|
| |||||||||||||||||||||||
During the past 30 years, while health care expenditures have risen in the United States from 4 percent to 14 percent, many cost-control strategies have been tried to varying degrees, without success4,5,6. The inflation of costs, influenced by technology, specialization, and a variety of other factors, has continued unabated6. Thoughtful criticisms of administrative cost-control solutions have been advanced, and it has been noted that these measures ration productive and nonproductive activities alike7. Despite their merits, proposed programs of reform introduce new problems, such as reduced access, increased costs, rationing, or adverse economic effects on small businesses2,3,4,8,9,10. On the face of it, broadening access and reducing costs at the same time is difficult.
A Theoretical Solution -- Reducing the Need and Demand for Medical Services
If there were no illness and no accidents, health care costs for a society would theoretically be zero. For much of this century, the decrease in acute illness and the proportional increase in chronic disease have fueled inflation in medical costs11. Preventing chronic illness would offer hope of a reduction in demand: if a coronary-artery bypass graft procedure costs $50,000, then avoiding that procedure could save up to $50,000, depending on the cost of the intervention, on whether the procedure is postponed or prevented, and other offsetting factors12. Approaches involving self-management could potentially yield similar benefits: if a visit to the emergency department for a cold, including cultures, x-ray films, and antibiotics, costs $130, then that amount is saved if the visit is not made. If they are based on well-documented and proved interventions, health policies directed at reducing the burden of illness constitute a positive approach: individuals become healthy and society may save money13,14.
How can we reduce the burden of illness and thus the need and demand for medical services? Ideally, a society of healthy people does not smoke, does not consume alcohol to excess, exercises regularly, eats wisely, uses seat belts, treats hypertension, provides other preventive health services, and sees that care at the end of life is humane15. People assume more responsibility for their own health by requesting health services when such services can be of help and avoiding them when they cannot. There is emphasis on both disease prevention and collective individual restraint. To be sure, there are offsetting factors involving cost as well, including the direct and indirect costs of the interventions themselves. With programs of early detection (screening) in particular, it has been notoriously difficult to document cost savings16. The same can be said of drug treatment for control of hypertension or reduction in cholesterol levels17,18,19. Not all "preventive" interventions save money.
We believe one requirement is broad access to wisely designed programs of health promotion, in which the concept of health promotion is expanded to include a goal of cost reduction. This expanded concept directly addresses the challenge of preventing illness as well as that of reducing health care costs.
The Potential for Reducing Demand
A health policy directed at reducing demand would be unlikely to make a major contribution to lowering costs if a number of conditions were not present: if preventable illness made up only a small fraction of the demand, if risky behavior were not expensive in terms of lifetime medical costs, if approaches involving self-management did not reduce costs, if the present system already linked the use of resources closely to the requirements of illness, or if health-promotion programs in the workplace increased overall health care costs.
Much Disease Is Preventable
Preventable illness makes up approximately 70 percent of the burden of illness and the associated costs. Well-developed national statistics such as those outlined in Healthy People 2000, Health U.S. 1991, and elsewhere document this central fact clearly1,15,20,21. McGinnis and Foege have carefully reclassified the causes of death in the United States, using underlying actual causes rather than the traditional disease-oriented classifications; they found that preventable causes account for eight of the nine leading categories and for 980,000 deaths per year. (McGinnis JM, Foege WH: personal communication).
Risky Behavior Costs Money
Lifetime medical costs, which average approximately $225,000 per person, are clearly linked to health habits22,23,24. For example, the lifetime costs for smokers, despite their shorter lives, are higher than those for nonsmokers22,25 by approximately one third20,22. Breslow and Breslow23 have demonstrated that poor health habits are strongly associated with greater burdens of illness and that these effects are similar in magnitude to the effects on mortality. Yen et al.26. showed that self-reported health habits strongly predicted annual claims costs over the next three years; people at low risk had average claims of $190, whereas those at high risk had claims averaging $1,550. Using cross-sectional data from the National Health Interview Survey, Wetzler and Cruess found that increased physical activity was associated with fewer visits to the doctor27. Tsai et al. confirmed statistically significant associations between smoking habits and overall morbidity; overall morbidity was higher by 60 percent among smokers28. Sokolov demonstrated that in a large corporate setting, people with three or more risk factors on a list that included smoking, obesity, hypertension, hypercholesterolemia, and diabetes had claims costs that were double those of people who had no risk factors29,30.
Variability in Regional Costs Implies Slack in the System
Wennberg and others have documented striking regional differences in the use of services31,32. Cesarean-section rates in the United States have risen from 5 percent in 1965 to over 25 percent in 1988; moreover, current rates range from 9.6 to 31.8 percent in different settings33. Hospital expenditures per capita are twice as high in Boston as in New Haven. In some communities in Maine, over 50 percent of men undergo prostatectomy, as compared with only 15 percent in nearby communities, yet the health outcomes in these men appear similar31,32,34. In the case of elective hospitalizations, admission rates correlate with the number of hospital beds per capita rather than the incidence of illness35,36. Educating the consumer so that more informed decisions are made decreases the frequency with which certain procedures are performed31,32. When patients are given information and alternatives, they have been shown, on average, to select less invasive (and less expensive) strategies than their physicians32. Health maintenance organizations cost as much as 20 percent less than traditional programs of health care delivery, with no discernible negative effects on health outcomes37.
