Page 1 of 1

Highlights from the National Healthcare Quality and Disparities Report

National Healthcare Quality Report, 2010

Health care seeks to prevent, diagnose, and treat disease and to improve the physical and mental well-being of all Americans. Across the lifespan, health care helps people stay healthy, recover from illness, live with chronic disease or disability, and cope with death and dying. Quality health care delivers these services in ways that are safe, timely, patient centered, efficient, and equitable.

Unfortunately, Americans too often do not receive care that they need, or they receive care that causes harm. Care can be delivered too late or without full consideration of a patient's preferences and values. Many times, our system of health care distributes services inefficiently and unevenly across populations. Some Americans receive worse care than other Americans. These disparities may be due to differences in access to care, provider biases, poor provider-patient communication, and poor health literacy.

Each year since 2003, the Agency for Healthcare Research and Quality (AHRQ) has reported on progress and opportunities for improving health care quality and reducing health care disparities. Guided by a subcommittee of AHRQ's National Advisory Council and a Department of Health and Human Services (HHS) Interagency Work Group,i past reports were built on more than 250 measures categorized across six dimensions: effectiveness, patient safety, timeliness, patient centeredness, efficiency, and access to care. As mandated by the U.S. Congress, the National Healthcare Quality Report (NHQR) focuses on "national trends in the quality of health care provided to the American people" (42 U.S.C. 299b-2(b)(2)) while the National Healthcare Disparities Report (NHDR) focuses on "prevailing disparities in health care delivery as it relates to racial factors and socioeconomic factors in priority populations" (42 U.S.C. 299a-1(a)(6)).

The 2010 reports and this summary incorporate a number of recommendations made by the Institute of Medicine (IOM). The IOM first provided guidance to AHRQ on the NHQR and NHDR in 2002. In 2008, AHRQ again asked the IOM to offer suggestions for enhancing future reports and associated products to ensure that these reports raise awareness of the performance of the U.S. health care system. In April 2010, AHRQ received advice from the IOM Committee on Future Directions for the National Healthcare Quality and Disparities Reports.ii

In the past, separate Highlights were produced for each report. This year, we have integrated findings from the 2010 NHQR and 2010 NHDR to produce a single summary document. This is intended to reinforce the need to consider simultaneously the quality of health care and disparities across populations when assessing our health care system. The National Healthcare Reports Highlights seeks to address three questions critical to guiding Americans toward the optimal health care they need and deserve:

  • What is the status of health care quality and disparities in the United States?
  • How have health care quality and disparities changed over time?
  • Where is the need to improve health care quality and reduce disparities greatest?

Consistent with past reports, the 2010 reports emphasize one of AHRQ's priority populations as a theme. This year, we present expanded analyses of care across the urban-rural continuum, and the National Healthcare Reports Highlights includes a summary of care received by residents of different types of geographic areas. Finally, this document summarizes information on eight national priorities identified by the IOM Committee and presents novel strategies for improving quality and reducing disparities from AHRQ's Health Care Innovations Exchange (HCIE).

Four themes from the 2010 NHQR and 2010 NHDR emphasize the need to accelerate progress if the Nation is to achieve higher quality and more equitable health care in the near future.

  • Health care quality and access are suboptimal, especially for minority and low-income groups.
  • Quality is improving; access and disparities are not improving.
  • Urgent attention is warranted to ensure improvements in quality and progress on reducing disparities with respect to certain services, geographic areas, and populations, including:
    • Cancer screening and management of diabetes.
    • States in the central part of the country.
    • Residents of inner-city and rural areas.
    • Disparities in preventive services and access to care.
  • Progress is uneven with respect to eight national priority areas:
    • Two are improving in quality: (1) Palliative and End-of-Life Care and (2) Patient and Family Engagement.
    • Three are lagging: (3) Population Health, (4) Safety, and (5) Access.
    • Three require more data to assess: (6) Care Coordination, (7) Overuse, and (8) Health System Infrastructure.
    • All eight priority areas showed disparities related to race, ethnicity, and socioeconomic status.

Health Care Quality and Access Are Suboptimal, Especially for Minority and Low-Income Groups

A key function of the reports is to summarize the state of health care quality, access, and disparities for the Nation. This undertaking is difficult, as no single national health care database collects a comprehensive set of data elements that can produce national and State estimates for all population subgroups each year. Rather, data come from more than three dozen databases that provide estimates for different population subgroups and data years. While most data are gathered annually, some data are not collected regularly or are old. Despite the data limitations, our analyses indicate that health care quality in America is suboptimal. The gap between best possible care and that which is routinely delivered remains substantial across the Nation.

In the reports, measures are classified as either process measures or outcome measures. Process measures are further subdivided, when possible, into preventive care, acute treatment, and chronic disease management.

On average, people received the preventive services tracked in the reports two-thirds of the time. Moreover, wide variation was found in receipt of different types of preventive services. For instance, 20% of high-risk adults ages 18-64 ever received pneumococcal vaccination, but 94% of children ages 19-35 months received 3 doses of polio vaccine.

On average, people received appropriate acute care services three-quarters of the time. Rates of receipt of acute care services ranged from a low of 8% among patients who needed and received treatment for an alcohol problem at a specialty facility to a high of 94% of hospitalized patients who indicated that communication with their doctors was good.

On average, patients received recommended chronic disease management services three-quarters of the time. Again, receipt of chronic disease management services varied widely, from 17% of dialysis patients being registered on a kidney transplant waiting list to 95% of hospice patients receiving the right amount of pain medication.

Access to care is also far from optimal. On average, Americans report barriers to care one-fifth of the time, ranging from 3% of people saying they were unable to get or had to delay getting prescription medications to 60% of people saying their usual provider did not have office hours on weekends or nights.

All Americans should have equal access to high-quality care. Instead, we find that racial and ethnic minorities and poor people often receive poorer quality of care and face more barriers when trying to access care. To assess disparities, we focus on a set of "core" measures,iii which includes the most important and scientifically supported measures in the full reports measure set.

For each measure, we examine the relative difference between a selected group and its reference group. Differences that are statistically significant, are larger than 10%, and favor the reference group are labeled as indicating poor quality or access for the selected group. Differences that are statistically significant, are larger than 10%, and favor the selected group are labeled as indicating better quality or access for the selected group. Differences that are not statistically significant or are smaller than 10% are labeled as the same between the selected group and the reference group.

 

Figure H.1. Distribution of core quality measures for which members of selected groups experienced better, same, or worse quality of care compared with reference group

Stacked bar chart. Comparison of performance on core quality measures for population subgroups. Number of measures: Black vs. White, 48, Asian vs. White, 19, American Indian or Alaska Native vs. White, 24, Hispanic vs. Non-Hispanic White, 43, Poor vs. High Income, 23. Black versus White: Worse, 21, Same, 21, Better, 6. Asian versus White: Worse, 4, Same, 10, Better, 5. American Indian or Alaska Native versus White: Worse, 9, Same, 10, Better, 5. Hispanic versus Non-Hispanic White: Worse, 24, Same, 13, Bette

Key: AI/AN = American Indian or Alaska Native; NHW = non-Hispanic White; n = number of measures.
Better = Population received better quality of care than reference group.
Same = Population and reference group received about the same quality of care.
Worse = Population received worse quality of care than reference group.

  • Disparities in quality of care are common:
    • Blacks and American Indians and Alaska Natives received worse care than Whites for about 40% of core measures.
    • Asians received worse care than Whites for about 20% of core measures.
    • Hispanics received worse care than non-Hispanic Whites for about 60% of core measures.
    • Poor people received worse care than high-income peopleiv for about 80% of core measures.

Figure H.2. Distribution of core access measures for which members of selected groups experienced better, same, or worse access to care compared with reference group

Stacked bar chart. Comparison of performance on core access measures for population subgroups. Number of measures: Black versus White, 6, Asian versus White, 5, American Indian or Alaska Native versus White, 5, Hispanic versus Non-Hispanic White, 6, Poor versus High Income, 6. Black versus White: Worse, 2, Same, 4, Better, 0. Asian versus White: Worse, 1, Same, 1, Better, 3. American Indian or Alaska Native versus White: Worse, 1, Same, 4, Better, 0. Hispanic versus Non Hispanic-White: Worse, 5, Same, 0, Be

Key: AI/AN = American Indian or Alaska Native; NHW = non-Hispanic White; n = number of measures.
Better = Population had better access to care than reference group.
Same = Population and reference group had about the same access to care.
Worse = Population had worse access to care than reference group.

