Health Disparities — Less Talk, More Action
http://www.100md.com
《新英格兰医药杂志》
During the past decade, hundreds of articles have been published documenting the existence of racial and ethnic disparities in health and health care — a data deluge that has led many observers to suggest that it is time to stop documenting disparities and turn our efforts to doing something about them. Three articles in this issue of the Journal once again document the existence of disparities — albeit with a slightly different twist — by looking at time trends.1,2,3 For most of the areas studied, disparities between white patients and black patients have not substantially improved during the past decade or so. Rather than simply dismissing these findings as more documentation, how might we use them to enhance our knowledge and inform strategies to eliminate disparities?
Jha et al.1 examined trends in the use of important surgical procedures from 1992 to 2001 and found that the gap between whites and blacks has not narrowed in most regions of the United States. Vaccarino et al.2 examined trends in the quality of care for myocardial infarction from 1994 to 2002 and report persistent racial differences in the use of reperfusion therapy and coronary angiography and in hospital mortality but no differences in the number of prescriptions given for aspirin and beta-blockers. Such enduring trends suggest that the adage "The system is perfectly designed to get the results that it does" is once again applicable here. As Satcher et al.4 point out in documenting the lack of substantial improvement in the racial gap in mortality during the past 40 years in the United States, "this complex system is consistently producing a predictable result." Aspects of the system are so ingrained that many physicians remain unaware of these disparities. As recently as last year, only a third of cardiologists who responded to a survey agreed with the statement that "clinically similar patients receive different cardiovascular care based on what their race and ethnic background is."5 In short, the articles by Jha et al. and Vaccarino et al. strongly suggest that we will not make progress in eliminating disparities simply by tinkering with the system.
On a more encouraging note, Trivedi et al.3 report on the quality of care for whites and blacks in Medicare managed-care plans from 1997 to 2003 and show that overall clinical performance improved on all measures for both white and black enrollees. In addition, the racial gap actually narrowed for seven of the nine measures that were studied. However, for two of the three outcome measures — control of levels of glycosylated hemoglobin for patients with diabetes and of low-density lipoprotein cholesterol after myocardial infarction — the disparities actually increased slightly. It is likely that some change in systems led to the improvements. Indeed, since 1997, managed-care plans administered by Medicare have been required to measure and issue public reports on such quality measures. This reporting has been shown to be associated with an improved quality of care. Although the behavior of physicians is notoriously hard to change, ordering a test or writing a prescription is a relatively uncomplicated activity and is amenable to change with appropriate incentives and modifications to systems. The control of levels of glucose and lipids and the performance of a revascularization procedure after myocardial infarction are much more complex challenges, suggesting that we need to approach these persistent racial disparities as multifaceted system issues. The level of complexity suggests that no one party is to blame, even though it has been convenient to blame doctors, patients, or society at large. As we have learned from other areas of quality improvement, it is not that simple.
These three articles tell us about fundamental components of systems that will be critical in the elimination of disparities in care. First, the research could not have been done without data on the race or ethnic background of patients; two of these studies relied on data available to Medicare. Second, measurement and reporting, and their associated quality-improvement activities, have led to improvement in the quality of care overall as well as to the narrowing of racial disparities. However, until very recently, the bulk of the delivery system had no data on race and ethnic background, so it has been virtually impossible to examine, let alone publicly report, data on the quality of care for various racial and ethnic groups. The use of available data and standard quality-improvement methods constitutes a good start, although these approaches alone may not get us where we need to go. Additional work will be required to understand why the process measures have changed even though the outcomes have not. This problem will probably require analysis of the root causes of the persistence of these differences in control — a study that is not easily accomplished with administrative data.
An important article that was published last year may shed additional light on the systems issue. Bach et al.6 reported that a minority of physicians care for the overwhelming majority of black patients in the United States. Physicians who care for black patients were more likely to report that they had difficulty mustering adequate resources — subspecialty referrals, diagnostic imaging, high-quality ancillary services, or hospital admission — for their patients. The systems analysis will need to examine structural factors in the delivery system and look for modifiable root causes that extend beyond the immediate control of the physician's office or hospital. Systems analysis, as well as systems change, may need to involve the community as well as the delivery system.
