Health Affairs, 22, no. 2 (2003): 60-71 doi: 10.1377/hlthaff.22.2.60 © 2003 by Project HOPE	New Online   Pay Cuts For Medicare Docs   Access To Care Woes   Public Coverage More Efficient   Empowering Consumers

This Article Abstract Reprint (PDF) Submit a response to this article Alert me when this article is cited Alert me when eLetters 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 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 ISI Web of Science (13) Citing Articles via Google Scholar Google Scholar Articles by Mehrotra, A. Articles by Dudley, R. A. Search for Related Content PubMed PubMed Citation Articles by Mehrotra, A. Articles by Dudley, R. A. Related Collections Business Of Health Hospitals Insurance - Business Coalitions Quality Of Care

Employers & Quality

Employers’ Efforts To Measure And Improve Hospital Quality: Determinants Of Success

Abstract

 
We examined eleven communities in which an employer coalition created a report card to compare the performance of the community’s hospitals. After interviewing employer coalition and hospital representatives from each community, we found great variability in report cards’ capacity to prompt quality improvement. Although some were successful, others had less impact because of conflicts between employer coalitions and hospitals. Areas of disagreement included selection of appropriate goals, methodology of quality measurement, whether report cards should be publicly released, and the use of economic incentives to improve quality. We describe these conflicts and offer recommendations for future hospital report cards.

Despite the effort that has gone into producing report cards, previous researchers have found they have variable success in improving quality.¹ Report card data are usually several years old and therefore often perceived as irrelevant.² Ratings may be unable to discriminate accurately between two hospitals’ performance.³ Consumers may have a difficult time interpreting the data, and physicians only rarely incorporate report card information into patient referral decisions.⁴

A number of researchers have studied report cards sponsored by Medicare, state governments, and the private company HealthGrades, but few have investigated a common report card sponsor: regional employer coalitions.⁵ Employers are the second-largest (after government) purchasers of health care in the United States and could play an important role in driving quality improvement.⁶ About 100 U.S. employer coalitions exist, and many have produced report cards, yet only three evaluations of hospital report cards—all limited to one report card—have been performed.⁷ Because employer coalitions are important but poorly understood actors, we studied their efforts to improve quality through hospital report cards and assessed the determinants of these initiatives’ success.

Study Methods

Top Study Methods Study Findings Conclusions And Policy... NOTES

 
We used an interview-based qualitative research design, a useful methodology for complex, rapidly changing phenomena and for exploratory studies in areas in which previous literature is limited.⁸ In each community we interviewed the leader of the employer coalition creating the report card and representatives of hospitals being evaluated. Hospital representatives, rather than consumers or individual physicians, were interviewed because the available evidence, although limited, suggests that they are the most responsive to report cards.⁹

Using a database maintained by the National Business Coalition on Health and interviews with experts in the field, we identified all employer-led hospital report cards produced through December 2001 that included more than three hospitals and the majority of hospitals in the community. We included both publicly released and private report cards as well as report cards that were no longer being produced. In communities having two report cards, we included both in our interviews and analysis. Report cards produced by the Leapfrog Group and the Hospital Profiling Project were offered in more than one community. We did not want perceptions of these two projects to dominate our analyses. Therefore, of the six communities in which one of these was the only report card, we included one community for Leapfrog and one for the Hospital Profiling Project, thereby excluding four areas (Atlanta, St. Louis, Michigan, and California). With those exceptions, our study population included all communities with employer-driven hospital report cards that had been produced through December 2001 (Exhibit 1).

Study Findings

Top Study Methods Study Findings Conclusions And Policy... NOTES

 
Overview of employer coalition hospital report cards. The eleven communities studied ranged from metropolitan areas such as Seattle and Cleveland to geographically larger, more rural areas such as northern Alabama and Maine. Most communities had more than one employer-led report card. Although we use "employer coalition" to refer to the organizers of the report card, in some cases the coalition included both employers and other groups (unions, health plans, or hospitals). Some coalitions are dominated by large employers, such as Disney or Boeing; others consist of numerous smaller employers.

The most commonly used performance measures were in-hospital mortality and length-of-stay. Sometimes these measures were reported for a particular diagnosis (such as pneumonia or heart failure); other times they were divided into specialty categories (such as cardiology or surgical oncology); on a few occasions the data were reported as one number for all hospitalized patients. Less frequently chosen measures were the results of patient satisfaction surveys, complication rates, accreditation scores, process measures (for example, percentage of myocar-dial infarction patients receiving aspirin), and the Leapfrog Group measures.¹¹ The reports were usually generated from billing data, but chart abstraction and hospital surveys were also used. Although we use the term report card, organizers often used other titles, such as "Patient Safety Report" or "Consumer Guide." Seven coalitions released the report card to the public. Only three report cards were initiated before 1998.

Report card success. Each interviewee was asked about hospitals’ response to the report card. We considered a report card successful if hospital or employer representatives gave examples of how the report card prompted or increased interest in quality improvement by hospitals. Our interviews suggest that a number of employer coalition report cards have been unsuccessful. Yet in other communities the report card catalyzed quality improvement, with hospitals reportedly devoting sizable resources to improving their performance with such projects as new computerized physician order entry (CPOE) systems. In communities with more successful report cards, interviewees also described a generally increased attention to quality: "The report card put quality assurance directors at the board of trustees meetings as frequently as the finance officers."

Barriers to success. Our analysis of the interviews found that there were six key issues limiting success: ambiguity of report card goals, conflicts over methods of measuring quality, the benefits of public release, and the use of economic incentives; lack of employer market power; and failure to do collaborative planning. Here we explore these six barriers in greater depth.

Ambiguity of goals. In many communities hospitals were unsure why mployers wanted a report card. For example, some coalitions claimed that they wanted to use report card projects to improve relationships with employees and unions but had no plan for how employees would actually use the data. Other coalitions initially informed hospitals that report cards were created to allow employers to select high-quality providers or plans, but then employers failed to use the data in contracting. One interviewee explained, "What [employers] got from the project was complex information they couldn’t interpret."

Hospital executives observed these discrepancies between stated intentions and initial plans and felt unsure about employers’ goals. Those executives who felt that the data were not being used were particularly critical about the data collection costs associated with report cards.

The largest contributor to ambiguity about goals, however, was hospitals’ perceptions about employers’ commitment to cost versus quality. Some coalitions were frank that their motivation was cost: "Quality is our last hope for reducing our health care expenditures. We have to do something about quality or just stop offering health care benefits." Other coalitions stated that improving quality itself was the primary goal. However, since most employer coalitions were originally formed to give employers greater market power to pursue price discounts, hospital representatives often suspected that quality initiatives were really about cost. One reported, "An organization that has been a negotiator of price first and foremost, that now declares that it’s focused on quality initiatives, is going to be a hard sell in this community," while another said, "Ultimately, you’re going to contract with me or not on the basis of price. End of story." This distrust of employers’ intentions was widespread.

Conflicts over how to measure quality. A recurring theme across the interviews was disagreement between the employer coalition and providers about report card methodology in three areas: case-mix adjustment, using outcomes versus process measures for quality improvement, and the validity and cost of the data collected for the report card. Although the extent of conflict on these topics varied, severe differences of opinion in some communities translated into a lack of faith in the report card and a refusal by hospitals to use the data.

Case-mix adjustment (also known as risk or severity adjustment) is the process of adjusting measurements of quality to account for variation in patient populations among hospitals. Without it, one cannot know whether a hospital with higher mortality provides worse care, has sicker patients, or both. In the ten communities using case-mix adjustment, most hospitals were concerned that the adjustment was inadequate. One coalition representative described "innumerable concerns raised about the inadequacy of the risk adjustment process for certain types of patients." Some hospital leaders expressed the view that "not in the near future, nor possibly ever, will we develop a reliable severity adjustment system." In contrast, most employers (and a few hospital executives) felt that some case-mix adjustment is better than none. One coalition leader said, "We have to move on with what we have today. When you find perfect, come back, and we’ll change immediately."

Some hospital representatives reported difficulty using report cards because of the type of data produced. For example, some report cards listed mortality for all cardiology patients rather than from a specific diagnosis such as myocardial infarction. Hospitals felt such data were too aggregated to be useful. Moreover, the outcome measures used in most report cards did not furnish clues as to the underlying quality problem: "Process measures are much more useable to hospitals in order to improve. If you have a [high] death rate, you still don’t know why the death rate is high." Many employers did not accept this complaint as legitimate: "People want longer-lasting batteries. Well, Duracell doesn’t stand there...and say, ‘Tell us how to make longer-lasting batteries.’ That’s the job of Duracell."

Often the most contentious issue between employer coalitions and hospitals involved the data used to generate report cards. Many coalitions use billing data, primarily from Medicare’s MedPar database or from state governments, to avoid having to ask hospitals for data. The use of billing data produced an almost universally negative reaction among hospital representatives, who viewed the data as financial, not quality, information: "We only put those things on the claim form that would have an impact on payment." Employer representatives were mixed in their response to this criticism, with some offering the counterargument, "It may not be accurate, but you got paid on it. That makes it pretty accurate in the real world."

The most common alternative to using billing data was chart review, but the majority of hospital representatives objected to its cost: "We have no money to hire people to go in and abstract charts." The phrase "unfunded mandate" was used by hospital leaders, who expressed frustration that they had to pay to gather data for report cards they did not want. Yet some hospital and employer representatives did not accept this complaint: "When you look at what the automotive industry puts into quality for vehicles, what we spend is extremely low." An employer coalition interviewee stated, "It’s pretty insulting if you’re saying that less than 2/1000 of a percent of a [cost of hospitalization] is too much. Maybe we shouldn’t be your customer because philosophically we’re pretty far apart." Disagreements about how to measure quality were common in communities in which report cards had been abandoned or ignored.

Conflicts about the utility of public release. Public release of a report card could improve quality by attracting the attention of consumers and the press, thereby forcing hospitals to respond. Among the seven communities with publicly released report cards, public release generated mixed responses. In the majority of communities the initial public release led to antagonism between hospitals and the employer coalition, sometimes derailing the report card. Hospitals were unhappy that a report card they perceived as inaccurate was shared with consumers. Media attention was described as uneven, with reporters focusing on controversies such as a well-regarded hospital performing poorly or certain hospitals’ dissatisfaction with the report card methodology.

Some hospitals were angered by what they saw as strong-arm tactics by employers: "Using this kind of information is more viable when it’s done in a less threatening model. You don’t use the press to beat people over the head." This anger was related to a common fear among interviewees that poor hospital performance would scare consumers, leading to "the public, uneducated about quality, stampeding the doors wanting change or getting our politicians involved."

Interviewees in communities with public report cards were generally less fearful of public release than were those in communities where data were kept confidential. Although the first release often generated a flurry of intense media interest, coverage waned over time. Hospitals also reported that consumers’ interest was relatively low. While this reduced the fear of stampeding consumers, it also led some hospitals to question whether it was worth collecting data at all. Other hospital and coalition leaders felt that the public’s interest would increase over time and stimulate change. One interviewee stated: "It’ll take three or four years of several guides before the consumer gets astute and starts paying attention." Another argued that a current low level of public interest is sufficient for prompting quality improvement: "Food labeling is the right metaphor. You want [to get] to one-and-a-half to three percent of the people each year. That gives hospitals time to fix their problems without horrible penalties...But if they ignore it for five years, all of a sudden you’re looking at a more significant market share shift."

Use of economic incentives. Some policymakers have proposed that paying more for good performance will lead to more rapid quality improvement.¹² None of the employer coalitions in our study based financial incentives on report card performance, and some interviewees cited this as an important deficiency that limited hospitals’ willingness to bear data collection costs. Those less supportive of economic incentives felt that health care providers are inherently interested in quality and do not need incentives.

Hospital leaders from communities with several years’ experience with report cards were more likely to support economic incentives: "In the beginning, recognition is good, but as you move forward, there has to be more. It’s about the money." One interviewee summarized, "It isn’t that people need rewards to do good work; they need rewards to help modify behavior and offset the costs they incur by devoting time to change."

Opinions also depended on the type of incentive considered. Hospital interviewees generally preferred increased reimbursement for high-quality providers to steerage (also called selective referral or evidence-based referral), whereby employers channel their employees to the better hospitals. Hospitals lose money on patients with certain diagnoses and are fearful that having more patients would increase losses. Further, some hospitals are full. Some also questioned the feasibility of requiring employees to go to higher-quality providers. One employer coalition leader said, "Employers are only in a position to threaten steerage. They’re not in a position to do it. Employers don’t have time to explain to every employee who wants to send their grandma to the nearest hospital why they can’t."

Variable degrees of employer market power. In theory, employer coalitions can use their buying power to force hospitals’ participation in report card programs.¹³ Our interviews suggest that the balance of power between the community’s employers, health plans, and hospitals is variable and that the relative power of employers vis-à-vis hospitals influenced the success of report card projects. In several communities a report card initiative was terminated or nearly terminated when hospitals withdrew from the program and employers did not have the market share to compel them to participate. In a few communities, though, employers had strong purchasing power and used it effectively: "We have what we call a velvet hammer. We’ve never had to do it, but we’ve had to go to the brink of threatening to move business out." This level of clout was the exception, however: "We’re not a huge purchasing entity that can force the hospitals to do anything."

Employers’ influence depends on the structure of the hospital and health plan market. In some communities hospitals have consolidated and gained the upper hand. One interviewee reflected that "market power does not remain constant. Sometimes purchasers are in the ascendancy and at other times, providers...Consolidation in the hospital market has...changed the balance of power here." Health plans may also eclipse employer clout. In theory, employers can dictate to a plan which providers should be included in its contracts. In reality, one or two health plans may dominate an area and become more important to hospitals than employers are. Moreover, health plans are under pressure to include all providers in their offerings regardless of quality: "For them to market their services in the southwest corner of our area, they needed a hospital in the southwest corner. The pricing they got from that hospital was more important than the quality findings."

Lack of collaborative planning. A recurring theme in our interviews was that of collaboration—whether hospitals believed that their voices were heard in the initial design and subsequent modifications of the report card: "We were part of this and we may not like our report this month, but we had a choice as to how the game was played." In several communities hospitals were frustrated because they did not understand how they were graded: "It’s like a big sausage maker. You take all this information, you feed it in the front, you get an answer out the back, and [you] have no idea what’s in the middle." Some coalition leaders were cognizant of the need for collaboration but were frustrated because hospitals appeared more interested in derailing the project than in commenting constructively. One hospital representative admitted that "most people like me are pretty resistant to this kind of thing. We’re not just jumping in there with both feet, saying, ‘Oh, let me share that piece of information.’"

Conclusions And Policy Implications

Top Study Methods Study Findings Conclusions And Policy... NOTES

 
As major purchasers of health care, employers have the potential to stimulate quality improvement. One way employers have tried to tap that potential is by creating hospital report cards. Although this trend is still in its infancy, our analysis found that only a few existing report cards have stimulated quality improvement. However, those that were successful demonstrate their potential impact. In several communities report cards appear to have changed the behavior and priorities of top management, and hospitals had come to accept performance reporting as meaningful and valuable.

There is no clear set of report card characteristics that predicted success, except that almost all successful report cards used data other than billing data. We hypothesize two possible explanations for this relationship: Hospitals ignored report cards based on billing data as inaccurate, or hospitals were never made partners in such initiatives. Ironically, employers sometimes used billing data precisely because they could get it from government databases without hospitals’ involvement. However, this may have increased the extent to which hospitals felt they had no stake in the process and were not collaborators.

Bringing employers and hospitals together. To be effective, employer coalitions must work collaboratively with hospitals on report card design. We found that despite their role as major purchasers, employers in most communities did not have sufficient market power to dictate report card methodology or enforce hospital participation. One strategy to achieve consensus and increase hospital involvement would be to choose quality measures that providers recognize as valid, such as the Joint Commission on Accreditation of Healthcare Organizations (JCAHO) core measures or provider-led quality measurement such as participation in the Society of Thoracic Surgeons database.

As of July 2002, JCAHO core measures are required for hospitals to receive accreditation; they focus on processes of care in four areas: heart failure, myocardial infarction, pneumonia, and obstetric care. The Society of Thoracic Surgeons database is a voluntary system that allows surgeons and hospitals to compare their performance with that of their peers in coronary artery bypass graft (CABG) and valve replacement surgery. There are similar voluntary reporting systems in other clinical areas. The National Quality Forum’s effort to establish a standard set of hospital quality measures also may create more consensus between employers and providers on what is a valid quality measure. The barriers we identified to use of these sources of quality information were that they require the voluntary submission of data from hospitals and that most employer leaders were unaware of their existence.

Public release—no panacea. Contrary to expectations, our analysis suggests that public release had limited impact on hospitals’ interest in quality improvement. Hospitals’ fears of consumer uproar and media frenzy appear exaggerated. When there was media interest, it was usually short-lived. Yet this lack of sustained response means that publicity has limited effectiveness as an incentive over time. In some cases, premature public release led to hospital backlash. In this regard, we agree with one interviewee’s advice: "I would insist that the process is validated before you start releasing data to the public."

Potential of economic rewards. Even when hospitals are responsive to performing poorly on the report card, our interviews suggest that interest will wane if hospitals are unable to recoup their investment in quality improvement. The creation of economic rewards for higher-quality providers may be an important watershed, serving as evidence that employers are serious about quality and not simply interested in cost. An unexpected finding was that many hospitals did not perceive increased patient volume as an economic reward. This was especially true of hospitals that were already full or were skeptical about the payment rates associated with selectively referred patients. Increased reimbursement may be a more effective economic reward; however, report card data must be accurate before becoming a yardstick that determines reimbursement.

Study limitations. The findings of this project are limited by the possibility that interviewees were unwilling to share their true feelings and that we did not directly measure the extent to which report cards stimulated change. Also, some report cards were released within the past year, and it may be premature to judge their impact. However, prior research supports several of our findings. Martin Marshall and colleagues conclude that there is "minimal agreement among the various stakeholders about expected gains from the release of comparative performance data."¹⁴ Report cards created with billing data, such as the Medicare report cards, which measured all patient hospital mortality, and the California Hospital Outcomes Project, which measured mortality after myocardial infarction, were also affected by hospitals’ anger over the methodology.¹⁵

The next step. With the continued growth in interest in health care quality, report cards will likely become more commonplace. However, they are only the first step in quality improvement. Providers need to transform report card data into effective internal quality improvement. Although it did not fit the entry criteria for this project, the Guidelines Applied in Practice (GAP) initiative illustrates the potential benefits of the next step. The GAP study was sponsored by the American College of Cardiology, the Michigan Peer Review Organization (PRO), and the local Detroit employer coalition.¹⁶ In the GAP study, hospitals received data about their myocardial infarction care and a toolkit for improving performance: clinical pathways, standard orders, use of opinion leaders, reminder systems, and feedback loops. This resource-intensive intervention resulted in modest but clinically and statistically significant improvement in quality. Projects such as GAP demonstrate that developing report cards and translating that information into quality improvement, although difficult, is possible. The production of report cards is a necessary but not a sufficient step toward improving the quality of hospital care.

Editor's Notes

 
Ateev Mehrotra is a third-year resident in the Harvard Combined Medicine/Pediatrics Program. He conducted this research last year as a fellow at the Institute for Health Policy Studies, University of California, San Francisco (UCSF). Tom Bodenheimer is a clinical professor in the UCSF Department of Family and Community Medicine. Adams Dudley is an assistant professor of medicine and health policy at UCSF.

This project was supported by the Robert Wood Johnson Foundation’s Improving Chronic Illness Care Program. A summary of our project’s findings was presented at the Annual Research Meeting of AcademyHealth, Washington, D.C., 23–25 June 2002.

NOTES

Top Study Methods Study Findings Conclusions And Policy... NOTES

 

