Chapter 2. Effectiveness

National Healthcare Disparities Report, 2010

As better understanding of health and sickness has led to superior ways of preventing, diagnosing, and treating diseases, the health of most Americans has improved dramatically. However, ample evidence indicates that some Americans do not receive the full benefits of high-quality care.

This chapter is organized around eight clinical areas (cancer, diabetes, end stage renal disease, heart disease, HIV and AIDS, maternal and child health, mental health and substance abuse, and respiratory diseases) and three types of health care services that typically cut across clinical conditions (lifestyle modification, functional status preservation and rehabilitation, and supportive and palliative care). The 11 sections of this chapter highlight a small number of core and supporting measures.

In this chapter, process measures are organized into several categories related to the patient's need for preventive care, treatment of acute illness, and chronic disease management. These are derived from the original Institute of Medicine categories: staying healthy, getting better, living with illness or disability, and coping with the end of life. There is sizable overlap among these categories, and some measures may be considered to belong in more than one category. Outcome measures are organized separately because prevention, treatment, and management can all play important roles in affecting outcomes.

Prevention

Caring for healthy people is an important component of health care. Educating people about healthy behaviors can help to postpone and avoid illness and disease. In addition, detecting health problems at an early stage increases the chances of effectively treating them, often reducing suffering and costs.

Treatment

Even when preventive care is ideally implemented, it cannot entirely avert the need for acute care. Delivering optimal treatment for acute illness can help reduce the effects of illness and promote the best recovery possible.

Management

Some diseases, such as diabetes and end stage renal disease, are chronic, which means they cannot simply be treated once; they must be managed across a lifetime. Management of chronic disease often involves promotion and maintenance of lifestyle changes and regular contact with a provider to monitor the status of the disease. For patients, effective management of chronic disease can mean the difference between healthy living and frequent medical problems.

Outcomes

Many factors other than health care influence health outcomes, including a person's genes, lifestyle, and social and physical environment. However, for many individuals, appropriate preventive services, timely treatment of acute illness or injury, and meticulous management of chronic disease can positively affect mortality, morbidity, and quality of life. 

 

Cancer

Importance

Mortality
Number of deaths (2007)562,8751
Cause of death rank (2007)2nd1
Prevalence
Number of living Americans who have been diagnosed with cancer (2007)11,713,7362
Incidence
New cases of cancer (2010)1,529,5603
New cases of breast cancer (2010)209,0603
Cost
Total costi (2010 est.)$263.8 billion4
Direct costsii (2010 est.)$102.8 billion4
Indirect costs (2010 est.)$161.0 billion4
Cost-effectivenessiii of colorectal cancer screening$35,000-$165,000/QALY5

Measures

Evidence-based consensus defining good quality care and how to measure it currently exists for only a few cancers and a few aspects of care. Breast and colorectal cancers have high incidence rates and are highlighted in alternate years of the report. The 2009 National Healthcare Disparities Report (NHDR) highlighted breast cancer; this year's focus is on colorectal cancer. The core report measures are:

  • Colorectal cancer screening.
  • Colorectal cancer first diagnosed at advanced stage.

As in previous reports, the 2010 NHDR includes one supporting measure for colorectal cancer care from the National Cancer Data Base that has been endorsed by the National Quality Forum:

  • Surgical resection of colon cancer that includes at least 12 lymph nodes.

Findings

Prevention: Colorectal Cancer Screening

Colorectal cancer is the third most common cancer in adults. Prevention of colorectal cancer includes modifying risk factors such as weight, physical activity, smoking, and alcohol use, as well as screening for early disease. Screening is important because early stages of colorectal cancer may not present any symptoms, and screening can detect abnormal growths before they develop into cancer.3,6 Early detection increases treatment options and the chances for survival. The U.S. Preventive Services Task Force recommends colorectal cancer screening for men and women age 50 and over. The screening measured in the NHDR includes having a fecal occult blood test in the past 2 years or ever having received flexible sigmoidoscopy, colonoscopy, or proctoscopy.