There are good reasons to be optimistic about some aspects of systems change. The leadership of health insurance plans, including the commercial insurers and Medicaid plans listed in Table 1, now understand that in order to make further improvement in the quality of care and respond to a more demographically diverse marketplace, they need to make progress in racial disparities. They have also realized that they cannot make progress without being able to measure and monitor that progress, which means that they need information about the race and ethnic background of enrollees. Working in two collaborative efforts,7 they have started to obtain and use such data. In the case of Medicaid plans, these data are now available through most state Medicaid agencies. The commercial insurers have either begun to ask enrollees to supply information voluntarily on their race or ethnic background or they have used sophisticated geocoding and surname-analysis techniques to estimate the racial composition of their enrollees. But regardless of the methodology, all participating health plans have begun to move toward systematically examining the quality of care for important subgroups of their enrollees with chronic illnesses. Furthermore, recognizing that measurement alone will be insufficient to produce results, they are designing and testing a variety of interventions to address the disparities they have found. Although it is too soon to know how successful these efforts will be, the plans, which collectively cover 90 million people, are consciously changing their systems.
Table 1. Members of Groups Working on Racial Disparities in Health Care.
Insurers and health care providers working together may be able to make substantial progress in health care disparities, but they cannot solve this problem alone. In many areas of the country, employers are joining in the effort. However, more widespread redesign of systems — particularly, outside of the traditional health care system — will be required to address the complex interplay of social determinants of health and health care outcomes, and this change will probably be longer in coming. Meanwhile, those of us working within the health care system need to test and implement effective strategies for the reduction of disparities. We will continue to rely on trend data, such as those reported in this issue, either to document our progress or to point us in additional directions for solutions. Regardless, we cannot give up until the job is done.
Source Information
From RAND, Arlington, Va.
References
Jha AK, Fisher ES, Li Z, Orav EJ, Epstein AM. Racial trends in the use of major procedures among the elderly. N Engl J Med 2005;353:683-691.
Vaccarino V, Rathore SS, Wenger NK, et al. Sex and racial differences in the management of acute myocardial infarction, 1994 through 2002. N Engl J Med 2005;353:671-682.
Trivedi AN, Zaslavsky AM, Schneider EC, Ayanian JZ. Trends in the quality of care and racial disparities in Medicare managed care. N Engl J Med 2005;353:692-700.
Satcher D, Fryer GE, McCann J, Troutman A, Woolf SH, Rust G. What if we were equal? A comparison of the black-white mortality gap in 1960 and 2000. Health Aff (Millwood) 2005;24:459-464.
Lurie N, Fremont A, Jain A, et al. Racial and ethnic disparities in care: the perspectives of cardiologists. Circulation 2005;111:1264-1269.
Bach PB, Pham HH, Schrag D, Tate RC, Hargraves JL. Primary care physicians who treat blacks and whites. N Engl J Med 2004;351:603-605.
Major health plans and organizations join AHRQ to reduce racial and ethnic disparities in health care. Press release of the Agency for Healthcare Research and Quality, Rockville, Md., December 14, 2004. (Accessed July 28, 2005, at http://www.ahrq.gov/news/press/pr2004/dispcolpr.htm.)(Nicole Lurie, M.D., M.S.P)
Jha et al.1 examined trends in the use of important surgical procedures from 1992 to 2001 and found that the gap between whites and blacks has not narrowed in most regions of the United States. Vaccarino et al.2 examined trends in the quality of care for myocardial infarction from 1994 to 2002 and report persistent racial differences in the use of reperfusion therapy and coronary angiography and in hospital mortality but no differences in the number of prescriptions given for aspirin and beta-blockers. Such enduring trends suggest that the adage "The system is perfectly designed to get the results that it does" is once again applicable here. As Satcher et al.4 point out in documenting the lack of substantial improvement in the racial gap in mortality during the past 40 years in the United States, "this complex system is consistently producing a predictable result." Aspects of the system are so ingrained that many physicians remain unaware of these disparities. As recently as last year, only a third of cardiologists who responded to a survey agreed with the statement that "clinically similar patients receive different cardiovascular care based on what their race and ethnic background is."5 In short, the articles by Jha et al. and Vaccarino et al. strongly suggest that we will not make progress in eliminating disparities simply by tinkering with the system.
On a more encouraging note, Trivedi et al.3 report on the quality of care for whites and blacks in Medicare managed-care plans from 1997 to 2003 and show that overall clinical performance improved on all measures for both white and black enrollees. In addition, the racial gap actually narrowed for seven of the nine measures that were studied. However, for two of the three outcome measures — control of levels of glycosylated hemoglobin for patients with diabetes and of low-density lipoprotein cholesterol after myocardial infarction — the disparities actually increased slightly. It is likely that some change in systems led to the improvements. Indeed, since 1997, managed-care plans administered by Medicare have been required to measure and issue public reports on such quality measures. This reporting has been shown to be associated with an improved quality of care. Although the behavior of physicians is notoriously hard to change, ordering a test or writing a prescription is a relatively uncomplicated activity and is amenable to change with appropriate incentives and modifications to systems. The control of levels of glucose and lipids and the performance of a revascularization procedure after myocardial infarction are much more complex challenges, suggesting that we need to approach these persistent racial disparities as multifaceted system issues. The level of complexity suggests that no one party is to blame, even though it has been convenient to blame doctors, patients, or society at large. As we have learned from other areas of quality improvement, it is not that simple.