Self-Management Can Result in Savings
Multiple studies have demonstrated that providing medical consumers with information and guidelines about self-management can lower rates of use of services, often by 7 to 17 percent, in association with modest interventions37,38,39,40,41. These interventions offer objective guidelines to help a person decide whether medical assistance is required for a particular problem and provide information about home treatment when appropriate. They appear to work through two mechanisms: better information and increased confidence that much illness can be self-limited. Education that increases confidence about health decisions has been shown to reduce the costs of long-term health care, even in people with chronic disease42,43.
Care for Terminal Illness Has Become Extraordinarily Expensive and Inhumane
The costs of medical care in the last year of life are high, and a portion of them represent overly intensive services in terminal illness44. Some 18 percent of lifetime costs for medical care, or over $40,000, is estimated to be incurred in the last year,45 and 29.4 percent of Medicare and Medicaid payments for those over the age of 65 are for people in the last year46. Imminent death is not always predictable, but sometimes the reality is clear. Seventy percent of people request no life-sustaining treatments for themselves when they are dying, and 89 percent desire living wills and other advance directives. Yet only 9 percent have made such directives47,48,49. Schneiderman and colleagues50 point out that not only signed documents but also physician-patient communication is important if dying patients are to receive less unwanted care. Roos and colleagues have discussed the factors associated with expensive and inexpensive terminal care51.
Health Promotion at Work Has Successfully Reduced Costs
A growing literature documents the potential of well-formulated health-promotion programs to decrease health care costs in the workplace38,39,40,41,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69. Multiple studies of such programs have shown substantial decreases in the number of sick days,52,53,54,56,59,60,65,68,69 outpatient costs,52,57,58,61,64,65,66,68,69 and hospitalization costs55,56,57,60,61,64,65,66,69. Many of these studies have used randomized or parallel control groups in similar plants or facilities for soundness of the experimental61,62,67or quasi-experimental53,55,59,63,69 design. Savings have frequently been confirmed by analyses of claims data60,61,62,65,67,69.
A recent review of 28 separate studies of health promotion in the workplace emphasized the effects on documented cost reduction, with the savings generally being three or more times greater than the program costs69. A low-cost intervention delivered by mail61,62 had a higher percentage return. A well-executed study sponsored by the National Institutes of Health in Birmingham, Alabama, showed a 10-to-1 return and held costs to the city constant over a five-year period67. Increasing collaboration by universities and corporations has substantially improved the quality of recent studies70. There already are model programs that improve health and decrease costs. It is not knowledge that is lacking, but penetration of these programs into a greater number of settings.
Redefining Health Promotion
We propose that health-promotion programs attempt to improve both physical and financial health. The central goal remains improvement in health habits, and ultimately the postponement and prevention of major chronic illnesses require reduction in risk factors. Available data, however, suggest a lag of two or three years between improvement in health habits and signs of better health and reduced costs12,71. Some studies have suggested that costs may even be increased during the first year of a program intended to improve health habits72.
There are, however, elements of a broadened health-promotion model that have the potential for more immediate savings. Materials on self-management that offer guidance in the appropriate use of services have been shown to produce savings in the first year,38,39,40,41 particularly those that raise people's confidence in making health-related decisions42. Programs intended to improve a person's level of confidence in dealing with chronic illness have been shown to lower costs both immediately and over a four-year period43. Interventions directed at reducing the number of low-birth-weight babies can theoretically have immediate effects: a stay in the intensive care unit for a single low-birth-weight baby can easily cost more than $100,000. Using advance directives, such as a living will or durable power of attorney, that emphasize humane and dignified care at the end of life and are coupled with appropriate communication could reduce costs as well as provide more humane care in a short period47,48,49. Dollar incentives for the appropriate use of health care, a measure sure to be controversial, may yet play a part in reducing costs and may do so over a brief period29,30,67,72.
Who Will Pay?
Widespread dissemination of programs such as those described here costs money -- not a great deal of money in comparison with overall health costs, but some money. The costs should be borne by those who will ultimately have the savings -- that is, those now paying the costs. Insurers, industry, and government can pay out of their potential savings. What is required is a widespread conviction that appropriately designed programs directed at reducing need and demand can actually save money13,14,69.
The Health Project
The Health Project is a voluntary consortium of business leaders, health insurers, policy scholars, and members of government, with representation from the Centers for Disease Control and Prevention, the Office of Health Promotion and Disease Prevention, and present and past administrations. Its members believe that reducing demand and need can have a substantial positive effect on health care costs. We propose a private-public partnership, with nominal federal expenditure, to promote improved health and reduce costs. The consortium believes that there already are programs capable of making a change in demand, but that except for demonstration projects, their availability now extends only to perhaps 1 or 2 percent of the population. Disseminating such programs is the issue.
