Can Reform's Prevention Incentives Help to Bend the Cost Curve?

Article excerpt

On my various trips to Boston for conferences and meetings, I make a point of having dinner with a close cousin. Our conversations invariably turn to the issue of health care and its reform. While hardly an advocate of comprehensive health reform, my cousin has taken the position that reform would be unnecessary if people took care of themselves. More specifically, he sees much of our current health care spending crisis as emanating from our sedentary lifestyles, poor health habits, lack of discipline regarding exercise and food consumption, and failure to follow recommendations for annual physical exams and other periodic preventive care. "The body is like a car," I recall him saying. "You need to maintain it and every once in a while you have to look under the hood."

My cousin does have a point regarding our reluctance to comply with good health practices and recommended preventive care, and I suspect that many others feel the same. Indeed, there can be ample personal rewards for such behavior as well as rewards to society at large. In the context of health reform, promoting preventive care and health-enhancing behaviors are part of the arsenal of cost-containment strategies in the Patient Protection and Affordable Care Act (PPACA), and a recent study sees the use of recommended preventive services as having the potential to reduce loss of life and achieve cost savings (Maciosek et al. 2010). However, in light of the fanfare and push for preventive care and other health-related practices, a critical issue is whether we can expect individuals to respond to reform's incentives for preventive activities, whether the vagaries of human behavior will impede such responses, and most importantly, whether such care can indeed be expected to reliably yield cost savings. In what follows, I explore the issue of whether we should expect incentives for more preventive care to yield substantive savings, and thus to what extent we can expect prevention and good health behaviors to make a meaningful contribution to bending the cost curve.

The PPACA and Incentives for Prevention

The PPACA includes a variety of provisions directed toward the use of preventive health services and wellness activities. A number of these measures entail administration and information gathering: creating administrative bodies to coordinate federal prevention efforts, establishing a national prevention strategy, expanding funds for public health programs, developing evidence-based recommendations for using clinical and community preventive services, and devising outreach campaigns for preventive benefits, among others (this and the following summary are drawn from the Kaiser Family Foundation 2010). The law also includes provisions to eliminate out-of-pocket costs for 41 preventive services rated "A" or "B" by the U.S. Preventive Services Task Force (2010). It does so by eliminating cost sharing for such services covered by Medicare, and by increasing federal medical assistance to states (by one percentage point) that offer A and B preventive services under Medicaid and also remove cost sharing. Additionally, the law authorizes Medicare coverage for personalized prevention plan services, provides incentives for Medicare and Medicaid beneficiaries to enroll in and complete behavior modification programs, and requires Medicaid to cover tobacco cessation programs for pregnant women. Finally, consistent with cost-containment goals, the PPACA permits the Secretary of the Department of Health and Human Services (DHHS) to modify or eliminate Medicare coverage of preventive services as recommended by the task force.

Reform also eliminates out-of-pocket costs for A and B preventive services for enrollees in qualified private health plans (beginning in September), and such plans are also required to provide, with no out-of-pocket costs, recommended immunizations, preventive care for children and adolescents, and additional preventive care and screening for women. Beginning in 2014, employers can reward employees through measures such as premium discounts and waivers of cost sharing (worth 30% of the cost of coverage) for enrolling in a wellness program and meeting specific health-related standards. Finally, enrollees in plans from state health insurance exchanges will also be free of out-of-pocket costs for A and B preventive services.

While the evidence-gathering provisions of health reform will accumulate information over time on the benefits of prevention and appropriate health behaviors, the short-run policy response relies primarily on changing the out-of-pocket costs for preventive care. On the surface, this appears to be a logical and straightforward approach to encouraging additional use of preventive services since many currently insured people incur out-of-pocket costs, and those who will become newly insured in 2014 currently face the full costs of such care. As national data reveal, both insured and uninsured groups exhibit fairly low rates of preventive service use. For example, data from the Medical Expenditure Panel Survey (MEPS) indicate that in 2005, only 40.4% of privately insured men ages 50 to 64 received a prostate-specific antigen test within the last year, and less than two-thirds (62.8%) did so within the last two years (Brown 2009). MEPS data for 2005 also indicate that nearly two-thirds (64.1%) of privately insured adults between ages 50 and 64 never had a colonoscopy, with rates of non-use generally higher for those on Medicare and considerably higher for those without health insurance (Soni 2007).

