Key Themes and Highlights From the National Healthcare Quality Report

Since 2003, the Agency for Healthcare Research and Quality (AHRQ), together with its partners in the Department of Health and Human Services (HHS), has reported on progress and opportunities for improving health care quality. With this fifth annual National Healthcare Quality Report (NHQR), these reports will have provided more than 50,000 data points about health care quality in the United States. Has it made a difference? Have Federal and State governmental agencies, provider organizations, insurers, and employers made progress in improving health care quality and safety? While every previous release of the NHQR has attempted to summarize the direction in which health care quality is going, this fifth report tries to summarize the progress that has been made and the remaining challenges to improve health care quality in this Nation.

The NHQR is built on 218 measures categorized across four dimensions of quality—effectiveness, patient safety, timeliness, and patient centeredness. This year's report focuses on the state of health care quality for a group of 41 core report measures that represent the most important and scientifically credible measures of quality for the Nation, as selected by the HHS Interagency Work Group.ⁱ The distillation of 42 core measures for the 2007 report provides a more readily understandable summary and explanation of the key results derived from the data.ⁱⁱ While the measures selected for inclusion in the NHQR are derived from the most current scientific knowledge, this knowledge base is not evenly distributed across health care. The analysis in the following pages centers on measures for which data are available from the baseline year of 2000 or 2001 and the comparison year of 2004 or 2005.

Three themes that emerge from the 2007 NHQR emphasize the need to accelerate progress in achieving high quality health care:

• Health care quality continues to improve, but the rate of improvement has slowed.
• Variation in quality of health care across the Nation is decreasing, but not for all measures.
• The safety of health care has improved since 2000, but more needs to be done.

---

ⁱ The HHS Interagency Work Group, which represents 18 HHS agencies and offices, was formed to provide advice and support to AHRQ and the National Reports team.

ⁱⁱ Data on all NHQR measures are available in the Data Tables Appendix at www.ahrq.gov/qual/measurix.htm.

---

Return to Contents

Health Care Quality Continues To Improve, But the Rate of Improvement Has Slowed

For the past 5 years, the NHQR has summarized trends in health care quality. This is a difficult undertaking, as there is no single national health care quality survey that collects a standard set of data elements from a uniform population over the same time period. Rather, data are available from a wide range of sources that focus on different populations and data years.

In order to track the progress of health care quality in this country, the NHQR presents an annual rate of change in quality, which represents how quickly the health care system is making improvements across the report's core measures. Another way to describe this is the speed of improvement in the U.S. health care system. Based on these core report measures, quality of care continues to improve. However, the rate of improvement seems to be slowing.

Figure H.1. Median rate of change overall and by care setting, 1994-2005 and 2000-2005

Note: Available data years for the 1994-2005 analysis vary based on the specific measure, as not all measures have data for every data year. Details on the measures included in these rates of change are presented in the NHQR Measure Specifications Appendix.

• The annual median rate of change for all core measures, which span the years 1994 to 2005, is 2.3%ⁱⁱⁱ (Figure H.1).
• More recently, the rate of improvement has slowed. From 2000 to 2005, the annual median rate of change for measures with available data was 1.5%.
• As reported in last year's NHQR, however, most measures show some improvement. Of the 41 core measures reported above with data that span 1994 to 2005, 27 improved, 6 declined, and 6 are unchanged.

---

ⁱⁱⁱ Not all data sources provide data for each year from 1994 through 2005: 1994 is the earliest data year for any data source reported in the 2007 NHQR Data Tables Appendix, and 2006 is the latest data year for any data source reported in the 2007 NHQR Data Tables Appendix.

---

When examining change across multiple diseases and care settings, it is difficult to understand why changes in performance occur. In the analysis of trends for this year's NHQR, it is clear that some areas have shown increasing rates of improvement while others have slowed. For example, the rate of improvement in heart disease treatment increased from 3.3% to 5.6% (1994-2005 versus 2000-2005). However, the rate of improvement in diabetes slowed from 1.2% to 0.6%. Initiatives such as public reporting and strong advocacy from multiple stakeholders in support of quality are circumstances that may influence broad system change and subsequent quality improvements in certain areas. However, these data show that sustaining a steady rate of improvement over time is a challenge.

Return to Contents

Variation in Quality of Health Care Across the Nation Is Decreasing, But Not for All Measures

One goal of quality improvement efforts nationally is to reduce differences in health care quality that patients receive in one State versus another. There is no justification, for example, for a patient hospitalized for a heart attack in California to have different care than a patient in Alabama. Yet, analyses from the NHQR, its companion National Healthcare Disparities Report (NHDR), other organizations, and the health care literature in general have shown that the quality of care that patients receive varies significantly across the country. This variability is evident in multiple dimensions, according to many different factors, such as social and demographic characteristics of patient populations, hospital types (e.g., urban, rural, teaching, non-teaching), and different clinical areas (e.g., heart disease, pneumonia, clinical preventive services).^{1, 2, 3}

For the past 20 years, a central focus of quality improvement efforts has been to bring care for all patients to a minimum quality standard based on evidence.⁴ Reporting on unwarranted variation across States in health care quality has been part of past NHQRs. The NHQR is a rich source of information on quality of care across the 104 measures for which State level information is available. This year's NHQR examines whether progress is being made at reducing variation in care for the time period 2000 to 2005.^iv

More measures have seen progress in terms of decreases in variation between the best performing State and the worst performing State between 2000 and 2005. Specifically:

• Variability decreased for 28 measures.
• Variability increased for 13 measures.
• There was no change for 18 measures.

For example, there has been progress in standardizing high quality care across the country, such as the percentage of heart attack patients given smoking cessation counseling while in the hospital (Figure H.2).

---

^iv In the past, the NHQR has presented variation in health care quality by showing measures with high ratios between the best and worst performing States. To examine whether variation was increasing, this year's NHQR examined these ratios across the 59 measures for which State data were available for 2000-2001 to 2004-2005 and for which the same States provided data for both data years. Then analysis was conducted to determine whether more measures had seen increases or decreases, or whether the ratio of best to worst State had not changed. Data were analyzed for measures on which the same States reported data for both time periods in order to ensure that appropriate comparisons were made across the same States across time.

---

Figure H.2. Percentage of heart attack patients given smoking cessation counseling while hospitalized, by State, 2002 and 2005

• In 2002, nationally, heart attack patients admitted to hospitals were counseled about quitting smoking less than 50% of the time.
• In 2002, the rate at which heart attack patients in hospitals were counseled to stop smoking by their doctor in the best performing State was 3.3 times higher than the rate for their counterparts in the worst performing State.
• However, over just 3 years, most States improved their performance on this measure. The latest national data (2005) show that heart attack patients admitted to hospitals are now counseled to quit smoking 91% of the time.
• In addition, the variation across the country on this measure has decreased. Figure H.2 shows that all States in 2005 had reached the level of the best performing States in 2002. In 2005, only two States were below 80% on this measure.

Although overall variation has decreased slightly, many of the measures with the most variation—that is, where patients are treated very differently based on the fact that they are in a different State—have not improved since the first NHQR. Moreover, if all States were able to achieve the highest State's performance even on these measures alone, it would mean significant advances in terms of patient care and outcomes. For example:

• In 2002, chronic care nursing home residents were restrained 8.3 times more frequently in the worst performing State than in the best performing State. In 2006, this ratio increased to 10. If all States had reached the average of the best performing State in 2006, at least 61,500 fewer residents would have been physically restrained nationwide.
• In 2000, diabetic patients were admitted to the hospital with uncontrolled diabetes 7.6 times more often in the worst performing State than in the best performing State. By 2004, this difference had nearly doubled, with uncontrolled diabetes admissions per 100,000 population in the worst performing State 14 times higher than in the best performing State. If all States had reached the level of the four best performing States in 2004, almost 39,000 fewer patients would have been admitted for uncontrolled diabetes, with a potential cost savings of $216.7 million.
• While some States may never be able to achieve the performance of the top States because of differences in their population's health risks, we know from this analysis and from other research that considerable costs,^{5, 6, 7} are incurred as a result of hospitalizations and other health services that are potentially avoidable.

Return to Contents

The Safety of Health Care Has Improved Since 2000, But More Needs To Be Done

For 5 years, the National Healthcare Quality and Disparities Reports have presented a snapshot of the safety of health care provided to the American people. This analysis has been based on an emerging set of databases that were created to respond to the need for information documented in such publications as the Institute of Medicine's landmark 2000 report To Err Is Human.⁸ Today a clearer picture of trends in health care safety is only beginning to emerge. In part, this results from the intense complexity required to accurately assess and ensure complete reporting on medical errors and patient safety events.

Key databases developed to date that have enhanced the current knowledge base on patient safety nationally include the Medicare Patient Safety Monitoring System, AHRQ's Healthcare Cost and Utilization Project's Patient Safety Indicators, and the National Nosocomial Infections Surveillance System.^v While these have been developed or expanded only in recent years, results from these efforts contribute to the trends that can be drawn from the 5 years of reporting on patient safety in the NHQR.

---

^v Data sources for patient safety information contained in the National Healthcare Quality and Disparities Reports:

• The Quality Improvement Organization Program, Centers for Medicare & Medicaid Services.
• The Medicare Patient Safety Monitoring System, Centers for Medicare & Medicaid Services.
• The Healthcare Cost and Utilization Project, AHRQ.
• The National Nosocomial Infections Surveillance System, a cooperative, nonfinancial relationship between hospitals and the Centers for Disease Control and Prevention.

---

Figure H.3. Median annual rate of change in quality by measurement area, 2000-2005

• Safety improved slightly between 2000 and 2005 at an annual rate of 1% improvement (Figure H.3).
• However, during that same time period, other areas of health care improved at significantly greater rates, with care for heart disease improving at 5.6% during this time and care across all measures with available data improving at nearly double the rate of improvement for safety.
• Areas where significant attention has been concentrated, such as the appropriate timing of antibiotics for surgery and reducing medication errors, have shown progress. Specifically:
  • Over 30% more patients received appropriate timing for antibiotics before and after surgery in 2005 than in 2004.
  • From 2004 to 2005, adverse drug events from war far in and low- molecular-weight heparin declined 21% and 28%, respectively.
  • Deaths following complications of care declined 2.4% from 1994 to 2004.

However, in many areas, a fraction of the information necessary to track patient safety effectively is available, and, in most cases, clinicians still lack basic monitoring information on medical errors and adverse events.⁹ For example:

• Only 25 of 41 safety measures in the full NHQR measure set have data available for tracking recent trends (2000 to 2005).
• No national data systems currently report data on safety at a State or local level.
• Because adverse events and medical errors are rare events from a statistical perspective, it is difficult, using current data, to compare across time or patient groups in order to examine potential causes and possible solutions to safety issues.

Return to Contents

National Standards, Neighborhood Solutions

The NHQR continues to present a broad examination of health care quality in the United States. As noted above, while quality continues to improve across most measures, the pace of change has slowed. Improvements have been demonstrated in such areas as patient safety and care for certain diseases and populations. However, these improvements have not been realized across all populations or across the entire country. Moreover, even in areas where progress has occurred, significant effort is needed to ensure that adequate systems are in place to track progress and use the data for improvement.

