February 2003: Letters

At the Intersection of Health, Health Care and Policy Cite this article as: David M. Driesen, Victor Fuchs and Sarah Rosen Frank Rational Public Policy Health Affairs, 22, no.1 (2003):278-279 doi: 10.1377/hlthaff.22.1.278

The online version of this article, along with updated information and services, is available at: http://content.healthaffairs.org/content/22/1/278.full.html

For Reprints, Links & Permissions: http://healthaffairs.org/1340_reprints.php E-mail Alerts : http://content.healthaffairs.org/subscriptions/etoc.dtl To Subscribe: http://content.healthaffairs.org/subscriptions/online.shtml

Health Affairs is published monthly by Project HOPE at 7500 Old Georgetown Road, Suite 600, Bethesda, MD 20814-6133. Copyright © 2003 by Project HOPE - The People-to-People Health Foundation. As provided by United States copyright law (Title 17, U.S. Code), no part of Health Affairs may be reproduced, displayed, or transmitted in any form or by any means, electronic or mechanical, including photocopying or by information storage or retrieval systems, without prior written permission from the Publisher. All rights reserved.

Not for commercial use or unauthorized distribution Downloaded from content.healthaffairs.org by Health Affairs on May 23, 2011 by guest

Letters

Letters

W

e welcome your responses to papers that appear in Health Affairs. We ask you to keep your comments brief (250–300 words) and sharply focused. Health Affairs reserves the right to edit all letters for clarity and length and to publish them in the bound copy or on our Web site. Health Affairs will not acknowledge receipt of unsolicited letters that are not published. Letters can be submitted via e-mail, [email protected], or the Health Affairs Web site, www.healthaffairs.org.

restrictions that limit purchases to a thirtyday supply. These fundamental barriers are relevant not only to contraceptives and other lifestyle drugs but also to many other prescription drugs. Like insurance coverage, they too must be addressed.

Viagra And Contraceptives

Jennifer S. Haas Brigham and Women’s Hospital/ Harvard Medical School Cambridge, Massachusetts

As with Viagra (Rudolf Klein and Heidrun Sturm, Nov/Dec 02), contraceptives have been considered by some to be a “lifestyle” drug, and some insurers do not cover them. The debate over insurance coverage for contraceptives intensified when it became known that some insurers covered Viagra but not contraceptives. Some state-level attempts to mandate such coverage have been successful, but federal laws have not passed, despite repeated attempts. Some of the lessons learned from the Viagra experience can be applied to the debate over coverage for contraceptives, but there are also critical differences. As the authors note, defining “lifestyle” drugs is controversial when the drugs involved relate to sexual activity. However, the purpose of contraceptives is to prevent unplanned pregnancies—a goal that is generally embraced by our society. Thus, it is difficult to argue that contraceptives serve only personal goals similar to the promotion of sexual function, which is the purpose of Viagra. Further, as with Viagra, research suggests that insurance coverage of contraceptives is unlikely to have a large impact on health care budgets.1 Finally, Viagra is more expensive than many contraceptives. Despite the importance of insurance coverage in determining whether people have reasonable access to medications that improve their well-being, several other factors influence medication use, especially use of contraceptives.2 These factors include the growing numbers of uninsured people, increasing copayment amounts, and

Kathryn A. Phillips and Joanne Spetz University of California, San Francisco

NOTES 1.

J. Darroch, “Cost to Employer Health Plans of Covering Contraceptives” (New York: Alan Guttmacher Institute, 1998). 2. K.A. Phillips et al., “Out-of-Pocket Expenditures for Oral Contraceptives and Number of Pill Packs Obtained Per Purchase” (Paper presented at the annual meeting of AcademyHealth, Washington, D.C., 23 June 2002).

Air Pollution And Medical Care Victor Fuchs and Sarah Frank (Nov/Dec 02) suggest that differences in airborne concentrations of particulate matter influence the medical expenditures of older Americans. Taking the estimates at face value, what does their result teach us about the value of controlling air pollution? Conventional estimates of the value of reducing particulates are dominated by the reduction in mortality risk. For example, the Environmental Protection Agency (EPA) estimates that by 2010 the value of reduced adverse health effects attributable to the 1990 Clean Air Act Amendments (CAAA) will be $110 billion (in 1990 dollars). Of this total, $100 billion will be due to reductions in mortality risk, calculated using a value per statistical life (VSL) of $4.8 million. Some commentators claim that this estimate is too large, arguing that the value of decreasing mortality risk to

H E A L T H A F F A I R S ~ Vo l u m e 2 2 , N u m b e r 1

Downloaded from content.healthaffairs.org by Health Affairs on May 23, 2011 by guest

277

Letters

elderly people with cardiopulmonary impairment (who are at greatest risk from air pollution) is less than the value of reducing risk to younger and healthier workers (for whom the VSL is estimated). The authors’ estimates appear to buttress the EPA’s estimates of health care costs. Using estimates for 2010, the authors’ results imply that the CAAA will reduce annual expenditures for respiratory admissions by $140 million and for total inpatients by $580 million. These values are remarkably close to EPA estimates ($130 million and $520 million, respectively), which are based on independent estimates of the frequency and cost of hospital admissions. The EPA estimates should be the larger, as they include costs of lost productivity and effects on people of all ages. However, these differences are probably small, because the EPA estimates the value of lost productivity as much smaller than the medical expense, and most of the hospital admissions are presumably concentrated in this older age group. The authors’ results are consistent with conventional estimates of air pollution’s effects on medical costs. It is important to note, however, that medical costs pale in comparison with estimates of the value to individuals of reduced mortality risk. James K. Hammitt Harvard University Cambridge, Massachusetts

Rational Public Policy Fuchs and Frank make a valuable contribution. Unfortunately, they claim that “a rational public policy should try to strike a balance between…benefits and…costs,” thereby implying that Congress acted irrationally when it required the EPA to protect public health from air pollution. Yet the health-protective approach has proved to be enormously successful, and government efforts to balance costs and benefits have produced paralysis. In my forthcoming book, The Economic Dynamics of Environmental Law (MIT Press, 2003), I argue that we need much more faith in industry’s capacity to adapt to stringent requirements with

278

cost-saving innovation. This may explain why a statute with a cost-blind goal-setting provision has produced such enormous public health gains—gains that have far outstripped costs. Ironically, the details of the authors’ study show why pollution control statutes should never rely on cost-benefit analysis. Reasonably enough, because of significant data gaps, the authors make no estimate of medical care costs incurred to protect those over age eighty-five, a population likely to generate enormous medical costs. When the government tries to balance costs and benefits, large uncertainties tend to yield a precise but completely wrong number, namely zero. David M. Driesen Syracuse University Syracuse, New York The authors respond: James Hammitt notes that our cross-area regression results imply a reduction in annual expenditures of $580 million in inpatient utilization from pollution reduction. This is very close to the EPA estimate based on alternative analyses. He does not mention that our regression shows an effect on outpatient utilization that is 1.31 times as large as the effect on inpatient utilization. Thus, the implied total estimated reduction in annual expenditures should be $1.35 billion (2.31 × $580 million). Hammitt also notes that the EPA estimate of the value of mortality reduction of $4.8 billion has been criticized because it takes no account of the age distribution of the mortality attributed to pollution. We agree with this criticism. Rather than a value per statistical life, a more meaningful measure can be obtained by assigning a value to each year of life extended (adjusted for quality of life, if possible). As Hammitt notes, much of the excess mortality attributable to pollution occurs among the elderly. Many of the victims already suffer from various chronic diseases and disabilities. Thus, the value of mortality reduction based on quality-adjusted years of life extended would be considerably less than one based on a value per statistical life of $4.8 million, a figure derived from estimates of the

Ja n u a r y/ Fe b r u a r y 2 0 0 3

Downloaded from content.healthaffairs.org by Health Affairs on May 23, 2011 by guest

Letters

Victor Fuchs Stanford University Stanford, California

to the consumer, exacerbating problems for those unlucky enough to need care. U.S. health and health care disparities are increasing; since health status indicators parallel economic and educational levels, the United States also ranks low in the global health league. A black male born in our nation’s capital can expect to live 57.9 years—15.5 fewer years than the average U.S. male and slightly fewer years than the average male in Ghana, Bangladesh, or Bolivia. Even a white male from the Bronx has a shorter life expectancy than his brothers in Cuba or Sri Lanka. All modern health care systems, including ours, are predominantly publicly financed. In systems financed directly through taxation or social insurance, costs are visible to citizens. Our complex “private” health insurance system militates against transparency (public tax expenditure subsidies and their regressive effects can remain invisible to taxpayers), while it multiplies non–health care costs. The current system is not only collapsing; it is expensive and unfair. Health care is important for individual and societal well-being and productivity. Isn’t it time to replace our failing system with a better social contract?