Demand Responses and Consequences

Since the post-reform change in out-of-pocket costs for preventive services will likely represent more than a marginal change for many people, it is difficult to use conventional estimates of price responsiveness to assess the magnitude of the demand response. Assume for the moment that the reform-induced price change elicits a large demand response from both insured and previously uninsured people, resulting in an increase in total spending for preventive care. The key question is whether these additional expenditures will yield benefits (cost savings) at least commensurate with new spending or, in the lexicon of health reform, whether we will obtain "value for money." Despite the confidence by some architects of reform that positive returns will be forthcoming, doubts remain.

As Russell (2007, 2009) and Cohen, Neumann, and Weinstein (2008) have reminded us, global statements about the cost savings from preventive care interventions have been overreaching and misleading. These writers note that cost-effectiveness analysis has generally revealed that preventive health interventions usually increase medical spending, adding more to costs than they save. In reviewing several hundred cost-effectiveness studies, these authors point out that less than 20% of preventive interventions saved money and the likelihood of cost savings from prevention was roughly equivalent to that from treatment.

There are several reasons for such a counter-intuitive assessment. For example, screening, a commonly recommended preventive intervention, typically requires periodic testing of large populations, thereby incurring expenditures each year or over every few years. Some individuals will develop the disease in question regardless of the screening, thereby negating the cost savings. Additionally, most individuals screened will not develop the disease, so society may incur large screening costs but fail to capture any savings from timely interventions or avoided treatment costs. Further, false positives are likely to appear, resulting in additional spending for repeat screening or more testing (Russell 2009) and psychological distress for the patient and family. Finally, and while difficult to assess, some abnormalities detected by screening and ultimately treated might have remained dormant, suggesting that any subsequent treatment costs incurred might have been unnecessary. This is not to discount the benefits to individuals when screening successfully identifies the early stage of a disease. Rather, the evidence suggests that we must be cautious in our expectations regarding any cost savings that may be forthcoming.

The summary evidence noted here and the reasons for a potentially low payoff to certain types of preventive care serve as a reminder to carefully target relevant at-risk populations and apply interventions proven to be cost effective. Indeed, the estimates of cost savings produced by Maciosek et al. (2010) are based on increasing the use of specific preventive services from current rates to 90% (which may be formidable given currently low use rates), and represent a savings of only .2% of U.S. personal health care spending. While social policy may dictate more widespread use of preventive interventions such as screening, since cost-effectiveness analysis does not capture other benefits (both economic and psychological) associated with the early detection of illness, policymakers need to recognize that the consequences of such preventive actions may not necessarily lead to large savings in health care costs.

Will Enrollees Respond to Reduced Cost Sharing?

I have noted that it might be difficult, a priori, to assess how eliminating cost-sharing provisions might affect the demand for preventive care given the potentially large reduction in out-of-pocket costs. In this regard, findings from the RAND Health Insurance Experiment (HIE) conducted during the 1970s may be instructive (Newhouse et al. 1993). The HIE randomly assigned study participants in several geographic locations to health plans with varying cost-sharing provisions, including a "free" plan without any coinsurance. Thus, comparisons of preventive care use (Pap smears, mammography, and sigmoidoscopy for adults, as well as immunizations, TB tests, and well-care examinations for children) between the free health plan and plans with cost sharing provide a rough approximation of how much completely eliminating out-of-pocket costs might spur the use of such services.

In this matter, the summary findings of the experiment reported by Newhouse and colleagues (page 178) were not encouraging. Overall, they reported that the utilization of preventive care by HIE participants was well below that recommended by health professionals, and that the wide discrepancy between these standards and enrollee behavior could not be explained primarily by differences in cost sharing. More telling was the fact that the percentage of people using any preventive care was only marginally higher for those on the free health plan compared to those with cost-sharing provisions, and that the majority of adult males on this free plan used no preventive services over the three-year study period. Such findings, along with the more recent estimates regarding the relatively low use of preventive services among insured populations and the weak impact of certain state insurance mandates for preventive care benefits (e.g., Li et al. 2010), do not bode well for strategies that seek to encourage preventive service use through changes in health insurance provisions.