  • Disparities in access are also common, especially among Hispanics and poor people:
    • Blacks had worse access to care than Whites for one-third of core measures.
    • Asians and American Indians and Alaska Natives had worse access to care than Whites for 1 of 5 core measures.
    • Hispanics had worse access to care than non-Hispanic Whites for 5 of 6 core measures.
    • Poor people had worse access to care than high-income people for all 6 core measures.

Quality Is Improving; Access and Disparities Are Not Improving

Suboptimal health care is undesirable, but we may be less concerned if we observe evidence of vigorous improvement. Hence, the second key function of the reports is to examine change over time. To track the progress of health care quality and access in this country, the reports present annual rates of change, which represent how quickly quality of and access to services delivered by the health care system are improving or declining. Another way to describe rate of change is the speed of improvement or decline in health care quality and access.

As in past NHQRs, regression analysis is used to estimate annual rate of change for each measure. Annual rate of change is calculated only for measures with at least 3 years of data. For most measures, trends include data points from 2001-2002 to 2007-2008. We label measures going in a favorable direction at a rate exceeding 1% per year as improving, going in an unfavorable direction at a rate exceeding 1% per year as worsening, and changing at a rate less than 1% per year as not changing.

This year, for the first time, we introduce a similar method for assessing change in disparities using regression results. When a selected group's rate of change is at least 1% higher than the reference group's rate of change, we label the disparity as improving. When a selected group's rate of change is at least 1% lower than the reference group's rate of change, we label the disparity as worsening. When the difference in rates is less than 1%, we label the disparity as no change.

 

Figure H.3. Change in quality and access over time

Stacked bar chart. Changes in quality and access measures. Quality: Worsening, 26, No Change 41, Improving, 112. Access: Worsening, 9, No change, 6, Improving, 7.

Improving = Quality or access is going in a positive direction at an average annual rate greater than 1% per year.
No Change = Quality or access is not changing or is changing at an average annual rate less than 1% per year.
Worsening = Quality or access is going in a negative direction at an average annual rate greater than 1% per year.

  • Quality is improving slowly. Across all 179 measures of health care quality tracked in the reports, almost two-thirds showed improvement. However, median rate of change was only 2.3% per year.
  • Access is not improving. Across the 22 measures of health care access tracked in the reports, about 70% did not show improvement and 40% were headed in the wrong direction. Median rate of change was -0.6% per year, indicating no change over time.

Figure H.4. Distribution of changes over time in racial, ethnic, and socioeconomic disparities for core quality measures

Stacked bar chart. Changes in disparities for quality measures. Number of measures: Black versus White, 47, Asian versus White, 19, American Indian or Alaska Native versus White, 22, Hispanic versus Non-Hispanic White, 41, Poor versus High Income, 20. Black versus White: Worsening, 2, No Change, 40, Improving, 5. Asian versus White: Worsening, 0, No Change, 13, Improving, 6. American Indian or Alaska Native versus White, Worsening, 6, No Change, 15, Improving, 1. Hispanic versus Non Hispanic White: Worsenin

Key: AI/AN = American Indian or Alaska Native; n = number of measures.
Improving = Disparity is getting smaller at a rate greater than 1% per year.
No Change = Disparity is not changing or is changing at a rate less than 1% per year.
Worsening = Disparity is getting larger at a rate greater than 1% per year.

  • Few disparities in quality of care are getting smaller.
    • Fewer than 20% of disparities faced by Blacks, American Indians and Alaska Natives, Hispanics, and poor people showed evidence of narrowing.
    • The Asian-White gap was narrowing for about 30% of core measures, the largest proportion of any group, but most disparities were not changing.

Figure H.5. Distribution of changes over time in racial, ethnic, and socioeconomic disparities for core access measures

Stacked bar chart. Changes in disparities for access measures. Black versus White: Worsening, 0, No Change, 6, Improving, 0. Asian versus White: Worsening, 0, No Change, 4, Improving, 1. American Indian or Alaska Native versus White: Worsening, 0, No change, 5, Improving, 0. Hispanic versus Non Hispanic White: Worsening, 0, No Change, 6, Improving, 0. Poor versus High Income: Worsening, 0, No Change, 6, Improving, 0.

Key: AI/AN = American Indian or Alaska Native.
Improving = Disparity is getting smaller at a rate greater than 1% per year.
No Change = Disparity is not changing or is changing at a rate less than 1% per year.
Worsening = Disparity is getting larger at a rate greater than 1% per year.

  • Almost no disparities in access to care are getting smaller.
    • Among disparities in core access measures, only one showed reduction.

Urgent Attention Is Warranted To Ensure Improvements in Quality and Progress on Reducing Disparities

The third key function of the reports is to identify which areas are in greatest need of improvement. Potential problem areas can be defined in terms of types of services, parts of the country, and populations at risk.

Variation Across Types of Services

Pace of improvement varies across preventive services, acute treatment, and chronic disease management.

 

Figure H.6. Change in quality over time by type of measure

Stacked bar chart. Change in quality by measure type. Number of measures: Preventive Care, 35, Acute Treatment, 21, Chronic Disease Management, 22, Outcome, 90. Preventive Care: Worsening, 4, No Change, 10, Improving, 21. Acute Treatment: Worsening, 2, No Change, 2, Improving, 17. Chronic Disease Management: Worsening, 3, No Change, 4, Improving, 15. Outcome: Worsening, 15, No Change, 25, Improving, 50.

Key: n = number of measures.
Improving = Quality is going in a positive direction at an average annual rate greater than 1% per year.
No Change = Quality is not changing or is changing at an average annual rate less than 1% per year.
Worsening = Quality is going in a negative direction at an average annual rate greater than 1% per year.

  • Measures of acute treatment are improving; measures of preventive care and chronic disease management are lagging.
    • While both process and outcome measures are improving, rates of improvement are faster among processes of care. Median rate of improvement of process measures was 2.6% per year compared with 1.6% per year for outcome measures.
    • Among process measures, the highest rate of improvement was in measures related to treatment of acute illnesses or injuries. Of the 21 process of care measures related to acute treatment, about 80% showed improvement.
    • In contrast, of the 35 process measures related to preventive services, only 60% showed improvement. Of the 22 process measures related to chronic disease management, about 70% showed improvement. This may reflect the high proportion of hospital measures included in acute treatment, many of which are tracked by the Centers for Medicare & Medicaid Services and publicly reported.

Table H1. Process measures getting worse over time

Preventive CareAcute TreatmentChronic Disease Management

Women age 40+ who received a mammogram in the last 2 years

Women age 18+ who received a Pap smear in the last 3 years

Adults age 50+ who received a fecal occult blood test in the last 2 years

Children ages 19-35 months who received 3 doses of Haemophilus influenzae type B vaccine

Emergency department (ED) visits lasting 6+ hours and resulting in admission to the hospital or transfer to another facility per 1,000 ED visits

Adults age 40+ with diabetes who received a hemoglobin A1c measurement in the calendar year

Adults age 40+ with diabetes who received a dilated eye examination in the calendar year

Adults age 40+ with diabetes who had their feet checked for sores in the calendar year

  • Several measures related to cancer screening and management of patients with diabetes have worsened over time.

Variation Across Parts of the Country

Quality of care varies not only across types of care but also across parts of the country. Knowing where to focus efforts improves the efficiency of interventions. Delivering data that can be used for local benchmarking and improvement is a key step in raising awareness and driving quality improvement. Since 2005, AHRQ has examined variation across States in the State Snapshots tool (http://statesnapshots.ahrq.gov). This Web site helps State health leaders, researchers, consumers, and others understand the status of health care quality in individual States and the District of Columbia. The State Snapshots are based on more than 100 NHQR measures, each of which evaluates a different aspect of health care performance and shows each State's strengths and weaknesses.

The 2010 reports introduce a new method for examining variation across States and benchmarking quality of care. For measures with State data, we calculate the benchmark as the average for the top 10% of States; this average is referred to as the "achievable benchmark." Achievable benchmarks are believed to be more actionable because they represent a level of performance that has been demonstrated in the real world. Here, we examine the frequency with which States perform in the top 10% of States and contribute to the achievable benchmark.

Data are not available for all States for all measures. Thus, States that have less health care data, either because it is not collected or because samples are too small to generate reliable estimates, have fewer opportunities to be in the top 10%. In addition, State data are more readily available from vital statistics and from hospitals, nursing homes, and home health agencies, while State data are much more limited for important topics such as quality of HIV or mental health care and care coordination. Policies that improve data collection at the State level would allow benchmarking across a broader array of measures. It is hoped that as health information technologies continue to expand, more information will become available for finer geographic units as well as for more granular subpopulations.

 

Figure H.7. Number of measures for which a State was in the top 10% by quartile

Number of measures for which a State was in the top 10 percent by quartile. Highest quartile: Colorado, Connecticut, District of Columbia, Hawaii, Maine, Massachusetts, Minnesota, New Hampshire, Rhode Island, South Carolina, Utah, Vermont. Second quartile: Alabama, Arkansas, Georgia, Idaho, Illinois, Kansas, Kentucky, Mississippi, Montana, Missouri, Oklahoma, South Dakota, Washington, West Virginia, Wisconsin. Third quartile: Alaska, Arizona, California, Delaware, Florida, Iowa, Louisiana, Maryland, Nebrask

Highest quartile = State in the top 10% for the largest number of measures.
Lowest quartile = State in the top 10% for the smallest number of measures.

  • Two parts of the country led in performance. While every State was in the top 10% for some measure and was part of a benchmark, States in the New England (CT, MA, ME, NH, RI, VT) and Pacific (AK, CA, HI, OR, WA) census divisions were benchmark States most often and States in the East North Central (IL, IN, MI, OH, WI), East South Central (AL, KY, MS, TN), and West South Central (AR, LA, OK, TX) divisions were benchmark States less often.

Table H2. States most often contributing to the top 10%

Preventive CareAcute TreatmentChronic Disease ManagementOutcome
Delaware
Massachusetts
New Hampshire
Maine
New Hampshire
New Jersey
Vermont
Hawaii
Minnesota
New Hampshire
South Carolina
Arizona
District of Columbia
Hawaii
Utah
  • New England did best on preventive care and acute treatment; western States did best on outcomes of care.

Variation Across Populations at Risk

Previous reports have emphasized one of AHRQ's priority populations as a theme. This year, we present expanded analyses of care across the urban-rural continuum. AHRQ is charged with examining the care received by residents of inner-city and rural areas. People who live in these areas often face unique barriers to care related to provider availability and transportation.  As in past reports, we have categorized areas into the following urban-rural categories:

  • Large central metropolitan statistical area (MSA): Central counties (inner city) in metropolitan area of 1 million+ inhabitants.
  • Large central metropolitan area: Outlying counties (suburbs) in metropolitan area of 1 million+ inhabitants.
  • Medium MSA: Counties in metropolitan area of 250,000 to 1,000,000 inhabitants.
  • Small MSA: Counties in metropolitan area of 50,000 to 250,000 inhabitants.
  • Micropolitan statistical area: Counties with an urban cluster of 10,000 to 50,000 inhabitants.
  • Noncore statistical area: Counties outside of metropolitan or micropolitan areas.

For comparisons across areas, residents of large fringe MSAs (large city suburbs) are used as the reference group since these counties have the lowest levels of poverty and typically have the best quality and access to health care.

 

Figure H.8. Distribution of quality measures for which residents of specific areas experienced better, same, or worse quality of care compared with residents of large fringe metropolitan areas

Stacked bar chart. Comparison of quality in metropolitan and rural areas relative to large fringe metropolitan areas. Large Central: Worse, 20, Same, 64, Better, 3. Medium: Worse, 8, Same, 75, Better, 4. Small: Worse, 16, Same, 61, Better, 6. Micropolitan: Worse, 26, Same, 54, Better, 6. Noncore: Worse, 21, Same, 46, Better, 9.

Key: n = number of measures.
Better = Population received better quality of care than reference group.
Same = Population and reference group received about the same quality of care.
Worse = Population received worse quality of care than reference group.
Note: Number of measures varies for each group because for some measures, data for some groups were not available.

  • Residents of the inner-city and rural areas sometimes receive worse quality of care. For most measures of quality of health care, differences across the urban-rural continuum were small. However, some disparities are noted. Compared with residents of large city suburbs, residents of large inner cities received worse care for about a quarter of quality measures tracked in the reports. Residents of micropolitan and noncore areas (areas typically regarded as "rural") received worse care for about 30% of measures.
  • Disparities related to preventive care were common across urban and rural areas while disparities related to diabetes were largest for residents of large inner cities and noncore rural areas.

Figure H.9. Distribution of access measures for which residents of specific areas experienced better, same, or worse access to care compared with residents of large fringe metropolitan areas

Stacked bar chart. Comparison of access in metropolitan and rural areas relative to large fringe metropolitan areas. Large Central: Worse, 9, Same, 15, Better, 0. Medium: Worse, 7, Same, 16, Better, 1. Small: Worse, 8, Same, 16, Better, 0. Micropolitan: Worse, 12, Same, 11, Better, 1. Noncore: Worse, 9, Same, 12, Better, 2.

Key: n = number of measures.
Better = Population received better quality of care than reference group.
Same = Population and reference group received about the same quality of care.
Worse = Population received poorer quality of care than reference group.
Note: Number of measures varies for each group because for some measures, data for some groups were not available.

  • Disparities in access to care across the urban-rural continuum tended to be more common than disparities in quality of care. Compared with residents of large city suburbs, residents of large inner cities had worse access to care for about 35% of access measures tracked in the reports. Residents of micropolitan areas had worse access to care for 50% of access measures. Residents of noncore areas had worse access to care for about 40% of access measures.

The NHDR focuses on disparities related to race, ethnicity, and socioeconomic status. Table H.3 summarizes the largest disparities for each major group tracked in the reports that are either getting larger or staying the same. The table shows that many groups experience disparities related to preventive services and access to care.

 

Table H3. Largest racial, ethnic, and socioeconomic disparities in core quality and access measures that are not improving

GroupsMeasureRR
Black compared with WhiteHospital admissions for lower extremity amputations per 1,000 population age 18 and over with diabetes2.0
Emergency department visits where patients left without being seen1.7
Adults age 65 and over who ever received pneumococcal vaccination1.5
Asian compared with WhiteAdults age 65 and over who ever received pneumococcal vaccination1.4
People with a usual primary care provider1.3
Adults who had a doctor's office or clinic visit in the last 12 months whose health providers listened carefully, explained things clearly, respected what they had to say, and spent enough time with them1.2
American Indian/Alaska Native compared with WhitePeople under age 65 with health insurance1.7
Hospital patients with heart failure who received recommended hospital care1.7
Adults age 50 and over who ever received colorectal cancer screening1.6
Hispanic compared with Non-Hispanic WhiteNew AIDS cases per 100,000 population age 13 and over3.3
People under age 65 with health insurance2.7
People with a specific source of ongoing care2.0
Poor compared with High IncomePeople under age 65 with health insurance4.7
Female Medicare beneficiaries age 65 and over who reported ever being screened for osteoporosis with a bone mass or bone density measurement3.6
People with a specific source of ongoing care2.9

Key: RR indicates rate relative to reference group.
Note: To compare RRs, measures were framed negatively. Hence, an RR greater than 1 indicates that a group is receiving poorer quality of care or facing larger problems with access to care compared with the reference group. For example, an RR of 1.3 indicates that Asians are 1.3 times more likely than Whites not to have a usual primary care provider.


Proceed to Next Section

Current as of February 2011
Internet Citation: Highlights from the National Healthcare Quality and Disparities Report: National Healthcare Quality Report, 2010. February 2011. Agency for Healthcare Research and Quality, Rockville, MD. http://www.ahrq.gov/research/findings/nhqrdr/nhqr10/Key.html