 

Figure 2.1. Adults age 50 and over who reported receiving colorectal cancer screening (received fecal occult blood test in past 2 years or ever received colonoscopy, sigmoidscopy, or proctoscopy), by race, ethnicity, and income, 2000-2008

Total, 2000, 49.8, 2003, 51.7, 2005, 55.5, 2008, 60.1. White, 2000, 50.8, 2003, 52.6, 2005, 56.8, 2008, 61.3. Black, 2000, 43.8, 2003, 47.3, 2005, 48.6, 2008, 54.4. Asian, 2000, 42.1, 2003, 37.6, 2005, 42.4, 2008, 53.6. AI/AN, 2000, 48.7, 2003, 41.6, 2005, 38.1, 2008, 36.7. More than one race, 2000, 53.7, 2003, 47, 2005, 55.8, 2008, 55.9. 2008 Achievable Benchmark: 67.1%.  Trend line chart, percentage of colorectal cancer screening by ethnicity for years 2000 - 2008. Non-Hispanic White, 2000, 51.7, 2003, 54, 2005, 58.5, 2008, 63.1. Hispanic, 2000, 34.8, 2003, 37.1, 2005, 37.3, 2008, 42.8. 2008 Achievable Benchmark: 67.1%.
Trend line chart, composite measure percentage of colorectal cancer screening for the uninsured for years 2000 - 2008. Non-Hispanic White , 2000, 26.2, 2003, 24.2, 2005, 26.7, 2008, 29.8. Non-Hispanic Black , 2000, 27.1, 2003, 25.3, 2005, 24.9, 2008, 31.3. Hispanic, 2000, 15.9, 2003, 14.3, 2005, 17.1, 2008, 13.3.

Key: AI/AN = American Indian or Alaska Native.
Source: Centers for Disease Control and Prevention, National Center for Health Statistics, National Health Interview Survey (NHIS), 2000, 2003, 2005, and 2008.
Denominator: Adults age 50 and over in the civilian noninstitutionalized population. Benchmark is derived from the Behavioral Risk Factor Surveillance System (BRFSS); refer to Introduction and Methods for details.

  • In all 4 years, the percentage of Whites who ever received colorectal cancer screening was significantly higher than that of Blacks and Asians (Figure 2.1)
  • In all 4 years, the percentage of non-Hispanic Whites who ever received colorectal cancer screening was significantly higher than that of Hispanics.
  • In all 4 years, the percentage of high-income individuals who ever received colorectal cancer screening was significantly higher than the percentage of poor, low-income, and middle-income individuals.
  • The top 5 State achievable benchmark was 67.1%.iv At the current annual rate of increase of approximately 1.5%, the benchmark could be achieved overall in 4.6 years.
  • Whites could reach the benchmark in 4.3 years while Blacks and Asians could reach the benchmark in about 10 years. Non-Hispanic Whites could reach the benchmark in 2.7 years but Hispanics would not reach the benchmark for 25 years.
  • Middle-income adults could reach the benchmark in about 7 years and low-income adults in approximately 8 years. High-income adults have already achieved the benchmark. There is no indication that poor adults are progressing toward the benchmark.

Also, in the National Healthcare Quality Report (NHQR):

  • In all available data years, the percentage of adults age 50 and over residing in large fringe metropolitan areas who ever received colorectal cancer screening was significantly higher than it was for adults residing in large central metropolitan and noncore areas.
  • Racial and ethnic minorities, as well as people with low incomes, have disproportionate rates of individuals with public insurance or no insurance. To distinguish the effects of race, ethnicity, and income on cancer screening, this measure is stratified by insurance status.

Figure 2.2. Composite measure: Adults ages 50-64 who reported receiving colorectal cancer screening (received fecal occult blood test in past 2 years or ever received colonoscopy, sigmoidscopy, or proctoscopy), by race and ethnicity, stratified by insurance, 2000-2008

Trend line chart, composite measure percentage of colorectal cancer screening for the privately insured for years 2000 - 2008. Non-Hispanic White, 2000, 47.7, 2003, 50.7, 2005, 55.6, 2008, 62.2. Non-Hispanic Black, 2000, 44.8, 2003, 47, 2005, 47.5, 2008, 55.1. Hispanic, 2000, 35.6, 2003, 38.5, 2005, 39.1, 2008, 44.4.           Trend line chart, composite measure percentage of colorectal cancer screening for the publicly insured for years 2000 - 2008. Non-Hispanic White, 2000, 48, 2003, 48.2, 2005, 51.3, 2008, 55.2. Non-Hispanic Black, 2000, 34.3, 2003, 43.9, 2005, 47.8, 2008, 50.4. Hispanic, 2000, 32.6, 2003, 36.5, 2005, 34.8, 2008, 48.2.  

Trend line chart, composite measure percentage of colorectal cancer screening for the uninsured for years 2000 - 2008. Non-Hispanic White , 2000, 26.2, 2003, 24.2, 2005, 26.7, 2008, 29.8. Non-Hispanic Black , 2000, 27.1, 2003, 25.3, 2005, 24.9, 2008, 31.3. Hispanic, 2000, 15.9, 2003, 14.3, 2005, 17.1, 2008, 13.3.

Source: Centers for Disease Control and Prevention, National Center for Health Statistics, National Health Interview Survey (NHIS), 2000, 2003, 2005, and 2008.
Denominator: Adults ages 50-64 in the civilian noninstitutionalized population.

  • Between 2000 and 2008, non-Hispanic Whites and non-Hispanic Blacks ages 50-64 who had private insurance showed significant improvement in the percentage of adults who reported receiving colorectal cancer screening (Figure 2.2). During the same time, non-Hispanic Blacks and Hispanics with public insurance also showed significant improvement in the percentage of adults who ever received colorectal cancer screening.
  • Non-Hispanic Whites ages 50-64 with private insurance had significantly higher rates of colorectal screening than non-Hispanic Blacks with private insurance in 2 of the four data years measured and significantly higher rates than Hispanics with private insurance in all 4 years.
  • In 2008, among individuals with public insurance ages 50-64, there were no statistically significant differences between non-Hispanic Whites and non-Hispanic Blacks or between non-Hispanic Whites and Hispanics.
Outcome: Advanced Stage Colorectal Cancer

Cancers can be diagnosed at different stages of development. Cancers diagnosed early before spread has occurred are generally more amenable to treatment and cure; cancers diagnosed late with extensive spread often have poor prognoses. The rate of cancer cases that are diagnosed at advanced stages is a measure of the effectiveness of cancer screening efforts and of adherence to followup care after a positive screening test. Because many cancers often take years to develop, changes in rates of late-stage cancer may lag behind changes in rates of screening.

Differences in rates may vary across racial and ethnic groups due to differences in underlying prevalence of colorectal cancer.

 

Figure 2.3. Colorectal cancer diagnosed at advanced stage (tumors diagnosed at regional or distant stage) per 100,000 population age 50 and over, by race and ethnicity, 2000-2007

Trend line chart, rate per 100,000 population of colorectal cancer diagnosed at advanced stage by race for years 2000 - 2007. White, 2000, 94.5, 2001, 93.2, 2002, 90.6, 2003, 87.7, 2004, 83.5, 2005, 80.9, 2006, 77.8, 2007, 75.2.  Black, 2000, 118.1, 2001, 118.1, 2002, 114, 2003, 115.1, 2004, 110.6, 2005, 106, 2006, 98.3, 2007, 96.4.  API, 2000, 79.8, 2001, 84.3, 2002, 79.8, 2003, 71.5, 2004, 69.1, 2005, 68.2, 2006, 65.3, 2007, 63.9. AI/AN, 2000, 53.9, 2001, 58.9, 2002, 57.2, 2003, 54.4, 2004, 60, 2005, 44.8  Trend line chart, rate per 100,000 population of colorectal cancer diagnosed at advanced stage by ethnicity for years 2000 - 2007. Non-Hispanic White, 2000, 96.7, 2001, 95, 2002, 92.7, 2003, 89.3, 2004, 85.1, 2005, 82.5, 2006, 79.2, 2007, 77.1. Hispanic, 2000, 73.4, 2001, 74.1, 2002, 70, 2003, 71.5, 2004, 67.7, 2005, 66.1, 2006, 65.2, 2007, 59.2.  NHDR 10 - Chapter 2 Figure 2-3b Image