Why Don't People Respond to Available Preventive Care Benefits?

Apart from the concern over the cost-saving potential of preventive care, the general question of why individuals fail to use such services, even when they don't have to pay anything, remains puzzling. As Ariely (2009) notes in reporting results from a variety of behavioral experiments, consumers respond quite significantly to the prospect of obtaining something for free, even if such a choice does not appear to be rational or in their best interests. If a zero price is so powerful in inducing consumption, why might it be less effective for health services like preventive care?

One obvious explanation is that the strong response to a zero price typically reflects the utility we obtain from consuming pleasurable or desirable commodities. By contrast, many preventive services are rarely viewed as pleasant or desirable experiences, often requiring drawing blood, fasting (for cholesterol screening), and invasive procedures that are unpleasant and take time from obligations considered more pressing. As Ariely and others (e.g., Kenkel 2000) note, such disutility leads to procrastination or the outright avoidance of such tests. Such procrastination may also reflect a self-control problem in which individuals exhibit what behavioral economists refer to as time-varying preferences, committing to preventive services (or other health-related behaviors) when they are scheduled well into the future, but failing to engage in such activities when the time for compliance is imminent.

As Kenkel (2000) has noted, there can be a variety of other behavioral reasons that limit responses to valued preventive care. For example, some individuals may be very present-oriented (in the economists' parlance, have high discount rates) and fail to recognize the value of "investing" in preventive activities whose benefits are likely to appear in the future. Those with health insurance may exhibit behavior characterized as ex ante moral hazard, whereby the insurance protection against illness-related financial loss reduces incentives for prevention. As Kenkel writes, should the price of insurance depart from actuarial fairness and fail to reflect an individual's preventive activities--as in the community rating of our current employment-based system and in some non-group insurance markets--individuals may have little pecuniary incentive to participate in prevention. Other factors leading to under-investment in prevention include high rates of preventive activities by others (e.g., vaccinating more people for a contagious disease reduces the likelihood that any one individual will contract the illness, and hence lowers one's perceived need for the vaccination); lack of consumer information and inaccurate assessments regarding the consequences of poor preventive behaviors and their associated risks (e.g., underestimation by some smokers, particularly the young, of the risks of smoking-related illnesses); and the failure of some providers to supply relevant information on prevention, the poor availability of basic primary care in some geographic locations and among some vulnerable groups, and the time costs of seeking such care. Finally, individual factors, such as age (the payoff period for prevention declines with age) and health status (those with superior health "endowments" may invest in more preventive care) can also affect decisions to engage in preventive care and activities.

Policies to Encourage Preventive Care and Behaviors

Given the low usage rates of preventive services and the potential weak response to eliminating cost sharing, what can public policy do to encourage health-enhancing behaviors and spur greater use of effective and cost-saving preventive interventions? While this issue could easily become the subject of another commentary, I will briefly mention some common approaches and alternatives.

As regards altering health behaviors, a frequently used strategy has been the application of "sin" taxes, typically levied on cigarette smoking and alcohol consumption, and more recently considered on soft drinks and for broader application to calorie-rich and energy-dense foods. Using smoking as an example, such taxes are usually justified on the basis of the health effects smoking may impose on others, as well as the possibility that plan enrollees with smoking-related illnesses can impact the group's health insurance premiums. More recently, the justification for taxes has been based on the need to counter problems that some individuals have in committing to prudent health behaviors due to lack of self-control and procrastination (Gruber 2002/2003). Additionally, much discussion has been directed to imposing excise taxes on soft drinks as a way to curtail demand, with a recent, carefully controlled experiment suggesting that a significant increase in soft drink prices could elicit a sizable decline in consumption (Block et al. 2010). Such a finding is consistent with a weak responsiveness shown in a number of state studies to relatively small price increases via excise taxes.

