rather than 36 James C. Robinson & Timothy T. Brown, Increases in Consumer Cost Sharing Redirect Patient Volumes and Reduce Hospital Prices for Orthopedic Surgery, 32 HEALTH AFF. 1392, 1392-93 (2013). 37 Id. at 1394-95. 38 Id. 39 Christopher T. Robertson, Scaling Cost-Sharing to Wages: How Employers Can Reduce Health Spending and Provide Greater Economic Security, 14 YALE J. HEALTH POL’Y L. & ETHICS 239, 244 (2014). http://dx.doi.org/10.1377/hlthaff.2013.0188 2016 357 CONTROLLING HEALTH CARE SPENDING: MORE PATIENT “SKIN IN THE GAME?” setting cost-sharing levels at fixed dollar amounts, insurers could calculate deductibles, annual caps, and other forms of cost-sharing as a percentage of income.40 Deductibles could be set at 1% of income, and annual caps could be set at 5% of income. B. Limits of Cost Containment While carefully-designed health insurance reforms can play a significant role in making patients more conscious of costs, these reforms can play only a limited role in cost containment. The impact of cost-sharing strategies dissipates when patients hit their annual cost-sharing maximums. Once a deductible is satisfied, for example, it no longer can have any influence, and once annual caps on total out-of-pocket expenses are exceeded, patients no longer need to worry about other cost-sharing policies such as copayments. Reference pricing would still matter even after annual caps on out-of-pocket spending are satisfied, but estimates indicate that reference pricing would reduce overall spending by less than two percent.41 In sum, it is useful to consider health insurance reforms that encourage greater cost-consciousness among patients, but policy makers will have to look elsewhere for major savings in health care spending. Might other patient-directed policies be useful? This article has focused so far on insurance plan design, but there are ways to influence patient behavior. The next section considers the potential role of employer wellness programs in containing health care costs. IV. EMPLOYER WELLNESS PROGRAMS In addition to lowering health care spending by sending patients higher bills for their visits to the doctor or the 40 Id. 41 PAUL FRONSTIN & M. CHRISTOPHER ROEBUCK, REFERENCE PRICING FOR HEALTH CARE SERVICES: A NEW TWIST ON THE DEFINED CONTRIBUTION CONCEPT IN EMPLOYMENT-BASED HEALTH BENEFITS 10 (Employee Benefit Research Institute 2014), available at http:// www.ebri.org/pdf/briefspdf/ebri_ib_398_apr14.refprcng.pdf [https:// perma.cc/26PW-Y3UA]. 358 INDIANA HEALTH LAW REVIEW Vol. 13:2 hospital, we might lower spending by encouraging patients to take better care of themselves. If people are healthier, they will not need as many appointments with their doctors or admissions to the hospital. Employers are increasingly using the skin-in-the-game approach to promote healthier behavior. Through financial incentives tied to “wellness programs,” the hope is that employees will eat more nutritiously, exercise more regularly, and require less health care.42 Wellness programs typically are divided into (1) screening initiatives and (2) intervention activities.43 Screening initiatives include questionnaires that ask individuals about their diet, exercise, and other health-related matters.44 Screening also can include clinical measurements such as a person’s weight, blood sugar, cholesterol, and blood pressure.45 If people realize that their weight, blood pressure, or other measurements are too high, they can follow up with a physician to see what kinds of action would be helpful. Or they might follow up with the wellness program’s intervention activities. These can include counseling about exercise and diet, smoking cessation programs, gym memberships, and healthy food offerings in cafeterias or vending machines.46 While many employers simply offer their wellness programs alone, other employers combine the programs with financial incentives, sometimes rewarding employees for participation in the programs, at other times rewarding employees for improvement in their weight, blood pressure, 42 Wellness programs can be implemented outside of the workplace. Governments, insurers, and individuals for themselves also can design wellness programs. Kristin M. Madison, Kevin G. Volpp & Scott D. Halpern, The Law, Policy & Ethics of Employers’ Use of Financial Incentives to Improve Health, 39 J.L. MED. & ETHICS 450, 450-51 (2011). But there are important advantages to employment-based programs, in large part because people spend much of their waking time at their workplace. Id. at 455. 43 David Orentlicher, Health Care Reform and Efforts to Encourage Healthy Choices by Individuals, 92 N.C. L. REV. 1637, 1648 (2014). 44 Id. 45 Id. 46 Id. http://dx.doi.org/10.1111/j.1748-720x.2011.00614.x 2016 359 CONTROLLING HEALTH CARE SPENDING: MORE PATIENT “SKIN IN THE GAME?” or other measures of health.47 Under federal law, there is no limit on the magnitude of incentives that can be used to encourage employees to participate in wellness programs.48 While a typical incentive might provide employees with a rebate on their health insurance premiums of $100 or $200 for checks of weight, blood pressure, blood sugar, and cholesterol, an employer could offer much higher rebates for participation—or impose surcharges of any amount for non- participation.49 Employers also might want to link their financial rewards or penalties to results. For example, a rebate or surcharge on insurance premiums might be tied to the losing of weight, the reduction of blood pressure, or the achievement of other health targets. For incentives tied to the satisfaction of health targets, the incentive may not be any higher than 30% of the cost of the employee’s health insurance coverage (with a 50% maximum for meeting smoking cessation targets).50 While wellness programs are sound in principle—an ounce of prevention is worth a pound of cure—there are significant problems with these programs in practice. For example, employers often do not choose effective programs.51 And even when wellness programs are successful, their results are modest. In one study, only one-third of employees lost at least five percent of weight.52 In another study, 47 Id. at 1648-49. 48 Id. (provisions regarding financial incentives for wellness programs are included in HIPAA and ACA.). 49 Id. 50 Id. at 1649. To protect employees from unfair discrimination, employers must offer reasonable alternative standards. For example, if the incentive is tied to weight loss, and a worker has a genetic disease that makes it very difficult to lose weight, the employer would have to revise the target for the employee. See id. 51 Karen Chan Osilla et al., Systematic Review of the Impact of Worksite Wellness Programs, 18 AM. J. MANAGED CARE e68, e78 (2012) (finding positive outcomes only one half of the time for wellness programs that were studied with a randomized controlled trial). 52 Caryn Zinn et al., A “Small-Changes” Workplace Weight Loss and Maintenance Program: Examination of Weight and Health Outcomes, 54 J. OCCUPATIONAL & ENVTL. MED. 1230, 1234-35 (2012). http://dx.doi.org/10.1097/jom.0b013e3182440ac2 360 INDIANA HEALTH LAW REVIEW Vol. 13:2 participants lost less than one percent of weight on average.53 And these modest results may be exaggerated. When programs are voluntary, “selection bias” may exaggerate their effectiveness.54 Hence, randomized controlled studies of wellness programs find smaller impacts than do non- randomized studies. In one review of wellness program studies, researchers found that exercise programs generated positive results 62% of the time, but only 43% of the time when the studies involved a randomized control group for comparison.55 Unfortunately, experts have not yet figured out how to design wellness programs that reliably deliver a high level of effectiveness. Ineffective programs are not only wasteful, they also can be harmful. In one of its most important provisions, the ACA promotes access to health care coverage by eliminating insurance premium surcharges for people with cancer, diabetes, heart disease, or other “pre-existing” medical conditions.56 No longer does a person’s health status affect the ability to afford health care coverage. But financial incentives tied to losing weight, lowering blood pressure, reducing blood sugar, or meeting other health targets will impose greater costs on persons with health problems, thereby undermining ACA’s protection of persons with pre- existing medical conditions. Indeed, an analysis of employer wellness programs suggests that savings on health care spending from the programs may simply reflect the shifting of costs to employees with higher risks of illness.57 ACA’s goal of affordable health care is further undermined by the fact that when person with health problems bear greater costs, the greater costs fall disproportionately on persons who are poor. 53 Susan B. Racette et al., Worksite Opportunities for Wellness (WOW): Effects on Cardiovascular Disease Risk Factors after 1 Year, 49 PREVENTIVE MED. 108, 110 tbl. 2 (2009). 54 Selection bias refers to the possibility that a voluntary program will attract especially motivated participants whose experiences will be different from the people who choose not to participate in the program. 55 Osilla et al., supra note 51, at e69. 56 See 42 U.S.C. § 300gg–4(a) (2016). 57 Jill R. Horwitz, Brenna D. Kelly & John E. DiNardo, Wellness Incentives in the Workplace: Cost Savings through Cost Shifting to Unhealthy Workers, 32 HEALTH AFF. 468, 469 (2013). http://dx.doi.org/10.1016/j.ypmed.2009.06.022 http://dx.doi.org/10.1377/hlthaff.2012.0683 2016 361 CONTROLLING HEALTH CARE SPENDING: MORE PATIENT “SKIN IN THE GAME?” While financial incentives tied to wellness programs often are ineffective and even harmful, there are some wellness incentives that can be useful. A number of features are important: When incentives are tied to short-term progress, they seem to work better than incentives calculated on an annual basis. People respond more readily to immediate rewards and penalties than to delayed rewards and penalties.58 Thus, in one study of financial incentives for weight loss, participants received lottery tickets or accumulated “deposit contract” rewards on a daily basis if they met their weight loss goals,59 and the incentives were effective at encouraging weight loss during the four months of the study.60 As this study also suggested, incentives may need to be maintained indefinitely. Within several months after the study ended, there was no significant difference in weight loss between the participants and a control group of people who had not received the financial incentives.61 Of course, this may simply reflect the fact that any strategies for weight loss need to be continued indefinitely, just as treatments for high blood pressure, diabetes, and other chronic medical conditions need to be continued indefinitely. Finally, program designers need to consider whether their incentives should be implemented as penalties for failure or rewards for success. Penalties often are more effective than rewards at eliciting changes in behavior. People worry more about losing something they already have than about gaining something they do not have.62 On the other hand, people 58 Orentlicher, supra note 43, at 1643, 1652. 59 Kevin G. Volpp et al., Financial Incentive-Based Approaches for Weight Loss: A Randomized Trial, 300 JAMA 2631, 2632-33 (2008) (describing a study with deposit contracts where participants committed a small amount of money each day that was matched at a higher amount by the study, with the total dollars paid to participants who achieved their weight loss goals). While deposit contract awards could be earned on a daily basis, the awards were actually paid out on a monthly basis. Id. at 2632. 60 Id. at 2634-35. 61 Id. at 2635. 62 Scott D. Halpern et al., Randomized Trial of Four Financial- Incentive Programs for Smoking Cessation, 372 NEW ENG. J. MED. 2108, 2109 (2015). http://dx.doi.org/10.1001/jama.2008.804 http://dx.doi.org/10.1056/nejmoa1414293 362 INDIANA HEALTH LAW REVIEW Vol. 13:2 prefer to be rewarded for success than penalized for failure, so reward-based incentives may be a more effective strategy overall.63 V. CONCLUSION In recent years, concerns about health care cost containment have led employers, insurers, and governments to give individuals more skin in their health care game. But the interest in patient incentives for cost consciousness has exceeded the benefits that these incentives can deliver. When used in a limited and properly designed fashion, the incentives can achieve some cost savings. But the overall savings will be small, and they can easily be offset by their own costs if the incentives are not well-designed. 63 Id. at 2114.