Sarah Rosen Frank National Bureau of Economic Research Stanford, California

Barbara Kivimae Krimgold Center for the Advancement of Health Washington, D.C.

value of life at all ages. Once the effect on use of outpatient care is included and a more meaningful value of mortality reduction is calculated, Hammitt’s conclusion that “medical costs pale in comparison with estimates of the value…of reduced mortality risk” is unwarranted. For him to insert “to individuals” is beside the point. Even if all of the medical costs are borne by third parties, surely it is the social benefit of pollution reduction that is relevant to social policy. If David Driesen believes that it is very difficult to measure benefits and costs accurately, we agree. If he believes that it is possible to choose rationally among alternative policies without estimating their benefits and costs, we strongly disagree. There is an unlimited supply of policy interventions that offer potential benefits for health, education, public safety, and the like. Which interventions are worth the expenditure of resources, either public or private? Without estimates of benefits and costs, policy choices will be arbitrary and capricious; the public will be the loser.

Health Insurance System Promotes Disparities Humphrey Taylor’s refreshingly candid piece (Nov/Dec 02) moves quickly past the smoke and mirrors of “consumer choice” and takes us straight to the facts: Health risks are predictable and unevenly distributed across groups; poorer, sicker, and socially disadvantaged groups have more health care needs and fewer resources. Because of the demise of community rating and progressive crosssubsidies, private health insurance markets discourage use by the poor, increase access for the well-off, and leave increasing numbers of working Americans uninsured. Definedcontribution plans shift even more risks back

Consumer-Driven Health Plans Taylor argues that the nascent market for consumer-driven health care plans augurs the demise of the employer-sponsored health insurance system. His analysis is misguided because he appears to misunderstand the nature of these plans as they are being implemented. He assigns grave consequences to the notion of plan choice, but 65 percent of covered workers already had two or more plans from which to choose in 2000.1 This does not include the option that all workers can exercise: to decline coverage when it is offered, which many do with regularity. By mentioning the New York and New Jersey individual insurance markets as examples of adverse selection, the author

H E A L T H A F F A I R S ~ Vo l u m e 2 2 , N u m b e r 1

Downloaded from content.healthaffairs.org by Health Affairs on May 23, 2011 by guest

279

Letters

unwittingly demonstrates what can go wrong when choice is restricted to only a handful of costly albeit full-service plans: Healthier people leave the market and sicker people join. As part of his argument Taylor takes a curious stance against educated consumers of health care, claiming that they are “our worst nightmare.” In his world, consumer-driven health plans would give consumers greater flexibility in health insurance and health care decisions as well as Web-based tools that would allow them to predict the future. He believes that such plans would thus permit people to opt out when they learn that they will not be diagnosed with cancer or heart disease or get hit by a bus. Even the most fervent supporters of consumer-driven health plans do not anticipate these benefits. While the jury is still out on whether these plans will be successful, their greater use has the potential to lower health insurance premiums.2 Lower premiums would make people at the margin more likely to take up health insurance and firms on the margin more likely to offer health benefits, leading to a greater number of insured people. Arming consumers with knowledge of their illnesses, combined with a modicum of frugality that a consumer-driven plan provides, could be the first step toward injecting rationality into a health care system that desperately needs it. Anthony T. Lo Sasso Northwestern University Evanston, Illinois NOTES 1.

Henry J. Kaiser Family Foundation and Health Research and Educational Trust, “Employer Health Benefits, 2000 Annual Survey” (Menlo Park, Calif., and Washington: KFF/HRET, 2000). 2. J. Gabel, A.T. Lo Sasso, and T. Rice, “ConsumerDriven Health Plans: Are They More than Talk Now?” (www.healthaffairs.org/WebExclusives/ Gabel_Web_Excl_112002.htm (21 November 2002).

The author responds: Giving employees a broad range of real choices leads to adverse selection and cream skimming (not to mention moral hazard),

280

based on two pieces of evidence: First is the decision that millions of employees have already made not to pay their share of premiums required by their employers’ plans. The numbers of these “voluntarily uninsured” are growing, depriving the insured pool of large numbers of low-risk lives whose premiums are needed to subsidize the cost of sicker people. Further evidence comes from an experiment by an insurance company (whose identity must remain confidential) that offered a choice of two plans—one with a high premium/low copay and another with a low premium/high copay. Not surprisingly, the former attracted many more heavy users and the latter many more healthy people. These people made rational decisions and chose the plans that were best for them. The insurer told me that adverse selection would not be a problem because the company could set the rates so that profits from the low-risk group would cover losses from the high-risk group. However, this kind of cross-subsidization breaks down as soon as an employer buys a less expensive plan from another insurer to cover only the low-risk group. Sophisticated employers, willing to take the long-term view, may therefore offer only a choice of plans from one insurer to make this cross-subsidization possible. But how many employers are that sophisticated when competing insurers appear to offer them a better deal? If employers believe that choice is good, and having more real choices is better, the death spiral of adverse selection can only be prevented by regulation and government subsidies (in Australia it took both). Humphrey Taylor Harris Interactive New York, New York

Male And Female Nurse Dissatisfaction We have just completed a secondary analysis of Vermont registered nurse (RN) relicensure surveys and find a sharp contrast to Julie Sochalski’s (Sep/Oct 02) findings that males are uniformly less satisfied. In our (not

Ja n u a r y/ Fe b r u a r y 2 0 0 3

Downloaded from content.healthaffairs.org by Health Affairs on May 23, 2011 by guest

Letters

yet published) study, males were not significantly more likely than females were to be intending to leave their positions (23 percent versus 20 percent). Similarly, stepwise logistical regression didn’t support gender as a predictor of intention to leave. However, males who were intending to leave their current position were more likely than women to be doing so for reasons related to “job dissatisfaction”—75 percent versus 51 percent, respectively. Among dissatisfied nurses, males cited “dissatisfaction with salary” most often. Intention to leave for reasons of career advancement was virtually identical for males (6.3 percent) and females (6.2 percent). In a separate analysis of licensed practical nurses, males were more likely to be intending to leave their position for career advancement. Overall, Sochalski’s conclusions are supported by these findings. The role of gender socialization, however, isn’t easily elucidated in national or state databases. The extent to which males feel more comfortable voicing dissatisfaction related to wages is unclear. But it is reasonable to conclude that females in “helping” professions are more reluctant to voice such dissatisfaction. Voiced or not, dissatisfaction exists, as evidenced in our study by the marked contrast in the intention to leave for career advancement versus job dissatisfaction. The career ladders, flexible hours, better wages, and more satisfying workplace conditions advocated by Sochalski would be a positive first step.

refer specifically to minimum staffing ratios for ICUs (Sep/Oct 02), those mandated in California are of concern because they could create a staffing ceiling that oversimplifies the complexity of nursing care and fails to consider the vital interplay between patients’ needs and nurses’ competencies. The American Association of Critical-Care Nurses (AACN) Synergy Model for Patient Care is built on the premise that synergy results when patients’ needs and characteristics are matched with a nurse’s competencies. Eight universal patient characteristics—including resiliency, vulnerability, and stability—determine the competencies required of nurses in eight focus areas: clinical judgment and reasoning, advocacy and moral agency, caring practices, collaboration, systems thinking, response to diversity, clinical inquiry and innovation, and facilitation of learning. As put forth in the AACN Staffing Blueprint, the key to appropriately staffing a patient care area depends on nurses’ ability to meet patients’ needs. The solution to nursing staff allocation problems should include a comprehensive strategic plan—driven by nurses—that links patients’ needs, cost delivery, competency of providers, and staff mix with patient outcomes. Ratio-driven models developed by administrators, regulators, or payers are a glaring example of the underrecognition and underuse of nurses’ knowledge to which Debbie Ward and Bobbie Berkowitz refer (also Sep/Oct 02).