While such tax-based approaches have been effective in discouraging smoking in certain groups and appear promising in other applications, they also confront two important issues. First, policymakers must recognize the possibility that a tax-induced price increase may also result in individuals substituting other undesirable consumption (e.g., shifting from cigarettes to smokeless tobacco); second, there must be sufficient political will to increase prices via taxes and set them at meaningful levels. The recent failure of New York to impose an excise tax on soft drinks illustrates how formidable the political process can be in impeding such a strategy.

Several other non-price strategies also warrant mention. The problem of lack of information regarding the fat and calorie content of specific food products can be addressed through nutritional postings on menus. Such postings have the potential to reduce calorie consumption by altering food choices (as in the widely noted case of Starbucks' calorie posting), and by inducing restaurants to change their menu choices (Bollinger, Leslie, and Sorenson 2010; Black 2010). The PPACA will implement such a strategy by requiring chain restaurants and vending machines to post the nutritional content of their food items.

Regarding incentives for using preventive services, several ideas for supplementing the reduction in out-of-pocket costs also warrant consideration. For example, the time costs of obtaining routine preventive care such as blood tests could be lowered by scheduling appointments at testing laboratories. This has recently happened in a facility that I use and has resulted in greatly reduced waiting times. Other innovative approaches might include periodic reminders for "scheduled maintenance," bundling medical tests akin to the servicing procedures followed by automobile dealers (Ariely 2009), and taking advantage of the influence of social networks and peers to encourage the use of preventive care and the adoption of good health habits (Singer 2010).

Disseminating Evidence-Based Recommendations: A Caution

As already mentioned, health reform also will support the development of evidence-based recommendations for preventive care and seek to disseminate such findings. In this regard, care must be exercised to avoid controversy, confusion, and political charges of health care rationing when introducing new guidelines that change well-regarded existing practices. The fallout from the controversial release of new guidelines for mammograms serves as a valuable lesson.

On Nov. 16, 2009, the task force's new recommendation--that women in their 40s should not have routine mammograms and that women between 50 and 74 should have the exam every two years rather than annually--resulted in a firestorm of controversy due in part to its conflict with many women's understanding of established practice and with recommendations from the American Cancer Society. What followed was a rather clumsy scene in which the Obama administration distanced itself from the recommendations and scurried to assure women that they could continue their usual practice, that public and private payers would continue to cover such procedures, and that state insurance mandates for mammograms (in all states but Utah) would remain intact (New York Times 2009). This unfortunate incident illustrates the perils that can occur when evidence-based recommendations seeking to reduce costs are not appropriately previewed and vetted across relevant interest groups, depart radically from long-standing practice, and do not adequately anticipate how their introduction may impact targeted population groups.


Following through on reform's promise of cost containment will be perhaps the most daunting challenge for its proponents and will have critical implications for our nation's fiscal health. In doing so, the contribution of every cost-saving provision in the PPACA will be scrutinized, additional ways to economize on reform's subsidies and new spending provisions will be debated, and accusations of health care rationing will inevitably appear. Encouraging the use of preventive care and health-enhancing behaviors has been viewed as a common-sense way of contributing to such cost containment. However, given the uncertainties about the cost-saving efficacy of prevention, the inertia associated with individual decisions concerning prevention, and the potential for confusion and misinterpretation when announcing new evidence-based recommendations, we must proceed judiciously and realistically in our expectations regarding prevention's impact on bending the cost curve.

So to my cousin and other like-minded individuals who view prevention as a path to achieve significant cost savings and avoid the necessity of reform, I would say the following. Be conscientious in your efforts to improve and monitor your health and encourage others to do so. But recognize this strategy's uncertainty as a more global cost-containment effort and acknowledge the need for the multiple cost-saving approaches that reform will bring. In many cases, these too will be uncertain, politically unpopular, and cumbersome to implement. However, it is through reform's evaluation of the proposed cost-saving interventions, its coordinated efforts to change the incentives faced by providers and patients, and its measures to eliminate wasteful spending that we will be able to reign in the unsupportable growth in national health care spending.