Key: API = Asian or Pacific Islander; AI/AN = American Indian or Alaska Native.
Source: National Cancer Institute, Surveillance, Epidemiology, and End Results Program, 2000-2007.
Denominator: Adults ages 50-64 in the civilian noninstitutionalized population.
Note: Advanced stage colorectal cancer is defined as local stage with tumor size greater than 2 cm diameter, regional stage, or distant stage. Data are SEER 2000-2007 limited-use database, 17 registries, released April 2010, based on the November 2009 submission. The 2005 data were adjusted for areas impacted by Hurricanes Katrina and Rita.

  • From 2000 to 2007, the rate of advanced stage colorectal cancer was significantly lower for Asians and Pacific Islanders (APIs) and American Indians and Alaska Natives (AI/ANs) than for Whites (Figure 2.3).
  • From 2000 to 2007, Blacks had significantly higher rates of colorectal cancer diagnosed at advanced stage compared with Whites. During the same period, Hispanics had significantly lower rates of advanced stage colorectal cancer compared with non-Hispanic Whites.

Also, in the NHQR:

  • From 2000 to 2007, the rate of advanced stage colorectal cancer in adults ages 50-64 significantly decreased, from 45.7 to 40.1 per 100,000 population.
  • During the same period, adults age 65 and over also saw a significant decrease, from 154.2 to 119.2 per 100,000 population.
  • In all years, adults age 65 and over had significantly higher rates of advanced stage colorectal cancer than adults ages 50-64.
Treatment: Recommended Care for Colorectal Cancer Patients

Different diagnostic and treatment options exist for various types of cancer. Some aspects of cancer care are well established as beneficial and are commonly recommended. The appropriateness of recommended care depends on different factors, such as the stage or the extent of the cancer within the body (especially whether the disease has spread from the original site to other parts of the body). Other types of care are important for accurate diagnosis, such as ensuring the adequate examination of lymph nodes when surgery (e.g., to remove color cancer) is performed.

 

Figure 2.4. Patients who received surgical resection of colon cancer that included at least 12 lymph nodes pathologically examined, by race and ethnicity, 2003-2007

Trend line chart, percentage, patients received resection, 12 lymph nodes examined, by race, 2003 through 2007. Total, 2003, 51.7, 2004, 55.1, 2005, 60.1, 2006, 66.9, 2007, 77. White, 2003, 51.4, 2004, 55, 2005, 60.1, 2006, 66.9, 2007, 77.1. Black, 2003, 52.4, 2004, 53.8, 2005, 58.4, 2006, 66, 2007, 76.3. Asian, 2003, 56.4, 2004, 59.7, 2005, 64.8, 2006, 69.2, 2007, 76.5. AI/AN, 2003, 55.6, 2004, 41.2, 2005, 50.8, 2006, 61.3, 2007, 72.8.           Trend line chart, percentage, patients received resection, 12 lymph nodes examined, by ethnicity, 2003 through 2007. Non-Hispanic White, 2003, 51.4, 2004, 55, 2005, 60.3, 2006, 66.9, 2007, 76.9. Hispanic, 2003, 52.5, 2004, 56.5, 2005, 61.2, 2006, 67.1, 2007, 77.2.

Key: AI/AN = American Indian or Alaska Native.
Source: Commission on Cancer, American College of Surgeons and American Cancer Society, National Cancer Data Base, 2000-2007
Denominator: U.S. population age 50 and over.

  • The overall percentage of adults diagnosed with colorectal cancer who received recommended care significantly increased from 51.7% in 2003 to 77.0% in 2007 (data not shown). Similar improvement was observed among all racial and ethnic groups during this period (Figure 2.4).