Betty Rambur, Mary Val Palumbo, Barbara McIntosh, and Joan Mongeon University of Vermont Burlington, Vermont

Wanda Johanson American Association of Critical-Care Nurses Aliso Viejo, California

Nurse Staffing Optimal nurse staffing is of serious concern to our 65,000 members. The highly vulnerable and unstable patients cared for by critical care nurses within and outside of intensive care units (ICUs) deserve our highest level of vigilance to assure their optimal outcome. Although Janet Coffman and colleagues did not

The authors respond: We wish to clarify an important element of California’s proposed nurse-to-patient ratios. In our paper we discuss minimum nurse-topatient ratios in ICUs, stating, “Under legislation enacted during the 1976–77 state legislative session, California hospitals must have a minimum ratio of one licensed nurse per two patients in intensive care and coronary care units. This legislation also requires that at least

H E A L T H A F F A I R S ~ Vo l u m e 2 2 , N u m b e r 1

Downloaded from content.healthaffairs.org by Health Affairs on May 23, 2011 by guest

281

Letters

half of licensed nurses working in intensive care and coronary care units be RNs.” The governor’s proposal calls for continuation of the nurse-to-patient ratios for ICUs and for implementation of new ratios for other types of inpatient units. Our discussion of the potential costs associated with the proposal focused on the new ratios for other inpatient units, because we sought to estimate the potential additional costs that the proposal might impose on hospitals. We could have stated more clearly that the governor’s proposal calls for the existing ICU ratios to remain in force. Anecdotal reports from staff nurses and nursing unions in California suggest that the ICU ratios promote greater job satisfaction among RNs working in ICUs, relative to those working in other hospital units. Many union officials believe that nurse staffing levels in ICUs are generally adequate; staff RNs who participated in a series of focus groups often cited the ICU ratios as an example of a policy that maintains RN workload at a reasonable level. In addition, hospitals treat the ICU ratios as minimum nurse-to-patient ratios and adjust staffing levels upward when ICU patients require more intensive nursing care. Janet M. Coffman for the authors University of California, Berkeley

Benefits Of Osteopathic Schools Edward Salsberg and Gaetano Forte (Sep/Oct 02) state that concerns since the 1980s about an impending physician surplus have led allopathic medical schools to restrain their output of new MDs. At the same time, they note that osteopathic medical graduates more than doubled between 1980 and 2000. They imply that osteopathic medical schools (and international medical graduates, or IMGs) undermined the response to a perceived oversupply problem. They also note that the physician surplus has not developed, and the continued restraint on allopathic output might require reconsideration. This situation again points out the wisdom and desirability of having two independent schools of physicians, allopathic and osteo-

282

pathic.1 Without an osteopathic medical discipline, current physician supply would certainly have been less adequate. Further, the growth in osteopathic output means that almost 12.7 percent of medical graduates in 2000 were osteopathic physicians, compared with 6.6 percent in 1980. An unintended benefit of the allopathic restraint is that the increased osteopathic market share should enhance competition between the two schools. Finally, the expansion of osteopathic graduates (and IMGs) may well lead allopathic schools to increase their output, which is a likely result of competition. Erwin A. Blackstone Temple University Philadelphia, Pennsylvania NOTE 1.

E.A. Blackstone, “The AMA and the Osteopaths: The Power of Organized Medicine,” Antitrust Bulletin 22, no. 2 (1977): 405, 440; and E.A. Blackstone, “Competition within the Physicians’ Services Industry: Osteopaths and Allopaths,” American Journal of Law and Medicine 8, no. 2 (1982): 137–150.

The authors respond: Although the increase in osteopathic medical schools over the past twenty years may help to address future physician workforce needs, we do not agree that ignoring national policy recommendations indicates wisdom or the desirability of having two approaches to medical education. It makes little sense to have some of the nation’s most respected medical schools limit enrollment growth in response to concerns over a potential oversupply of physicians, while another group of schools ignores policy recommendations and while incentives exist for hospitals to increase the inflow of IMGs. Edward Salsberg and Gaetano Forte University at Albany, State University of New York

Ja n u a r y/ Fe b r u a r y 2 0 0 3

Downloaded from content.healthaffairs.org by Health Affairs on May 23, 2011 by guest