In another matter, I am delighted to report that a recent article published in Inquiry and authored by Sandra L. Decker, a senior service fellow at the National Center for Health Statistics, has been honored with two prestigious awards by the Centers for Disease Control and Prevention (CDC). "Medicaid Physician Fees and Ambulatory Care of Medicaid Patients," published in Inquiry Vol. 46, No. 3 (Fall 2009), is the recipient of the 2010 Charles C. Shepard Science Award for scientific achievement in prevention and control, as well as the 2010 Kaafee Billah Memorial Award in Economics Research. The Billah award recognizes outstanding scientific contributions in economic research at CDC, specifically for applying economic methods, theories, and knowledge to analyze a public health problem.

Sandra's paper investigated the effect of the generosity of Medicaid physician payment levels on the volume and site of ambulatory care for Medicaid patients compared to the privately insured. Her findings indicate that cuts in Medicaid fees lead to reductions in the number of physician visits for Medicaid patients compared to those with private insurance, and a shift from care at physician offices toward hospital emergency and outpatient departments. Her work is highly relevant for provisions under the Patient Protection and Affordable Care Act, which will raise physician fees under Medicaid to Medicare levels for some selected services.

From all of us at Inquiry---congratulations, Sandra!


Ariely, D. 2009. Predictably Irrational." New and Revised Edition. New York: Harper Perennial.

Black, J. 2010. Chain Restaurants such as KFC, UNO, and Starbucks are Finding that Calories Count. Washington Post January 6:E01.

Block, J. P., A. Chandra, K. D. McManus, and W. C. Willett. 2010. Point-of-Purchase Price and Education Intervention to Reduce Consumption of Sugary Soft Drinks. American Journal of Public Health 100(8): 1427-1433.

Bollinger, B., P. Leslie, and A. Sorenson. 2010. Calorie Posting in Chain Restaurants. National Bureau of Economic Research Working Paper #15648. Cambridge, Mass.: National Bureau of Economic Research (NBER).

Brown Jr., E. 2009. Screening for Prostate Cancer with the Prostate-Specific Antigen Test--United States, 2006. Statistical Brief #233. January. Rockville, Md.: Agency for Healthcare Research and Quality.

Cohen, J. T., P. J. Neumann, and M. C. Weinstein. 2008. Does Preventive Care Save Money? Health Economics and the Presidential Candidates. New England Journal of Medicine 358(7):661 663.

Gruber, J. 200212003. Smoking's "Internalities." Regulation (Winter):52 57.

Kaiser Family Foundation. 2010. Focus on Health Reform: Summary of New Health Reform Law. April 21. Washington, D.C.: Kaiser Family Foundation.

Kenkel, D. S. 2000. Prevention. In Handbook of Health Economics, Volume 1B, A. J. Culyer and J. P. Newhouse, eds. Amsterdam, The Netherlands: Elsevier North Holland.

Li, R., P. Zhang, L. Baker, and D. Hartsfield. 2010. The Impact of State Mandatory Insurance Coverage on the Use of Diabetes Preventive Care. BMC Health Services Research 10:133.

Maciosek, M. V., A. B. Coffield, T. J. Flottemesch, N. M. Edwards, and L. I. Solberg. 2010. Greater Use of Preventive Services in U.S. Health Care Could Save Lives at Little or No Cost. Health Affairs 29(9): 1656-1660.

Newhouse, J. P., and the Insurance Experiment Group. 1993. Free for All? Lessons from the RAND Health Insurance Experiment. Cambridge, Mass.: Harvard University Press.

New York Times. 2009. Editorial: The Controversy over Mammograms. November 20.

Russell, L. B. 2007. Prevention Potential for Slowing the Growth of Medical Spending. Report to the National Coalition on Health Care. October.

--, 2009. Preventing Chronic Disease: An Important Investment, But Don't Count on Cost Savings. Health Affairs 28(1):42~15.

Singer, N. 2010. Better Health, With a Little Help from Our Friends. New York Times September 19.

Soni, A. 2007. Screening Colonoscopy Among U.S. Noninstitutionalized Adult Population Age 50 and Older, 2005. Statistical Brief 4~188. November. Rockville, Md.: Agency for Healthcare Research and Quality.

U.S. Preventive Task Force Recommendations. 2010. prevention/taskforce. Accessed August 18, 2010.

Alan C. Monheit, Ph.D.