1. M.N. Marshall et al., "The Public Release of Performance Data: What Do We Expect to Gain? A Review of the Evidence," Journal of the American Medical Association 283, no. 14 (2000): 1866–1874.[Abstract/Free Full Text]
2. D.M. Eddy, "Performance Measurement: Problems and Solutions," Health Affairs (July/Aug 1998): 7–25.
3. H.M. Krumholz et al., "Evaluation of a Consumer-Oriented Internet Health Care Report Card: The Risk of Quality Ratings Based on Mortality Data," Journal of the American Medical Association 287, no. 10 (2002): 1277–1287.[Abstract/Free Full Text]
4. J.H. Hibbard et al., "Increasing the Impact of Health Plan Report Cards by Addressing Consumers’ Concerns," Health Affairs (Sep/Oct 2000): 138–143; Marshall et al., "The Public Release of Performance Data"; and E.C. Schneider and A.M. Epstein, "Use of Public Performance Reports: A Survey of Patients Undergoing Cardiac Surgery," Journal of the American Medical Association 279, no. 20 (1998): 1638–1642.[Abstract/Free Full Text]
5. D.M. Berwick and D.L. Wald, "Hospital Leaders’ Opinions of the HCFA Mortality Data," Journal of the American Medical Association 263, no. 2 (1990): 247–249; [Abstract]P.S. Romano, J.A. Rainwater, and D. Antonius, "Grading the Graders: How Hospitals in California and New York Perceive and Interpret Their Report Cards," Medical Care 37, no. 3 (1999): 295–305; and [Medline]Krumholz et al., "Evaluation of a Consumer-Oriented Internet Health Care Report Card."
6. H. Darling, "Containing Costs: Corporate Employers’ Perspective," 23 January 2002, www.healthaffairs.org/WebExclusives/Darling_Perspective_Web_Excl_012302.htm (12 December 2002); and T. Bodenheimer and K. Sullivan, "How Large Employers Are Shaping the Health Care Marketplace," First of Two Parts, New England Journal of Medicine 338, no. 14 (1998): 1003–1007.[Free Full Text]
7. I. Fraser et al., "The Pursuit of Quality by Business Coalitions: A National Survey," Health Affairs (Nov/Dec 1999): 158–165; G.E. Rosenthal et al., "Using Hospital Performance Data in Quality Improvement: The Cleveland Health Quality Choice Experience," Joint Commission Journal on Quality Improvement 24, no. 7 (1998): 347–360; G.E. Rosenthal, L. Quinn, and D.L. Harper, "Declines in Hospital Mortality Associated with a Regional Initiative to Measure Hospital Performance," American Journal of Medical Quality 12, no. 2 (1997): 103–112; and J.D. Clough et al., "Lack of Relationship between the Cleveland Health Quality Choice Project and Decreased Inpatient Mortality in Cleveland," American Journal of Medical Quality 17, no. 2 (2002): 47–55.[Abstract]
8. C. Pope and N. Mays, "Reaching the Parts Other Methods Cannot Reach: An Introduction to Qualitative Methods in Health and Health Services Research," British Medical Journal 311, no. 6996 (1995): 42–45.[Free Full Text]
9. Marshall et al., "The Public Release of Performance Data."
10. M.K. Giacomini and D.J. Cook, "Users’ Guides to the Medical Literature: XXIII. Qualitative Research in Health Care, A. Are the Results of the Study Valid? Evidence-Based Medicine Working Group," Journal of the American Medical Association 284, no. 3 (2000): 357–362.[Abstract/Free Full Text]
11. As part of its accreditation process, the Joint Commission on Accreditation of Healthcare Organizations (JCAHO) evaluates hospitals in depth every three years. After the evaluation, hospitals receive a score based on their performance in meeting JCAHO standards. The Leapfrog Group evaluates hospitals based on three criteria: (1) Does the hospital have a CPOE system? (2) Is the hospital’s intensive care unit (ICU) staffed by ICU specialists? (3) What is the hospital’s yearly volume of cases in five different areas, and how does that volume compare with what is considered a high-volume hospital? The emphasis on volume arises from prior research demonstrating that hospitals with a high volume for a specific procedure have better outcomes for that procedure.
12. M.J. Coye, "No Toyotas in Health Care: Why Medical Care Has Not Evolved to Meet Patients’ Needs," Health Affairs (Nov/Dec 2001): 44–56.
13. R.H. Miller and R.R. Bovbjerg, "Efforts to Improve Patient Safety in Large, Capitated Medical Groups: Description and Conceptual Model," Journal of Health Politics, Policy and Law 27, no. 3 (2002): 401–440.[Abstract]
14. Marshall et al., "The Public Release of Performance Data."
15. Romano et al., "Grading the Graders"; and Berwick and Wald, "Hospital Leaders’ Opinions."
16. R.H. Mehta et al., "Improving Quality of Care for Acute Myocardial Infarction: The Guidelines Applied in Practice (GAP) Initiative," Journal of the American Medical Association 287, no. 10 (2002): 1269–1276.[Abstract/Free Full Text]

                      What's this?

This article has been cited by other articles:

				C. P. Wen, S. P. Tsai, and W.-S. I. Chung A 10-Year Experience with Universal Health Insurance in Taiwan: Measuring Changes in Health and Health Disparity Ann Intern Med, February 19, 2008; 148(4): 258 - 267. [Abstract] [Full Text] [PDF]

				M. B. Rosenthal and R. A. Dudley Pay-for-Performance: Will the Latest Payment Trend Improve Care? JAMA, February 21, 2007; 297(7): 740 - 744. [Full Text] [PDF]

				H. H. Pham, J. Coughlan, and A. S. O'Malley The Impact Of Quality-Reporting Programs On Hospital Operations Health Aff., September 1, 2006; 25(5): 1412 - 1422. [Abstract] [Full Text] [PDF]

				N. J.O. Birkmeyer and J. D. Birkmeyer Strategies for improving surgical quality--should payers reward excellence or effort? N. Engl. J. Med., February 23, 2006; 354(8): 864 - 870. [Full Text] [PDF]

				E. L. Arsura, M. Lewis, A. K. Jha, J. B. Perlin, and R. A. Dudley The Quality of Care in the Veterans Affairs Health Care System N. Engl. J. Med., September 11, 2003; 349(11): 1093 - 1094. [Full Text] [PDF]