Also, in the NHQR:

  • From 2003 to 2007, the percentage of colorectal cancer patients who received recommended care significantly increased in all residence locations. The percentage of large metropolitan colorectal cancer patients who received recommended care was significantly higher in all years than that of micropolitan patients and noncore patients. 

 

Diabetes

Importance

Mortality
Number of deaths (2007)71,8321
Cause of death rank (2007)7th1
Prevalence
Total number of Americans with diabetes (2007)23.6 million7
Number of people with diagnosed diabetes (2007)17.9 million7
Number of people with undiagnosed diabetes (2007)5.7 million7
Incidence
New cases (age 20 and over, 2007)1.6 million7
Cost
Total cost (2007 est.)$174 billion7
Direct medical costs (2007 est.)$116 billion7

Measures

Routine monitoring of blood glucose levels with hemoglobin A1c (HbA1c)v tests and dilated eye and foot examinations have been shown to help prevent or mitigate complications of diabetes, such as diabetic neuropathy, retinopathy, and vascular and kidney disease.8 With more than half a million discharges in 2006, diabetes is one of the leading causes of hospitalization in the United States.9 However, with appropriate and timely ambulatory care, it may be possible to prevent many hospitalizations for diabetes and related complications.

The core measure reported in this section examines the extent to which individuals with diabetes receive care needed to prevent complications or slow the disease's progression:

  • Receipt of three recommended diabetes services.

In addition, three supporting outcome measures are presented. Two of these measures are included in AHRQ's Prevention Quality Indicators (PQIs).vi PQIs may be used to estimate rates of potentially avoidable hospitalizations among ambulatory care-sensitive conditions. These are hospitalizations that may have been prevented with high-quality ambulatory care and treatment.

The supporting measures from the PQIs are:

  • Hospitalization for short-term diabetes complications (PQI 1).
  • Hospitalization for lower extremity amputation (PQI 16).

The final supporting measure also offers insight into the adequacy of diabetes management:

  • Control of HbA1c, cholesterol, and blood pressure.

Findings

Management: Receipt of Three Recommended Diabetes Services

The NHDR uses a composite measure to track the national rate of the receipt of all three recommended annual diabetes interventions: an HbA1c test, an eye examination, and a foot examination. These are basic process measures that provide an assessment of the quality of diabetes management.

 

Figure 2.5. Composite measure: Adults age 40 and over with diagnosed diabetes who received three recommended services for diabetes in the calendar year (hemoglobin A1c test, dilated eye examination, and foot examination), by race, ethnicity, family income, and education, 2002-2007

Trend line chart, percentage, adults over 40 (with diabetes, received A1c, eye, foot exams in year) by race, 2002 through 2007. White, 2002, 43.3, 2003, 46.4, 2004, 43.7, 2005, 41.0, 2006, 42.2, 2007, 38.73. Black, 2002, 42.8, 2003, 36.8, 2004, 43.3, 2005, 37.0, 2006, 36.6, 2007, 31.68. 2008 Achievable Benchmark: 51.4%. Trend line chart, percentage, adults over 40 (with diabetes, received A1c, eye, foot exams in year) by ethnicity, 2002 through 2007. Non-Hispanic White, 2002, 45.1, 2003, 47.8, 2004, 45.4, 2005, 42.4, 2006, 44.6, 2007, 41.49. Hispanic, 2002, 33.7, 2003, 39.3, 2004, 35.7, 2005, 33.8, 2006, 31.6, 2007, 27.29. 2008 Achievable Benchmark: 51.4%. Trend line chart, percentage, adults over 40 (with diabetes, received A1c, eye, foot exams in year) by income, 2002 through 2007. Poor, 2002, 38.6, 2003, 31.8, 2004, 33.1, 2005, 30.3, 2006, 33.4, 2007, 22.46. Near Poor, 2002, 35, 2003, 30.5, 2004, 35.1, 2005, 28.5, 2006, 31.9, 2007, 28.63. Middle Income, 2002, 41.4, 2003, 46.5, 2004, 37.9, 2005, 38.4, 2006, 42.7, 2007, 33.13. High Income, 2002, 52.1, 2003, 55.1, 2004, 55.3, 2005, 52.6, 2006, 47.8, 2007, 51.36. 2008 Achievable Benchmark: 51.4%.  Trend line chart, percentage, adults over 40 (with diabetes, received A1c, eye, foot exams in year) by education, 2002 through 2007. Less than High School, 2002, 34.1, 2003, 36.0, 2004, 32.2, 2005, 31.5, 2006, 31.4, 2007, 29.21. High School Graduate, 2002, 43.1, 2003, 48.2, 2004, 41.7, 2005, 39.9, 2006, 42.9, 2007, 31.94. At Least Some College, 2002, 51.3, 2003, 48.3, 2004, 52.3, 2005, 47.7, 2006, 46.4, 2007, 47.1. 2008 Achievable Benchmark: 51.4%.