The consortium plans to review the evidence, identify the best programs, and promote their replication in additional workplaces and communities. In an initial review of more than 200 health-promotion programs in the workplace, there was a convincing documentation of savings in 8: Johnson and Johnson,55 Du Pont,53 Tenneco,52 Blue Shield of California,61,62 Travelers,39,40 Southern California Edison,29,30 the County of Ventura (California), and Coors58. Collectively, these programs had developed many of the features that influence costs directly over the short term. Two had a heavy emphasis on self-management, two used cost incentives, one emphasized advance directives for terminal care, several contained elements directed at enhancing self-confidence with regard to health-related decisions, and several had defined approaches to people at high risk.
Problems and Caveats
Why have solutions aimed at reducing costs by reducing demand not been implemented more widely? Two common answers are the popular belief that high costs are necessary if we are to have the finest medical system in the world and the idea that technology is essential to improved health and longevity. Economists have focused on administrative and financing issues. Across-the-board solutions have been sought, rather than selective ones, partly because they are easier to conceptualize. Advocates of health promotion have themselves caused delays, first by not making cost reduction a primary goal and second by neglecting rigorous economic evaluation. The government has not funded evaluative research to any substantial degree. It can be argued that academic conservatism has held preventive policy to a more stringent standard of proof than that generally applied in other areas of health policy.
In addition, there has been the suggestion that health-promotion programs might actually increase costs or that at best the reductions would be small73,74,75,76. Smoking cessation, for example, might increase costs by promoting greater longevity with its attendant costs. Furthermore, proposals to teach people to use hospitals and doctors less can be seen as raising another kind of barrier to access, even though the intention is quite different. Health-promotion programs have also been viewed as intrusive and as jeopardizing privacy. There has been doubt that large segments of the population can learn to practice self-management. Concern that powerful interest groups (doctors, hospitals, industry, or the medical-industrial complex) will effectively oppose such approaches has engendered a sense of hopelessness.
In the end, however, the payers are consumers, who can learn to act in their own interests to control costs and who ultimately constitute the strongest interest group77. Moreover, the proposals outlined here are not antimedicine; rather, they are in the tradition of a profession devoted to improving health. Ultimately, a service profession must applaud this initiative, and many if not most physicians already support these strategies. It is time to confront concern directly about whether efforts to reduce need and demand can be of real value. The evidence is that they can.
Conclusions
Reducing the need and demand for medical services is theoretically plausible and practically documented, and there is a funding mechanism in place, through the savings accruing to the present payers. The approach complements multiple proposals for the reform of health care financing that are now under consideration, and indeed it is essential to any such plan, for all face the question of costs. The Health Project Consortium believes that widespread implementation requires ever broader collaboration among business, labor, the insurance industry, government, and the university. This approach does not directly address many other important issues in medical reform, including access, overspecialization, and the development of a two-tiered system, although it may provide indirect help in some of these areas. Nor does it, as now conceived, adequately address the issues of health promotion and reduction of demand for services as they affect the unemployed, the uninsured, and the poor; to the extent that it can free funds, however, it may provide benefits for them. Reducing the need and demand for medical services is a positive solution, one that will bring better health for the individual, and that will ultimately lower medical costs.
We are indebted to Lester Breslow, Robert Cihak, Milton Friedman, Sarah Tilton-Fries, Victor Fuchs, Halsted Holman, Paul Leigh, Deborah Lubeck, Harold Luft, and Kenneth R. Pelletier for critical review and suggestions.
Source Information
From Stanford University, Stanford, Calif. (J.F.F.); C. Everett Koop Institute, Hanover, N.H. (C.E.K.); William M. Mercer, New York (C.E.B.); Prudential Insurance Company, Newark, N.J. (P.P.C.); Washington Business Group on Health, Washington, D.
(M.J.E.); Health Net, Woodland Hills, Calif. (R.F.G.); Sokolov Strategic Alliance, Los Angeles (J.J.S.); and the Tau Group, New York (D.W.). In addition to the study authors, the following persons are members of the Health Project Consortium: Charles B. Arnold, M.D., Metlife; Charles R. Buck, Jr., Sc.D., General Electric; Bruce Fried, Clinton Transition Team; Ron Hartwig, Hill and Knowlton; James Harrell, Dept. of Health and Human Services; Karen Ignagni, AFL-CIO; Dorothea R. Johnson, M.D., AT&T; Johannes Kuttner, the White House; James Marks, M.D., Centers for Disease Control and Prevention; Richard F. O'Brien, General Motors; Robert E. Patricelli, Value Health; Roger B. Porter, M.D., Office of Policy Development, the White House; Dallas L. Salisbury, Employee Benefit Research Institute; Jack Shelton, Ford Motor Company; John F. Troy, the Travelers Insurance Companies; and Reed Tuckson, M.D., Charles R. Drew University.
Address reprint requests to Dr. Fries at 1000 Welch Rd., Suite 203, Palo Alto, CA 94304.
References
| |||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |