Health Affairs, 25, no. 4 (2006): 1070-1078 doi: 10.1377/hlthaff.25.4.1070 © 2006 by Project HOPE	New Online   Senate Health Reform Bill   Rewarding Providers   Public Option Policy Brief   Health Reform & Abortion   Delivery System Reform

This Article Abstract Figures Only Reprint (PDF) Submit a response to this article Alert me when this article is cited Alert me when Comments are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to Citation Manager Reprints & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (10) Citing Articles via Google Scholar Google Scholar Articles by Remler, D. K. Articles by Glied, S. A. Search for Related Content PubMed PubMed Citation Articles by Remler, D. K. Articles by Glied, S. A. Related Collections Health Reform Insurance Coverage Managed Care Business Of Health Health Spending Consumer Issues Insurance Market Related Blog Posts

Health Tracking

MARKETWATCH

How Much More Cost Sharing Will Health Savings Accounts Bring?

Abstract

 
Proponents of health savings accounts (HSAs) contend that they will reduce medical expenditures. In practice, however, the effect of HSAs, and the high-deductible health plans that must accompany them, will depend on the actual provisions of those plans and of the plans they replace. We show that typical plans in the market today already contain substantial cost sharing. We find that many HSA/high-deductible arrangements would actually reduce cost sharing for many groups. In particular, the group responsible for half of all medical spending would see no change or a decline in cost sharing at the margin and on average.

A considerable literature evaluates the possible benefits and costs of various configurations of MSAs from the last period when they were of major policy interest.⁴ Although the simulation methods vary and the policies examined differ somewhat from one another and from the policies today, none of the simulations predicted major impacts on medical spending. As we demonstrate here, examination of the cost-sharing features of HSA/high-deductible plans and typical plans in the current market show that substantial spending reductions are even less likely today.

What Are Health Savings Accounts?

Top What Are Health Savings... Health Insurance Plans Today Forms Of Cost Sharing:... Where's The Money? Conclusions NOTES

 
The conditions. Under the HSA legislation, Americans who select a high-deductible health insurance plan may establish, make contributions to, and make payments from a tax-free HSA. The plan must include a yearly deductible of $1,000–$5,000 for an individual or $2,000–$10,000 for a family. An eligible person or his or her employer may contribute to the account up to the value of the deductible (but not more than $2,600 for an individual or $5,100 for a family). The yearly out-of-pocket maximum is capped at $5,000 for an individual or $10,000 for a family.

Contributions to HSAs may be made from pretax income, as long as a person is eligible (that is, covered only by a high-deductible plan). Withdrawals that are used to pay for medical expenses, incurred while a person is eligible, are not taxable. Withdrawals used for other purposes are taxed and subject to an additional 10 percent penalty. Balances accumulate tax-free over time (including during periods when a person is not covered by a high-deductible plan).⁵ Thus, HSAs are a way of extending the tax subsidy available for health insurance premiums to out-of-pocket medical spending.

The rationale. Health economists have long criticized the tax subsidy to health insurance premiums as encouraging too much health insurance. The favorable tax treatment of health insurance was seen as "worse" than that of, say, mortgage interest payments, because it had the further effect of encouraging health care spending to be made through insurance, rather than out of pocket. According to this theory, third-party payment for health care insulates people from feeling the cost when treatment decisions are made, because patients and doctors are spending other people’s money; this, in turn, increases overall medical care spending. Empirically, the RAND HIE found that more-generous insurance does indeed increase medical spending.⁶

The question. A reduction in health care spending is frequently touted as HSAs’ primary advantage. Whether or not this would occur and to what extent depends on how responsive people are to cost sharing, the form and extent of cost sharing in place now, the form and extent of cost sharing under HSAs/high-deductible plans, who would face the increased cost sharing under such plans, and what share of medical spending those people would be responsible for.

Health Insurance Plans Today

Top What Are Health Savings... Health Insurance Plans Today Forms Of Cost Sharing:... Where's The Money? Conclusions NOTES

 
BLS surveys. According to the Bureau of Labor Statistics (BLS), cost sharing at the point of service has been rising in the past several years.⁷ Cathy Baker found that in 2000, mean annual deductibles were $334 (individual) for full-time participants in non–health maintenance organizations (HMOs) in private industry, compared with $219 in 1992–93.⁸ The mean out-of-pocket maximum spending was $1,803 (individual) in 2000 but only $1,416 in 1992–93.⁹

The most recent BLS survey, for 2002–03, is based on somewhat different questions and so is not fully comparable with the earlier survey.¹⁰ It found that 79 percent of non-HMO plans had a deductible (a mean of $240); 81 percent had an out-of-pocket maximum (a mean of $1,688); and the mean coinsurance among everyone was 12–16 percent (Exhibit 1).

The 2005 KFF/HRET survey asked explicitly about high-deductible plans that qualify for HSAs and found that the mean deductible among these plans was $1,901 and the mean out-of-pocket maximum was $2,551.¹² To compare such plans with typical employer-based plans, we used the individual plan-level data from the KFF/HRET survey to determine what share of people with employer-sponsored insurance have as much cost sharing as these plans have, or more (columns 3 and 4 of Exhibit 1). We found that 5 percent of insured workers have deductibles of $1,000 or more, 21 percent have no out-of-pocket maximum, and 55 percent have no out-of-pocket maximum or a maximum of $2,000 or more.

Plans in the RAND HIE versus plans today. In contrast to the substantial amounts of cost sharing in health plans today, the most generous plan in the RAND HIE provided entirely free care with no coinsurance, no deductible, and thus no out-of-pocket maximum. The second-most-generous plan had 25 percent coinsurance, no deductible, and an out-pocket maximum of $1,000 ($2,185 in 2000 dollars), or 5 percent of family income ($1,480 in 2000 dollars for the mean-income family), whichever was lower.

Plans today typically also have substantial managed care features. For example, utilization review (UR) was rare in the 1970s but is common today even in "non–managed care" indemnity, fee-for-service plans.¹³ Thus, health care plans today already contain both substantial cost sharing and managed care measures that are likely to reduce spending.

Forms Of Cost Sharing: Marginal And Average Prices

Top What Are Health Savings... Health Insurance Plans Today Forms Of Cost Sharing:... Where's The Money? Conclusions NOTES

 
The prototypical comprehensive plan today has a deductible of $350, a coinsurance rate of 20 percent, and an out-of-pocket maximum of $1,800. What are the cost-sharing consequences of such a policy coupled with the existing provisions for the tax-deductibility of health insurance premiums?

Marginal price. For amounts up to the deductible of $350, the individual must pay every dollar out of pocket with after-tax dollars (no deduction available).¹⁴ Therefore, to receive an additional dollar’s worth of medical care services, a person must forgo a dollar’s worth of consumption somewhere else, such as for education, entertainment, food, or housing. Economists refer to the out-of-pocket amount that must be sacrificed for additional health care as the marginal price of health care. Here the marginal price is one dollar of other goods for each dollar of health care.

Once the deductible is reached, the insured person pays the coinsurance rate and so obtains another dollar of medical care by forgoing only twenty cents of consumption elsewhere. In this range, the marginal price of health care is twenty cents of other goods and services sacrificed for every dollar of health care.

This marginal price continues until the out-of-pocket maximum is reached, which occurs at a total medical spending level of $7,600.¹⁵ After that point, all further care costs nothing out of pocket. From that point on, the marginal price of health care is zero (Exhibit 2).¹⁶

View larger version (17K): [in this window] [in a new window] EXHIBIT 2 High-Deductible Versus Comprehensive Plan (Without HSA): Relationship Between Medical Spending For An Individual And Spending For All Other Goods And Services

Impact of a high-deductible plan without an HSA. Moving to a high-deductible plan is expected to increase cost sharing. We take as our prototypical high-deductible policy a simple one with no coinsurance, a deductible of $2,500, and an out-of-pocket maximum of $2,500.¹⁸ Although the high-deductible policy has greater cost sharing both at the margin and on average for spending up to $2,500, after that, the more comprehensive policies have greater cost sharing on the margin (Exhibit 2). If the deductible is increased without increasing the out-of-pocket maximum, the maximum is met at a lower level of total medical spending, reducing the level of medical spending after which the marginal cost of medical care is zero. Thus, the maximum out-of-pocket level and its interaction with the other cost-sharing provisions are critical.

Furthermore, the same out-of-pocket maximum will affect a different range of medical spending under a pure deductible system than it will under a coinsurance system. To reach $2,500 in out-of-pocket spending under a plan with a $2,500 deductible, the beneficiary must spend a total of $2,500 on medical care. Beyond that level, additional care does not impose any out-of-pocket cost on the beneficiary. To reach the same out-of-pocket maximum in a plan with no deductible but with 20 percent coinsurance, the beneficiary must spend a total of $12,500 on medical care before reaching the level where no additional out-of-pocket costs are incurred. Using coinsurance rather than deductibles permits a much higher spending level to be affected for any given maximum level of out-of-pocket spending.¹⁹

Effect of HSA/high-deductible plans on cost sharing. With an HSA, all out-of-pocket expenses become tax-advantaged.²⁰ HSAs level the playing field between medical care provided through insurance and that provided through out-of-pocket spending (from the HSA) by making both tax-favored, neutralizing the distortion between insurance-funded care and out-of-pocket-funded care. However, as currently designed, HSAs increase the overall distortion between medical care and all other goods and services. Mark Pauly has described this type of policy as "two wrongs" and questioned whether "two wrongs make a right."²¹

Specifically, with an HSA/high-deductible plan, until the deductible is met, one dollar of medical care can be purchased with a pretax dollar, resulting in a price on the margin for medical care of (1–marginal tax rate). Because HSA contributions are shielded from state and federal income taxes, as well as the payroll taxes of Social Security and Medicare, this can be a substantial rate.

To illustrate the effect of this subsidy, we took a prototypical total marginal tax rate of 40 percent. Exhibit 3 illustrates the budgetary impacts of the HSA/high-deductible plan and comprehensive policies using our prototypical plans and tax rate. Critical levels of medical care spending, where either the marginal or the average price of health care changes places under the two regimes, are marked with dotted lines.

View larger version (19K): [in this window] [in a new window] EXHIBIT 3 High-Deductible Plan With HSA Versus Comprehensive Plan: Relationship Between Medical Spending For An Individual And Spending For All Other Goods And Services

One of the most striking features is how raising the deductible without further increasing the out-of-pocket maximum substantially moves the level of spending at which all further care no longer comes from out of pocket. To the extent that there are choices to be made about what to do in expensive episodes, cost sharing will have been greatly undermined. Moreover, the RAND experiment shows that the largest effects on behavior occur in the shift from no coinsurance to 25 percent coinsurance. Therefore, increasing cost sharing from zero to 25 percent is likely to have a larger effect than further increases.

Where’s The Money?

Top What Are Health Savings... Health Insurance Plans Today Forms Of Cost Sharing:... Where's The Money? Conclusions NOTES

 
Health care spending is heavily concentrated among a few people.²² Moreover, a sizable share of this concentration persists over time, with people with high expenditures continuing to be high spending for many years.²³ Since HSAs will cause cost sharing to increase for some and to decrease for others, we examined how many people are in each of spending category and what share of health care spending the categories represent.

Using the 2001 Medical Expenditure Panel Survey (MEPS), we found that only 30 percent of individuals and 16 percent of all medical spending are for those who face increased cost sharing both on average and on the margin; only 18.7 percent of spending is for those who will face increasing cost sharing on the margin; and only 42.2 percent of spending is for those who face any kind of increased cost sharing, whether on the margin or on average (Exhibit 4). Moreover, fully half of spending is for those who face reduced cost sharing on average. Thus, when considering the plans in existence today and comparing them with the types of plans associated with the new legislation, it is not clear that HSAs live up to their advertised increase in cost sharing.

If people are not required to use HSAs, not all will choose them. Two of the earlier MSA studies also looked at the likely effect on spending, incorporating a different selection of MSAs and plans depending on health status. One study estimated that spending would rise 1–2 percent; another, that it would fall 6.8 percent.²⁵ The American Academy of Actuaries (AAA) predicted a decrease of 2–11 percent.²⁶ Given the further increases in cost sharing that have occurred since the 1990s, we would expect even somewhat smaller effects than those predicted. A recent AAA study that used the experience of similar health reimbursement arrangements (HRAs) to predict the likely effect of HSAs also predicted only modest effects.²⁷ Finally, early examination of consumer-driven health plans, which include greater cost-sharing features, finds that relative to HMOs, they increase medical spending.²⁸

Conclusions

Top What Are Health Savings... Health Insurance Plans Today Forms Of Cost Sharing:... Where's The Money? Conclusions NOTES

 
Because today’s health insurance plans also contain substantial utilization review of hospitalizations, the most important margin for cost saving under the RAND HIE (having a hospitalization or not) is likely to be less price-sensitive. Today’s health insurance policies are far from the generous, no-questions-asked policies of the RAND HIE era. The point is not that the lessons of the RAND experiment have changed—people still behave in the same way—but that the contracts are different, because insurers have already incorporated the HIE’s message.

Current HSA/high-deductible plans represent an increase in cost sharing for those in the middle of the spending distribution but a decrease for those at the very low end and in much of the high end. In fact, maintaining or introducing an out-of-pocket maximum while increasing the deductible can greatly reduce cost sharing for those high spenders who are responsible for a large share of overall health spending. Thus, we find that current HSA/high-deductible plans fall short of the accompanying rhetoric.

One response to this analysis might be to encourage further increases in cost sharing and give HSA/high-deductible policies more bite. However, as our results show, sizable increases in cost sharing would be required if a large share of medical spending were to be made susceptible to cost sharing. Raising incentives for cost-consciousness necessarily increases financial risk, by the very nature of those incentives, and it might undermine the access to care that we wish to preserve.²⁹ Both political and market realities indicate that people value reductions in financial risk and access to care when they are seriously ill. Whether the benefits of increasing the cost sharing of HSA/high-deductible plans outweigh the costs depends on the magnitudes of those effects.

Editor's Notes

 
Dahlia Remler (Dahlia_Remler{at}baruch.cuny.edu) is an associate professor in the School of Public Affairs, Baruch College, City University of New York. Sherry Glied is a professor and chair of the Department of Health Policy and Management, Mailman School of Public Health, at Columbia University in New York City.

The authors thank Bisundev Mahato for helpful research assistance; Cathy Baker, Isadora Gil, and Jeremy Pickreign for discussions about the survey methods and questions; and Kathy Swartz and participants in the Columbia Health Alliance seminar, the Mount Sinai Medical School Health Policy Seminar, and the Association for Public Policy Analysis and Management Fall 2005 conference for useful comments. This work was supported (for Glied) by the Commonwealth Fund and (for Remler) (in part) by the City University of New York PSC–CUNY Research Award Program.

NOTES

Top What Are Health Savings... Health Insurance Plans Today Forms Of Cost Sharing:... Where's The Money? Conclusions NOTES

 

1. For Bush administration proposals, see, for example, White House, "Fact Sheet: Guidance Released on Health Savings Accounts (HSAs)," 22 December 2003, http://www.whitehouse.gov/news/releases/2003/12/20031222-1.html (accessed 10 April 2006).For policy analysts’ proposals, see, for example, J.F. Cogan, R.G. Hubbard, and D.P. Kessler, "Healthy, Wealthy, and Wise," Hoover Digest, no. 3 (2004): 161–167; M. Feldstein, "Health and Taxes," Wall Street Journal, 19 January 2004; and D. Gratzer, "HSA Man vs. Healthzilla," Wall Street Journal, 12 October 2004.
2. Current proponents include Gratzer, "HSA Man vs. Healthzilla." Original proponents include J.C. Goodman and G.L. Musgrave, Patient Power: The Free-Enterprise Alternative to Clinton’s Health Plan (Washington: Cato Institute, 1994).
3. J.P. Newhouse, "Consumer-Directed Health Plans and the RAND Health Insurance Experiment," Health Affairs 23, no. 6 (2004): 107–113.[Abstract/Free Full Text]
4. See, for example, American Academy of Actuaries, Medical Savings Accounts: Cost Implications and Design Issues (Washington: AAA, May 1995); E.B. Keeler et al., "Can Medical Savings Accounts for the Nonelderly Reduce Health Care Costs?" Journal of the American Medical Association 275, no. 21 (1996): 1666–1671[Abstract/Free Full Text]; L. Ozanne, "How Will Medical Savings Accounts Affect Medical Spending?" Inquiry 33, no. 3 (1996): 225–236[Web of Science][Medline]; D. Zabinski et al., "Medical Savings Accounts: Microsimulation Results from a Model with Adverse Selection," Journal of Health Economics 18, no. 2 (1999): 195–218[CrossRef][Web of Science][Medline]; and L.M. Nichols, M. Moon, and S. Wall, Tax-Preferred Medical Savings Accounts and Catastrophic Health Insurance Plans: A Numerical Analysis of Winners and Losers, Research Report no. 406690 (Washington: Urban Institute, 1996).
5. Further details are available in S.A. Glied and D.K. Remler, The Effect of Health Savings Accounts on Health Insurance Coverage, Pub. no. 811 (New York: Commonwealth Fund, 2005).
6. J.P. Newhouse and the Insurance Experiment Group, Free for All? Lessons from the RAND Health Insurance Experiment (Cambridge, Mass.: Harvard University Press, 1993).
7. C.A. Baker, "Cost Sharing in Medical Insurance Plans," 31 March 2004, http://www.bls.gov/opub/cwc/content/cm20040326ar01p1.stm (accessed 17 May 2006); Bureau of Labor Statistics, National Compensation Survey: Employee Benefits in Private Industry in the United States, 2000, January 2003, http://www.bls.gov/ncs/ebs/sp/ebbl0019.pdf (accessed 5 June 2006); Henry J. Kaiser Family Foundation and Health Research and Educational Trust, Employer Health Benefits: 2004 Annual Survey, Section 7: Employee Cost Sharing, 9 September 2004, http://www.kff.org/insurance/7148/sections/index.cfm (accessed 10 April 2006); and V. Goff, Consumer Cost Sharing in Private Health Insurance: On the Threshold of Change, Issue Brief no. 798 (Washington: National Health Policy Forum, 2004).
8. The 1992–93 value is $251, adjusted to 2000 dollars using the gross domestic product (GDP) deflator. Baker, "Cost-Sharing in Medical Insurance Plans."
9. The 1992–93 value is $1,627 in 2000 dollars. The out-of-pocket maximums in the BLS tables, ibid., do not include the deductible. We added the mean deductible to the total out-of-pocket maximum. The deductibles and out-of-pocket maximums are the averages among those with such provisions.
10. BLS, National Compensation Survey: Employee Benefits in Private Industry in the United States, 2002–2003, January 2005, http://www.bls.gov/ncs/ebs/sp/ebbl0020.pdf (accessed 20 February 2005).
11. These results include HMOs, which often do not have coinsurance that would expose members to very large out-of-pocket spending. However, 17 percent of those in conventional plans, 23 percent of those in POS plans, and 11 percent of those in PPOs have no out-of-pocket maximum.
12. G. Claxton et al., "What High-Deductible Plans Look Like: Findings from a National Survey of Employers, 2005," Health Affairs 24 (2005): w434–w441 (published online 14 September 2005; 10.1377/hlthaff.w5.434).
13. D.K. Remler et al., "What Do Managed Care Plans Do to Affect Care? Results from a Survey of Physicians," Inquiry 34, no. 3 (1997): 196–204[Web of Science][Medline]; S.A. Glied, "Managed Care," in Handbook of Health Economics, ed. A.J. Culyer and J.P. Newhouse (Amsterdam: North Holland, 2000), 707–753; and G.P. Mays, G. Claxton, and J. White, "Managed Care Rebound? Recent Changes in Health Plans’ Cost Containment Strategies," Health Affairs 23 (2004): w427–w436 (published online 11 August 2004; 10.1377/hlthaff.w4.427).
14. Some people hold flexible spending accounts (FSAs) through their employers, which allow deductibles and other out-of-pocket health care spending to be made with pretax dollars. FSAs have a number of provisions that make them less attractive than HSAs.
15. The spending at which all further out-of-pocket payments cease is [MaxOOP – (1 – coinsurance) (deductible)]/coinsurance.
16. The zero marginal price (and the other marginal prices) refer only to the financial price; medical care also has opportunity costs, particularly travel and time costs.
17. E.B. Keeler and J.E. Rolph, "The Demand for Episodes of Treatment in the Health Insurance Experiment," Journal of Health Economics 7, no. 4 (1988): 337–367[CrossRef][Web of Science][Medline]; and Newhouse and the Insurance Experiment Group, Free for All?
18. This prototypical policy differs somewhat from the KFF/HRET data but is actually more constraining.
19. See M. Feldstein, "Balancing the Goals of Health Care Provision," Presentation at the International Health Economics Association meeting, 6 January 2006, http://healtheconomics.org/ihea-news/06-01-11/index.html (accessed 10 April 2006).
20. We assume that individuals keep their MSAs fully funded, with sufficient funds to cover the maximum out-of-pocket spending. However, depending on the specific terms of the HSA and plan, this probably would not occur in the initial start-up period, and those with high medical spending from the start of an HSA might not be able to reach this point.
21. M.V. Pauly, An Analysis of Medical Savings Accounts: Do Two Wrongs Make a Right? (Washington: AEI Press, 1994).Extending the tax subsidy to out-of-pocket health care spending encourages the purchase of health insurance plans with greater cost sharing. Under certain assumptions, the ensuing higher cost sharing will result in lower medical spending. See W. Jack and L. Sheiner, "Welfare-Improving Health Expenditure Subsidies," American Economic Review 87, no. 1 (1997): 206–221[Web of Science]; and B.E. Dowd and R. Feldman, "Voluntary Reduction in Health Insurance Coverage: A Theoretical Analysis," Eastern Economic Journal 13, no. 3 (1987): 215–312.
22. W.W. Yu and T.M. Ezzati-Rice, "Concentration of Health Care Expenditures in the U.S. Civilian Noninstitutionalized Population," Statistical Brief no. 81, May 2005, http://www.meps.ahrq.gov/papers/st81/stat81.pdf (accessed 10 April 2006); and M.L. Berk and A.C. Monheit, "The Concentration of Health Care Expenditures, Revisited," Health Affairs 20, no. 2 (2001): 9–18.[Abstract/Free Full Text]
23. A.C. Monheit, "Persistence in Health Expenditures in the Short Run: Prevalence and Consequences," Medical Care 41, no. 7 Supp. (2003): III53–III64.[Medline]
24. See, for example, Keeler et al., "Can Medical Savings Accounts for the Nonelderly Reduce Health Care Costs?"; Ozanne, "How Will Medical Savings Accounts Affect Medical Spending?"; and Nichols et al.,, Tax-Preferred Medical Savings Accounts.
25. Keeler et al., "Can Medical Savings Accounts for the Nonelderly Reduce Health Care Costs?"; and Zabinski et al.,, "Medical Savings Accounts."
26. AAA, Medical Savings Accounts.
27. AAA, The Impact of Consumer-Driven Health Plans on Health Care Costs: A Closer Look at Plans with Health Reimbursement Accounts (Washington: AAA, January 2004).
28. S.T. Parente, R. Feldman, and J.B. Christianson, "Evaluation of the Effect of a Consumer-Driven Health Plan on Medical Care Expenditures and Utilization," Health Services Research 39, no. 4, Part 2 (2004): 1189–1210.[CrossRef][Web of Science][Medline]
29. J.A. Nyman, "Is ‘Moral Hazard’ Inefficient? The Policy Implications of a New Theory," Health Affairs 23, no. 5 (2004): 194–199.[Abstract/Free Full Text]

                      What's this?

Related Blog Posts:

This article has been cited by other articles:

				D. M. Herrick Consumer-Driven Health Care: The Changing Nature of Health Insurance American Journal of Lifestyle Medicine, September 1, 2009; 3(5): 394 - 406. [Abstract] [PDF]

				C. L. Barry, M. R. Cullen, D. Galusha, M. D. Slade, and S. H. Busch Who Chooses A Consumer-Directed Health Plan? Health Aff., November 1, 2008; 27(6): 1671 - 1679. [Abstract] [Full Text] [PDF]

				T. Brennan and L. Reisman Value-Based Insurance Design And The Next Generation Of Consumer-Driven Health Care Health Aff., March 1, 2007; 26(2): w204 - w207. [Abstract] [Full Text] [PDF]

				M. B. Buntin, C. Damberg, A. Haviland, K. Kapur, N. Lurie, R. McDevitt, and M. S. Marquis Consumer-Directed Health Care: Early Evidence About Effects On Cost And Quality Health Aff., November 1, 2006; 25(6): w516 - w530. [Abstract] [Full Text] [PDF]

				M. G. Bloche Consumer-Directed Health Care N. Engl. J. Med., October 26, 2006; 355(17): 1756 - 1759. [Full Text] [PDF]