A number of major efforts have begun to accomplish this goal of using data for improvement. Specifically:

The President signed the Patient Safety and Quality Improvement Act of 2005 (Patient Safety Act) to spur the development of voluntary, provider-driven initiatives to improve the quality, safety, and outcomes of patient care. The Act, which will be implemented by AHRQ and the HHS Office for Civil Rights, addresses many of the current barriers to improving patient care. All too often, health care providers fear that if they participate in the analysis of medical errors or patient care processes, the findings may be used against them in court or harm their professional reputations. The reluctance of providers to participate in improvement initiatives, combined with the difficulty of aggregating and sharing data confidentially across facilities or State lines, limits our current ability to aggregate data in sufficient numbers to identify rapidly the most prevalent risks and hazards in the delivery of patient care, their underlying causes, and practices that are most effective in mitigating them.

The Patient Safety Act addresses these barriers to improvement through the following goals and mechanisms:

• Encourage greater health care provider participation in improvement initiatives by establishing strong, nationally uniform confidentiality and privilege protections for the patient safety information that these initiatives assemble or develop.
• Expand the analytic expertise that is available to health care providers to analyze safety and quality issues by encouraging the formation of new patient safety organizations (PSOs) with which providers can voluntarily work.
• Improve our ability to rapidly recognize and address the underlying causes of risks and hazards to patient care by facilitating the aggregation and analysis of large numbers of patient safety events.

The Patient Safety Act directs the Secretary to develop regional and national statistics and trends for reporting in future NHQRs. AHRQ will carry out these activities on behalf of the Secretary and is required to develop this information through the aggregation and analysis of non-identifiable patient safety data that PSOs, providers, and others voluntarily contribute to a network of patient safety databases that the statute envisions.

Creating ways that data can be used for benchmarking and improvement is a key step in driving quality improvement. AHRQ's State Snapshots Web tool was launched in 2005. It is an application that helps State health leaders, researchers, consumers, and others understand the status of health care quality in individual States. The 51 State Snapshots—for every State plus the District of Columbia—are based on 129 quality measures, each of which evaluates a different segment of health care performance and shows each State's strengths and weaknesses. Although the measures are the products of complex statistical formulas, they are expressed on the Web site as simple, five-color graphical "performance meters." The State Snapshots provide summaries that measure health care quality in three different contexts: by type of care (such as preventive, acute, or chronic care), by setting of care (such as nursing homes or hospitals), and by clinical area (such as care for patients with cancer, diabetes, or respiratory diseases). After selecting a State on a national map, users can view whether that State has improved or worsened compared with other States in a particular area of health care delivery. The 129 measures range from preventing bed sores to screening for diabetes-related foot problems to providing antibiotics quickly to hospitalized pneumonia patients. The State Snapshots also allow users to compare a State's performance against that of other States in the same region and to see how a State compares against best performing States.

The release of this year's State Snapshots is complemented by the launch of NHQRnet and NHDRnet, a pair of new, interactive Web-based tools for searching AHRQ's storehouse of national health care data. These online search engines allow users to create spreadsheets and customize searches of information in the NHQR and NHDR.

Return to Contents

Efforts To Improve Measures and Data

The measures and data used in these reports come from a variety of sources within HHS and elsewhere. Performance measures are central to ongoing efforts to improve the quality of care within this country. Yet, despite progress in quality improvement, a number of challenges remain related to the current state of the science in the performance measurement field, including the following:

• Inadequately developed measures that lack reliability and validity can lead to inaccuracies in determining the presence or absence of quality.
• The existence of multiple and, at times, conflicting performance measures that are neither uniform nor standardized may cause confusion.
• Measure maintenance is an issue as the evidence base of measures is refined and expanded.
• The burden of data collection for performance measurement is a growing concern.

Efforts by HHS to address these and other concerns related to improving performance measures and data have been ongoing. For instance, results of validation studies for measures, including field testing, expert panel review, focus group testing, and the like, have provided insights that significantly contributed to measure refinement and, in other cases, demonstrated the reliability and accuracy of the existing performance measures.

Research that develops new scientific evidence has also contributed to the improvement of measures and can provide fruitful avenues for measure development. Agencies within HHS have also worked with a number of stakeholders, such as measure developers, consumers, purchasers, various quality alliances, and the National Quality Forum, to come to consensus around a core set of measures that can be used to report quality. This collaboration helps to harmonize measures and provides a forum for discussions on the current state of measurement. A key area is identifying gaps that, if filled, will provide a comprehensive picture of performance in health care. Another key area is identifying state-of-the-art methods, such as risk adjustment methodologies, that can enhance performance measures to provide an unbiased representation of quality.

HHS is also exploring other opportunities to reduce the burden of collecting the data necessary for performance measurement and quality improvement activities. Of particular interest is research that explores cost- efficient approaches to data collection, such as standardization of common data elements contained in the electronic health record. These data elements may enhance existing performance measures, making them more reliable and accurate. They also can make performance data collection a useful and automatic byproduct of the care delivered rather than an independent activity. This change can enable doctors, nurses, and other health care professionals to spend more time caring for patients.

Performance measurement is a tool for continuous quality improvement, and the broad areas cited above depict some of the continuing activities HHS is currently supporting in its efforts to improve and enhance the quality of care delivered within the United States.

Return to Contents

Efforts To Make Patients' Lives Better

The NHQR concentrates on the national view of health care quality. This view of health care quality is often far removed from the daily reality faced by health care providers and patients in clinics and hospitals. At the same time, however, the statistics that are reported in the National Healthcare Quality and Disparities Reports reflect the everyday experiences of patients and their doctors and nurses across the Nation. It makes a difference in people's lives when breast cancer is diagnosed early with timely mammography; when a patient suffering from a heart attack is given the correct life-saving treatment in a timely fashion; when medications are correctly administered; and when doctors listen to their patients, show them respect, and answer their questions.

These are the statistics that are reported in this year's NHQR. This report documents important progress in making patients' lives better. At the same time, however, it highlights many areas where much more could be done to use the data in the National Healthcare Quality and Disparities Reports to target policy and clinical interventions to improve care. Each of the 50,000 data points that have been produced and reported during the past 5 years represents groups of patients across the country. The hope is that the next 5 years will see greater use of data for decisionmaking, so that those patients begin to experience true quality improvement in American health care.

Return to Contents

References

1. Agency for Healthcare Research and Quality. State Snapshots. Available at: http://statesnapshots.ahrq.gov/statesnapshots/index.jsp?menuId=1&state=. Accessed August 6, 2007.

2. Cantor JC, Schoen C, Belloff D, et al. Aiming higher: results from a state scorecard on health system performance. The Commonwealth Fund Commission on a High Performance Health System; June 2007.

3. Kahn CN, Ault T, Isenstein H, et al. Snapshot of hospital quality reporting and pay- for- performance under Medicare. Health Aff. 2006 Jan-Feb;25(1):148-62.

4. Vladeck BC. Everything new is old again. Health Aff. 2004;Suppl Web Exclusives: VAR108-11.

5. Klonoff D, Schwartz D. An economic analysis of interventions for diabetes. Diabetes Care. 2000;23(3):390-404.

6. Herman W, Eastman R. The effects of treatment on the direct costs of diabetes. Diabetes Care. 1998;21(Suppl 3):C19-C24.

7. Beaulieu N, Cutler D, Ho K, et al. The business case for diabetes disease management at two managed care organizations: a case study of Health Partners and Independent Health Association. New York: The Commonwealth Fund; 2003. Available at www.cmwf.org/programs/quality/beaulieu_diabetesdiseasemanagement_610.pdf. Accessed December 17, 2003.

8. Kohn LT, Corrigan JM, Donaldson MS, eds. To err is human: building a safer health system. A report of the Committee on Quality of Health Care in America, Institute of Medicine. Washington, DC: National Academy Press; 2000.

9. Berwick DM. Errors today and errors tomorrow. N Engl J Med. 2003;348(25):2570-72.

Return to Contents