Source: Agency for Healthcare Research and Quality, Medical Expenditure Panel Survey, 2002-2007.
Denominator: Civilian noninstitutionalized population with diaagnosed diabetes, age 40 and over.
Note: Data include people with both type 1 and type 2 diabetes. Rate is age adjusted to the 2000 U.S. standard population. Benchmark is derived from the Behavioral Risk Factor Surveillance System (BRFSS); refer to Introduction and Methods for details.

  • Between 2002 and 2007, Blacks showed a significant decrease in the percentage of adults diagnosed with diabetes who received recommended care (42.8% to 31.7%; Figure 2.5).
  • With the exception of 2003, when results were not statistically significant, across all years, the percentage of adults with diabetes who received recommended services was significantly lower for Hispanics than for non-Hispanic Whites.
  • An association was noted between income and receipt of diabetes services. Relative to those with a high income, the percentage of adults with diabetes who received all recommended services was significantly lower among poor, near-poor, and middle-income individuals. Among near-poor individuals, this finding was observed in all years; among middle-income people, lower percentages were found except in 2003 and 2006.
  • In all years, the percentage of adults age 40 and over with diabetes who received three recommended services was significantly lower for people with less than a high school education compared with adults with at least some college education.
  • In 2007, 74.2% of high-income adults diagnosed with diabetes had their feet examined in the calendar year compared with only 62.1% of near-poor adults. Between 2002 and 2007, the percentage of near-poor individuals who had their feet examined in the calendar year significantly decreased from 73.4% to 62.1% (data not shown).
  • Between 2002 and 2007, Hispanic, poor, and near-poor adults diagnosed with diabetes all had significant decreases in the percentage of adults who received HbA1c measurement in the calendar year (data not shown).
  • The 2008 top 4 State achievable benchmark was 51.4%.vii The percentages of Whites, Blacks, non-Hispanic Whites, and Hispanics receiving recommended care show no progress toward the benchmark. High-income individuals already have achieved the benchmark but middle-income, near poor, and poor individuals show no progress toward the benchmark. Individuals with at least some college education achieved the benchmark in 2004 but individuals with a high school education or less show no progress toward the benchmark.

Also, in the NHQR:

  • With the exception of 2004, adults age 40 and over who reside in large fringe metropolitan areas were significantly more likely than those in noncore areas to receive recommended services.

Multivariate analyses were conducted to identify the independent effects of race and socioeconomic factors on several measures. Adjusted percentages are shown for receipt of diabetes services after controlling for race/ethnicity, family income, education, health insurance status, and location.

 

Figure 2.6. Composite measure: Adjusted percentages of adults ages 40-64 with diagnosed diabetes who received three recommended services for diabetes in the calendar year, by race/ethnicity, family income, education, insurance status, and residence location, 2002-2007

Multi-part bar chart (five parts), percentage, adults 40 to 64, with diabetes (three services received in calendar year), 2002 through 2007. Part One, by race/ethnicity. Non-Hispanic White, 42.1. Non-Hispanic Black, 39.2. Hispanic, 36.5. Part Two, by family income. Poor, 32.7. Low Income, 32.4. Middle Income, 37.5. High Income, 49.7. Part Three, by education. <High School, 38.1. High School Grad, 39. Some College, 43.8. Part Four, by insurance status. Private Insurance, 42.5. