Improving the Quality of Hospital Care in America
http://www.100md.com
《新英格兰医药杂志》
Over the past few years, several large studies have shown,1,2,3,4 and the Institute of Medicine has emphasized,5 that the quality of health care in the United States is not nearly at the level that we should expect from the world's most expensive health care system. Problems with quality are pervasive throughout both outpatient and inpatient settings and may be responsible for thousands of deaths each year.3,4
In this issue of the Journal, Williams and colleagues,6 from the Joint Commission on Accreditation of Healthcare Organizations (JCAHO), present time-series data from their "core measures" program, in which hospitals seeking accreditation must collect and submit data on clinical performance according to standardized, evidence-based measure sets. They bring us good news. During the period from late 2002 through early 2004, continuously participating hospitals improved significantly on 15 of the 18 measures for acute myocardial infarction, heart failure, and pneumonia. These hospitals improved significantly on every measure that involved appropriate use of medications or provision of counseling, and the trend was favorable for all three of the measures that did not show statistically significant improvement (i.e., mean time to thrombolysis and inpatient mortality after acute myocardial infarction and blood cultures for pneumonia). Furthermore, hospitals with the poorest performance at baseline improved the most over this two-year period. For five measures, involving the use of aspirin and beta-blockers for acute myocardial infarction and oxygenation assessment for pneumonia, U.S. hospitals are approaching optimal performance.
Williams and colleagues' findings are reassuring but not surprising. In 2003, Jencks and colleagues, from the Centers for Medicare and Medicaid Services (CMS), reported national changes in hospital performance for Medicare beneficiaries on 22 measures of quality,7 of which 12 were substantively similar to those reported in this issue of the Journal. According to data collected by Medicare Quality Improvement Organizations through contracted clinical data abstraction centers, the median state's performance improved on 20 of the 22 measures, by an average of 12 percent, from 1998 and 1999 to 2000 and 2001. For all but one measure, states with poor performance at baseline improved more (in absolute terms) than states with good performance at baseline. Although the specifications of the 12 similar measures have changed slightly, Williams and colleagues present convincing evidence that the improvements documented among Medicare beneficiaries before 2001 have continued through 2004 and have benefited non-Medicare patients as well. Similar improvements have been reported from the Department of Veterans Affairs health care system.8
Also in this issue of the Journal, Jha and colleagues9 describe variations in clinical performance during the first half of 2004 among hospitals participating in the CMS Hospital Quality Alliance program, which began as a voluntary public reporting program with limited participation but grew substantially after the implementation of financial incentives for participation. Although Jha and colleagues report on several of the same measures as Williams and colleagues, the JCAHO and CMS specifications differed slightly until this year, making it difficult to compare results side by side. Nonetheless, the work of Jha and colleagues complements that of Williams and colleagues by identifying settings in which quality remains below average. Specifically, midwestern and northeastern hospitals perform better, on average, than southern and western hospitals in all three clinical domains (i.e., acute myocardial infarction, congestive heart failure, and pneumonia). Northern and central communities (e.g., Boston; Indianapolis; Kansas City, Mo.; and Camden, N.J.) dominate the list of "top-ranked performers," whereas southeastern and southwestern communities (e.g., Little Rock, Ark.; Orlando, Fla.; Miami; San Bernardino, Calif.; and San Diego, Calif.) dominate the list of "bottom-ranked performers." Nonprofit hospitals slightly but consistently outperform for-profit hospitals; teaching hospitals outperform nonteaching hospitals with respect to cardiac care but not pneumonia care.
What are the implications of these two studies for the future of improvements in quality in the U.S. hospital industry? First, they establish that very high levels of adherence with evidence-based guidelines are achievable, with sufficient education of physicians and hospital managers and sufficient attention from outside entities such as JCAHO and CMS. Of course, these high levels of adherence may be somewhat illusory, since physicians and hospitals have become increasingly clever at documenting questionable contraindications to standard therapies, thereby excluding many patients who might have benefited from them.
Second, these studies confirm that variation in hospital performance, at least for three medical conditions, has generally shrunk in recent years. Poorly performing hospitals improved more than highly performing hospitals during the period from 2002 to 2004, and regional disparities appear to be smaller now than they were in 1998 and 1999.10 Of course, variation in quality persists, and the data reported by Williams et al. and Jha et al. do not address more pernicious disparities related to insurance status and other sociodemographic characteristics.
Third, these studies confirm the long-recognized fact that repeatedly observed behaviors improve over time — a phenomenon known as the "Hawthorne effect," after the location of the facility in which it was described.11 CMS and JCAHO chose to focus on acute myocardial infarction, congestive heart failure, and pneumonia because these conditions frequently lead to hospitalization and death and because medical therapy for these conditions is clearly effective. However, these conditions do not account for the majority of hospitalizations in the United States. We have no idea whether care for other conditions has deteriorated, even while care for acute myocardial infarction, congestive heart failure, and pneumonia has improved. It is even possible that unmeasured aspects of care for these three conditions have deteriorated. Williams and colleagues' inability to find any decrease in inpatient mortality related to acute myocardial infarction, neonatal mortality, or obstetric lacerations (unpublished data) suggest that the effect of the reported improvements on public health may be modest.
Despite the progress that we have made in hospital care for acute myocardial infarction, congestive heart failure, and pneumonia, we still face major challenges. Medical knowledge continues to expand every year, so practice guidelines and the measures of quality on which they are based require continual updating. Quality measures that are not updated to reflect current research findings will lose their value, and sponsors will lose their credibility.
As hospitals and physicians become more sophisticated in "gaming" quality measures, sponsors must also become more sophisticated in monitoring accuracy. For example, the largest absolute improvements between 2002 and 2004 were documented for smoking-cessation counseling and discharge instructions, which are measures that hospitals can manipulate through check-off forms that nurses complete when they discharge patients with the target conditions. Although any patient documented as receiving aspirin almost certainly received it, we have no such confidence with regard to smoking-cessation counseling. Educational interventions that are effective in a clinical trial may fail abysmally when they are transformed into a check box on a discharge form. Patient and family surveys may be helpful to monitor the delivery of such interventions.
Finally, we must not rest on our laurels and assume that we have solved the problem of quality by improving 15 measures for three conditions in about 1400 to 2000 acute care hospitals. Jha and colleagues show that performance remains mediocre at hospitals that do not meet the sample-size requirement for public reporting.9 Brennan and colleagues showed that there is substantial room for improving safety among surgical patients.3 McGlynn and colleagues showed that patients receive only 50 to 60 percent of indicated interventions across multiple domains of predominantly outpatient care.1 Indeed, performance improved by a median of 5.4 percent on 24 measures of hospital quality between the 2003 and 2004 editions of the National Healthcare Quality Report but by only 1.4 percent on 49 measures of the quality of ambulatory care.12 We have barely begun to touch quality-related problems in mental health and substance-abuse care, pediatric care, and home health care. As physicians and health professionals, we have made a little progress, but we still have far to go in closing the "quality chasm" that the Institute of Medicine recognized in 2001.5
Source Information
From the Division of General Medicine and the Center for Health Services Research in Primary Care, University of California–Davis, Sacramento.
References
McGlynn EA, Asch SM, Adams J, et al. The quality of health care delivered to adults in the United States. N Engl J Med 2003;348:2635-2645.
Fisher ES, Wennberg DE, Stukel TA, Gottlieb DJ, Lucas FL, Pinder EL. The implications of regional variations in Medicare spending. 1. The content, quality, and accessibility of care. Ann Intern Med 2003;138:273-287.
Brennan TA, Leape LL, Laird NM, et al. Incidence of adverse events and negligence in hospitalized patients: results of the Harvard Medical Practice Study I. N Engl J Med 1991;324:370-376.
Thomas EJ, Studdert DM, Burstin HR, et al. Incidence and types of adverse events and negligent care in Utah and Colorado. Med Care 2000;38:261-271.
Committee on Quality of Health Care in America, Institute of Medicine. Crossing the quality chasm: a new health system for the 21st century. Washington, D.C.: National Academies Press, 2001.
Williams SC, Schmaltz SP, Morton DJ, Koss RG, Loeb JM. Quality of care in U.S. hospitals as reflected by standardized measures, 2002-2004. N Engl J Med 2005;353:255-264.
Jencks SF, Huff ED, Cuerdon T. Change in the quality of care delivered to Medicare beneficiaries, 1998-1999 to 2000-2001. JAMA 2003;289:305-312.
Jha AK, Perlin JB, Kizer KW, Dudley RA. Effect of the transformation of the Veterans Affairs health care system on the quality of care. N Engl J Med 2003;348:2218-2227.
Jha AK, Li Z, Orav EJ, Epstein AM. Care in U.S. hospitals -- the Hospital Quality Alliance program. N Engl J Med 2005;353:265-274.
Jencks SF, Cuerdon T, Burwen DR, et al. Quality of medical care delivered to Medicare beneficiaries: a profile at state and national levels. JAMA 2000;284:1670-1676.
Roethlisberger FJ, Dickson WJ. Management and the worker: technical vs. social organization in an industrial plant. Cambridge, Mass.: Harvard University Press, 1934.
2004 National Healthcare Quality Report (NHQR). Rockville, Md.: Agency for Healthcare Research and Quality, 2004. (AHRQ publication no. 05-0013.)(Patrick S. Romano, M.D., )
In this issue of the Journal, Williams and colleagues,6 from the Joint Commission on Accreditation of Healthcare Organizations (JCAHO), present time-series data from their "core measures" program, in which hospitals seeking accreditation must collect and submit data on clinical performance according to standardized, evidence-based measure sets. They bring us good news. During the period from late 2002 through early 2004, continuously participating hospitals improved significantly on 15 of the 18 measures for acute myocardial infarction, heart failure, and pneumonia. These hospitals improved significantly on every measure that involved appropriate use of medications or provision of counseling, and the trend was favorable for all three of the measures that did not show statistically significant improvement (i.e., mean time to thrombolysis and inpatient mortality after acute myocardial infarction and blood cultures for pneumonia). Furthermore, hospitals with the poorest performance at baseline improved the most over this two-year period. For five measures, involving the use of aspirin and beta-blockers for acute myocardial infarction and oxygenation assessment for pneumonia, U.S. hospitals are approaching optimal performance.
Williams and colleagues' findings are reassuring but not surprising. In 2003, Jencks and colleagues, from the Centers for Medicare and Medicaid Services (CMS), reported national changes in hospital performance for Medicare beneficiaries on 22 measures of quality,7 of which 12 were substantively similar to those reported in this issue of the Journal. According to data collected by Medicare Quality Improvement Organizations through contracted clinical data abstraction centers, the median state's performance improved on 20 of the 22 measures, by an average of 12 percent, from 1998 and 1999 to 2000 and 2001. For all but one measure, states with poor performance at baseline improved more (in absolute terms) than states with good performance at baseline. Although the specifications of the 12 similar measures have changed slightly, Williams and colleagues present convincing evidence that the improvements documented among Medicare beneficiaries before 2001 have continued through 2004 and have benefited non-Medicare patients as well. Similar improvements have been reported from the Department of Veterans Affairs health care system.8
Also in this issue of the Journal, Jha and colleagues9 describe variations in clinical performance during the first half of 2004 among hospitals participating in the CMS Hospital Quality Alliance program, which began as a voluntary public reporting program with limited participation but grew substantially after the implementation of financial incentives for participation. Although Jha and colleagues report on several of the same measures as Williams and colleagues, the JCAHO and CMS specifications differed slightly until this year, making it difficult to compare results side by side. Nonetheless, the work of Jha and colleagues complements that of Williams and colleagues by identifying settings in which quality remains below average. Specifically, midwestern and northeastern hospitals perform better, on average, than southern and western hospitals in all three clinical domains (i.e., acute myocardial infarction, congestive heart failure, and pneumonia). Northern and central communities (e.g., Boston; Indianapolis; Kansas City, Mo.; and Camden, N.J.) dominate the list of "top-ranked performers," whereas southeastern and southwestern communities (e.g., Little Rock, Ark.; Orlando, Fla.; Miami; San Bernardino, Calif.; and San Diego, Calif.) dominate the list of "bottom-ranked performers." Nonprofit hospitals slightly but consistently outperform for-profit hospitals; teaching hospitals outperform nonteaching hospitals with respect to cardiac care but not pneumonia care.
What are the implications of these two studies for the future of improvements in quality in the U.S. hospital industry? First, they establish that very high levels of adherence with evidence-based guidelines are achievable, with sufficient education of physicians and hospital managers and sufficient attention from outside entities such as JCAHO and CMS. Of course, these high levels of adherence may be somewhat illusory, since physicians and hospitals have become increasingly clever at documenting questionable contraindications to standard therapies, thereby excluding many patients who might have benefited from them.
Second, these studies confirm that variation in hospital performance, at least for three medical conditions, has generally shrunk in recent years. Poorly performing hospitals improved more than highly performing hospitals during the period from 2002 to 2004, and regional disparities appear to be smaller now than they were in 1998 and 1999.10 Of course, variation in quality persists, and the data reported by Williams et al. and Jha et al. do not address more pernicious disparities related to insurance status and other sociodemographic characteristics.
Third, these studies confirm the long-recognized fact that repeatedly observed behaviors improve over time — a phenomenon known as the "Hawthorne effect," after the location of the facility in which it was described.11 CMS and JCAHO chose to focus on acute myocardial infarction, congestive heart failure, and pneumonia because these conditions frequently lead to hospitalization and death and because medical therapy for these conditions is clearly effective. However, these conditions do not account for the majority of hospitalizations in the United States. We have no idea whether care for other conditions has deteriorated, even while care for acute myocardial infarction, congestive heart failure, and pneumonia has improved. It is even possible that unmeasured aspects of care for these three conditions have deteriorated. Williams and colleagues' inability to find any decrease in inpatient mortality related to acute myocardial infarction, neonatal mortality, or obstetric lacerations (unpublished data) suggest that the effect of the reported improvements on public health may be modest.
Despite the progress that we have made in hospital care for acute myocardial infarction, congestive heart failure, and pneumonia, we still face major challenges. Medical knowledge continues to expand every year, so practice guidelines and the measures of quality on which they are based require continual updating. Quality measures that are not updated to reflect current research findings will lose their value, and sponsors will lose their credibility.
As hospitals and physicians become more sophisticated in "gaming" quality measures, sponsors must also become more sophisticated in monitoring accuracy. For example, the largest absolute improvements between 2002 and 2004 were documented for smoking-cessation counseling and discharge instructions, which are measures that hospitals can manipulate through check-off forms that nurses complete when they discharge patients with the target conditions. Although any patient documented as receiving aspirin almost certainly received it, we have no such confidence with regard to smoking-cessation counseling. Educational interventions that are effective in a clinical trial may fail abysmally when they are transformed into a check box on a discharge form. Patient and family surveys may be helpful to monitor the delivery of such interventions.
Finally, we must not rest on our laurels and assume that we have solved the problem of quality by improving 15 measures for three conditions in about 1400 to 2000 acute care hospitals. Jha and colleagues show that performance remains mediocre at hospitals that do not meet the sample-size requirement for public reporting.9 Brennan and colleagues showed that there is substantial room for improving safety among surgical patients.3 McGlynn and colleagues showed that patients receive only 50 to 60 percent of indicated interventions across multiple domains of predominantly outpatient care.1 Indeed, performance improved by a median of 5.4 percent on 24 measures of hospital quality between the 2003 and 2004 editions of the National Healthcare Quality Report but by only 1.4 percent on 49 measures of the quality of ambulatory care.12 We have barely begun to touch quality-related problems in mental health and substance-abuse care, pediatric care, and home health care. As physicians and health professionals, we have made a little progress, but we still have far to go in closing the "quality chasm" that the Institute of Medicine recognized in 2001.5
Source Information
From the Division of General Medicine and the Center for Health Services Research in Primary Care, University of California–Davis, Sacramento.
References
McGlynn EA, Asch SM, Adams J, et al. The quality of health care delivered to adults in the United States. N Engl J Med 2003;348:2635-2645.
Fisher ES, Wennberg DE, Stukel TA, Gottlieb DJ, Lucas FL, Pinder EL. The implications of regional variations in Medicare spending. 1. The content, quality, and accessibility of care. Ann Intern Med 2003;138:273-287.
Brennan TA, Leape LL, Laird NM, et al. Incidence of adverse events and negligence in hospitalized patients: results of the Harvard Medical Practice Study I. N Engl J Med 1991;324:370-376.
Thomas EJ, Studdert DM, Burstin HR, et al. Incidence and types of adverse events and negligent care in Utah and Colorado. Med Care 2000;38:261-271.
Committee on Quality of Health Care in America, Institute of Medicine. Crossing the quality chasm: a new health system for the 21st century. Washington, D.C.: National Academies Press, 2001.
Williams SC, Schmaltz SP, Morton DJ, Koss RG, Loeb JM. Quality of care in U.S. hospitals as reflected by standardized measures, 2002-2004. N Engl J Med 2005;353:255-264.
Jencks SF, Huff ED, Cuerdon T. Change in the quality of care delivered to Medicare beneficiaries, 1998-1999 to 2000-2001. JAMA 2003;289:305-312.
Jha AK, Perlin JB, Kizer KW, Dudley RA. Effect of the transformation of the Veterans Affairs health care system on the quality of care. N Engl J Med 2003;348:2218-2227.
Jha AK, Li Z, Orav EJ, Epstein AM. Care in U.S. hospitals -- the Hospital Quality Alliance program. N Engl J Med 2005;353:265-274.
Jencks SF, Cuerdon T, Burwen DR, et al. Quality of medical care delivered to Medicare beneficiaries: a profile at state and national levels. JAMA 2000;284:1670-1676.
Roethlisberger FJ, Dickson WJ. Management and the worker: technical vs. social organization in an industrial plant. Cambridge, Mass.: Harvard University Press, 1934.
2004 National Healthcare Quality Report (NHQR). Rockville, Md.: Agency for Healthcare Research and Quality, 2004. (AHRQ publication no. 05-0013.)(Patrick S. Romano, M.D., )