These three articles tell us about fundamental components of systems that will be critical in the elimination of disparities in care. First, the research could not have been done without data on the race or ethnic background of patients; two of these studies relied on data available to Medicare. Second, measurement and reporting, and their associated quality-improvement activities, have led to improvement in the quality of care overall as well as to the narrowing of racial disparities. However, until very recently, the bulk of the delivery system had no data on race and ethnic background, so it has been virtually impossible to examine, let alone publicly report, data on the quality of care for various racial and ethnic groups. The use of available data and standard quality-improvement methods constitutes a good start, although these approaches alone may not get us where we need to go. Additional work will be required to understand why the process measures have changed even though the outcomes have not. This problem will probably require analysis of the root causes of the persistence of these differences in control — a study that is not easily accomplished with administrative data.
An important article that was published last year may shed additional light on the systems issue. Bach et al.6 reported that a minority of physicians care for the overwhelming majority of black patients in the United States. Physicians who care for black patients were more likely to report that they had difficulty mustering adequate resources — subspecialty referrals, diagnostic imaging, high-quality ancillary services, or hospital admission — for their patients. The systems analysis will need to examine structural factors in the delivery system and look for modifiable root causes that extend beyond the immediate control of the physician's office or hospital. Systems analysis, as well as systems change, may need to involve the community as well as the delivery system.
There are good reasons to be optimistic about some aspects of systems change. The leadership of health insurance plans, including the commercial insurers and Medicaid plans listed in Table 1, now understand that in order to make further improvement in the quality of care and respond to a more demographically diverse marketplace, they need to make progress in racial disparities. They have also realized that they cannot make progress without being able to measure and monitor that progress, which means that they need information about the race and ethnic background of enrollees. Working in two collaborative efforts,7 they have started to obtain and use such data. In the case of Medicaid plans, these data are now available through most state Medicaid agencies. The commercial insurers have either begun to ask enrollees to supply information voluntarily on their race or ethnic background or they have used sophisticated geocoding and surname-analysis techniques to estimate the racial composition of their enrollees. But regardless of the methodology, all participating health plans have begun to move toward systematically examining the quality of care for important subgroups of their enrollees with chronic illnesses. Furthermore, recognizing that measurement alone will be insufficient to produce results, they are designing and testing a variety of interventions to address the disparities they have found. Although it is too soon to know how successful these efforts will be, the plans, which collectively cover 90 million people, are consciously changing their systems.
Table 1. Members of Groups Working on Racial Disparities in Health Care.
Insurers and health care providers working together may be able to make substantial progress in health care disparities, but they cannot solve this problem alone. In many areas of the country, employers are joining in the effort. However, more widespread redesign of systems — particularly, outside of the traditional health care system — will be required to address the complex interplay of social determinants of health and health care outcomes, and this change will probably be longer in coming. Meanwhile, those of us working within the health care system need to test and implement effective strategies for the reduction of disparities. We will continue to rely on trend data, such as those reported in this issue, either to document our progress or to point us in additional directions for solutions. Regardless, we cannot give up until the job is done.
Source Information
From RAND, Arlington, Va.
References
Jha AK, Fisher ES, Li Z, Orav EJ, Epstein AM. Racial trends in the use of major procedures among the elderly. N Engl J Med 2005;353:683-691.
Vaccarino V, Rathore SS, Wenger NK, et al. Sex and racial differences in the management of acute myocardial infarction, 1994 through 2002. N Engl J Med 2005;353:671-682.
Trivedi AN, Zaslavsky AM, Schneider EC, Ayanian JZ. Trends in the quality of care and racial disparities in Medicare managed care. N Engl J Med 2005;353:692-700.
Satcher D, Fryer GE, McCann J, Troutman A, Woolf SH, Rust G. What if we were equal? A comparison of the black-white mortality gap in 1960 and 2000. Health Aff (Millwood) 2005;24:459-464.
Lurie N, Fremont A, Jain A, et al. Racial and ethnic disparities in care: the perspectives of cardiologists. Circulation 2005;111:1264-1269.
Bach PB, Pham HH, Schrag D, Tate RC, Hargraves JL. Primary care physicians who treat blacks and whites. N Engl J Med 2004;351:603-605.
Major health plans and organizations join AHRQ to reduce racial and ethnic disparities in health care. Press release of the Agency for Healthcare Research and Quality, Rockville, Md., December 14, 2004. (Accessed July 28, 2005, at http://www.ahrq.gov/news/press/pr2004/dispcolpr.htm.)(Nicole Lurie, M.D., M.S.P)