Letters

Market Demand Matters The papers by Kevin Grumbach, Uwe Reinhardt, and Salsberg and Forte (Sep/Oct 02) all fail to consider several important demand-side facts in the U.S. health care environment. Grumbach discusses the analytic challenge of treating “the anticipated incidence of disease.” This misrepresents what in today’s health care marketplace constitutes a sizable proportion, if not a majority, of providers’ work output. Today’s medical paradigm is not the treatment of established disease, but disease prevention and the counseling of those already afflicted. Screening studies for cervical, colorectal, and dermatological malignancies as well as coronary occlusive disease constitute huge segments of demand for providers time. Often, these studies are marginally compensated. Counseling for sexually transmitted diseases, chronic hepatitis, coronary occlusive disease, hypertension, and an almost innumerable array of other maladies has drastically altered and added to providers’ duties and, hence, demand. The mere chronological computation of physician-to-population ratios provides almost no meaningful input in a market primarily characterized by an ever burgeoning demand. To portray such demand as providerdriven or inappropriate reflects, at best, misplaced hubris and, at worst, an ignorance of medicine’s history, which often reveals today’s medical fad to be tomorrow’s standard of care. Any attempt to impose Edmund Burke’s dictum that “society cannot exist unless a controlling power upon will and appetite be placed somewhere” ignores both political and market realities. Barry Kisloff University of Pittsburgh Pittsburgh, Pennsylvania The author responds: I am sympathetic with Barry Kisloff’s observation that many factors have “added to providers’ duties” but find fault with his logic that more physicians are therefore needed to

respond to this “ever burgeoning demand.” A variety of factors—biomedical innovations, evidence about effective standards of care, patients’ expectations—add to physicians’ potential workload.1 But to conclude that this added workload means that more physicians are required to meet the health needs of the public rests on three assumptions: All the work currently performed by physicians is clinically appropriate and improves patient outcomes; physicians are currently operating at maximal productivity; and none of the work performed by clinicians could be performed as effectively by less expensive personnel. None of these three assumptions is met in the U.S. health care system. Although evidence suggests that physicians should be doing more of some things (such as screening patients for colon cancer, prescribing beta-blockers after heart attacks, and counseling patients about smoking cessation), evidence also suggests that physicians should be doing less of others (such as annual physical exams for adults, total body screening scans, and aggressive hospital interventions for patients preferring palliative end-of-life care). Productivity is far less than optimal because of ill-designed practice systems that are inefficient. Part of this inefficiency results from the lack of a suitable division of labor among clinical team members. Although physicians should have a hand in the counseling activities mentioned by Kisloff, much of this counseling could be more efficiently—and often more effectively—provided by nurses, health educators, and other team members. Before jumping to the conclusion that more physicians per capita are required, physicians and policymakers need to critically assess the effectiveness of the work now being performed by physicians and think more creatively about reengineering practice organizations to enhance clinical performance and productivity. Kevin Grumbach University of California, San Francisco

H E A L T H A F F A I R S ~ Vo l u m e 2 2 , N u m b e r 1

Downloaded from content.healthaffairs.org by Health Affairs on May 23, 2011 by guest

283

Letters

NOTE

NOTES

1.

1.

K. Grumbach and T. Bodheimer, “A Primary Care Home for Americans: Putting the House in Order,” Journal of the American Medical Association 288, no. 7 (2002): 889–893.

Moonlighting Residents I commend Daniel Derksen (Sep/Oct 02) for a moving essay on the dilemma of balancing resident work hours with patient care, resident education, and stress. However, his suggestion to incorporate moonlighting into a residency warrants several caveats. First, residents need to know that when moonlighting as an attending, they will be legally held to the standard of an attending.1 Second, emphasis should be placed on determining which environments are appropriate for moonlighting. Working in the capacity of a physician assistant by taking routine medical admissions or seeing patients in a community clinic and then presenting to an attending would be appropriate. Covering a hospital and being responsible for cardiac arrests, when you’ve only done five intubations on a mannequin, would not. Finally, there is a misconception that emergency departments (EDs) are fine places for any resident to moonlight. The unpredictability, breadth, and often sheer volume of pathology seen in emergency medicine can challenge even the most experienced emergency physician. When 20 percent of moonlighting emergency medicine residents report feeling unprepared for the rigors they encounter, what quality of care might an ED patient receive from a moonlighting resident trained in a non–emergency medicine specialty?2 A landmark 1994 Josiah Macy report stated that “the presence of physicians in EDs who are neither adequately nor appropriately trained is not conducive to high quality emergency care.”3 Eight years later, this statement still holds. Resident moonlighting can be a means of alleviating stress and debt for residents, but both residents and patients must be protected from the clinical limitations of neophyte physicians. David T. Huang University of Pittsburgh Pittsburgh, Pennsylvania

284

L. Berlin, “Liability of the Moonlighting Resident,” American Journal of Roentgenology 171, no. 3 (1998): 565–567. 2. M.I. Langdorf et al., “National Survey of Emergency Medicine Resident Moonlighting,” Academic Emergency Medicine 2, no. 4 (1995): 308–314. 3. “The Role of Emergency Medicine in the Future of American Medical Care” (New York: Josiah Macy Jr. Foundation, April 1994).

The author responds: I agree that there are ways to reform residency training without eliminating important learning experiences or exposing residents or patients to unsafe work conditions or illprepared physicians. As David Huang observes, moonlighting residents are held to the standard of an attending and should be prepared for the moonlighting environment. By incorporating resident moonlighting into our institution, such assessment and oversight are possible. As director of the University of New Mexico Locum Tenens Program (primary care) and Specialty Extension Services, I spend several days a month in sites that request our practice relief services. Through direct experience, I assess which sites are appropriate for residents and recent graduates. In addition, when the site is billed, we survey each site about the performance of the physician and the program. Our medical director reviews charts completed by residents in these sites. Our programs assure that residents have state licenses and are privileged and credentialed through the requesting site. Approvals to moonlight are obtained through the graduate medical education (GME) office and the appropriate residency training program. The residency director and GME office receive a summary of moonlighting activity by resident. Such documentation will facilitate compliance with Accreditation Council for Graduate Medical Education (ACGME) resident dutyhour standards. Finally, as Huang points out, EDs can challenge even seasoned physicians. Last year when an emergency medicine resident provided coverage in a rural ED, we heard some complaints—not about his medical skills but

Ja n u a r y/ Fe b r u a r y 2 0 0 3

Downloaded from content.healthaffairs.org by Health Affairs on May 23, 2011 by guest

Letters

his interpersonal skills. The resident was concerned about the ED’s policies and procedures related to chest pain management and voiced these concerns in a direct manner. In the end, however, the rural ED instituted the resident’s suggestions, and the resident learned about more diplomatic ways to interact with staff and patients in a small town. Daniel Derksen University of New Mexico Albuquerque, New Mexico

Liposuction Procedures: How, Not Where The American Academy of Dermatology (AAD) believes that the issue in office-based medicine is how, not where, some doctors are performing certain surgical procedures (Elizabeth Lapetina and Elizabeth Armstrong, July/Aug 02). The AAD is concerned about the assertion that “aesthetic plastic surgeons…are the specialty most qualified to perform liposuction procedures.” Dermatology is the only specialty that includes liposuction training in its residency requirements. Tumescent liposuction, which uses local anesthesia administered with controlled volumes of fluid, was developed by a dermatologist to maximize patient safety. As the first medical specialty organization to develop guidelines of care for liposuction, the AAD recommends that liposuction should not be performed under general anesthesia, should be performed on healthy people, is not a procedure to treat obesity, should not be used to extract large amounts of fat, and should not be performed along with other procedures. Contrary to the authors’ assertions, data reviewed in a letter to the Journal of the American Medical Association showed that adverse incidents occur during office-based procedures when the above guidelines are not followed.1 We also remind the authors that the AAD’s recommendations for lidocaine usage are within the range of safely recommended dosages for anesthesia during this procedure. The Food and Drug Administration’s maximum dosage for regional anesthesia was sanctioned

in 1947, well before the development of tumescent liposuction, and has not been updated since. We share the researchers’ concern about patient safety and a lack of data regarding the safety records of office-based procedures. Mandatory reporting of adverse incidents would further ensure the safety of such procedures. However, adequate safety mechanisms need to be incorporated so that physicianpatient confidentiality is not violated. The AAD continues to support regulation that ensures the best care for patients but does not hamper the ability to provide it. Fred F. Castrow II American Academy of Dermatology Schaumburg, Illinois NOTE 1.

B. Coldiron, “Patient Injuries from Surgical Procedures Performed in Medical Offices,” Journal of the American Medical Association 285, no. 20 (2001): 2582.

The authors respond: The AAD’s efforts to develop appropriate care guidelines for liposuction are commendable. However, Fred Castrow’s letter highlights the importance not only of guidelines, but of more and better data on the incidence of adverse events in the outpatient setting. The research letter that he cites is, in fact, one of the few published reports on data collected by the state of Florida in the first twelve months after it mandated reporting in February 2000. However, the fifty-five reports, including six deaths, analyzed by that letter writer are insufficient to generate definitive conclusions about safety in the office setting, especially since they are contradicted by peer-reviewed data published elsewhere.1 Moreover, we are confused by Castrow’s invocation of patient-physician confidentiality as a potential barrier to reporting office-based adverse events. Until a wider spectrum of data on error in the outpatient setting becomes available, we reiterate our belief that patient safety is dependent upon both how and where procedures are performed.

H E A L T H A F F A I R S ~ Vo l u m e 2 2 , N u m b e r 1

Downloaded from content.healthaffairs.org by Health Affairs on May 23, 2011 by guest

285

Letters

Elizabeth M. Armstrong Princeton University Princeton, New Jersey Elizabeth M. Lapetina Lewin Group Falls Church, Virginia

condition. If we want to account for disability in priority setting, we should measure individual burden. Estimating BOD is a waste that comes at the opportunity costs of calculating more valuable measures.2 Afschin Gandjour University of Cologne Cologne, Germany

NOTE 1.

C.J. Coté et al., “Adverse Sedation Events in Pediatrics: A Critical Incident Analysis of Contributing Factors,” Pediatrics 104, no. 4 (2000): 805–814; and R.B. Rao, S.F. Ely, and R.S. Hoffman, “Deaths Related to Liposuction,” New England Journal of Medicine (13 May 1999): 1471–1475.

Burden-Of-Disease Estimates Benjamin Druss and colleagues (Jul/Aug 02) use their burden-of-disease (BOD) calculations for several conditions in the United States, to discuss to what degree national expenditures for these conditions are justified. Thus, they judge current allocation practice. They justify this use of BOD estimates by referring to John Rawls’s “Difference Principle,” which states that we should maximize the long-term expectations of the worst-off. I believe that there is no justification for using BOD estimates to guide resource allocation in health care. The authors are right to state that from a utilitarian or economic point of view, assessing BOD makes no sense. But a justification based on Rawls’s “Difference Principle” is also problematic: Rawls himself did not consider health when assessing who is worstoff. Instead, he relied on income and wealth. Kenneth Arrow pointed out that considering health would justify shifting resources to the worst-off until the rest of society is reduced to poverty.1 But even if we accepted that health matters and that the worst-off patients should be prioritized, BOD is not the right measure. It combines individual burden or disability with the prevalence rate of the disease. But is it justifiable to give priority to a patient simply because a large number of patients are affected by the same condition? Obviously not. Individual disability should matter, regardless of how many other people are affected by the same

286

NOTES 1.

K.J. Arrow, “Some Ordinalist-Utilitarian Notes on Rawls’s Theory of Justice,” Journal of Philosophy (May 1973): 245–263. 2. G. Mooney and V. Wiseman, “Burden of Disease and Priority Setting,” Health Economics ( July 2000): 369–372.

The author responds: Afschin Gandjour identifies several key questions: Why should those with the greatest BOD receive priority for health care resources? What is the appropriate way to measure this burden? And how should equity concerns be balanced against the goal of maximizing the cost-effectiveness of care? These important and complex issues continue to be explored by philosophers, economists, and bioethicists.1 It certainly seems premature to simply conclude that “estimating BOD is a waste.” People intuitively sense that particular concern needs to be accorded to the needs of the sickest individuals in setting priorities.2 Instead of dismissing this intuition as irrational or irrelevant, we should seek to better understand it and what it tells us about allocating limited resources. Benjamin Druss Yale University New Haven, Connecticut NOTES 1.

D. Brock, “Priority to the Worst Off in Health Care Resource Prioritization,” in Health Care and Social Justice, ed. M. Battin, R. Rhodes, and A. Silvers (New York: Oxford University Press, 2001); and J. Olsen, “Theories of Justice and Their Implications for Priority Setting in Health Care,” Journal of Health Economics 16, no. 6 (1997): 625–639. 2. E. Nord, Cost-Value Analysis in Health Care (Cambridge: Cambridge University Press, 1999).

Ja n u a r y/ Fe b r u a r y 2 0 0 3

Downloaded from content.healthaffairs.org by Health Affairs on May 23, 2011 by guest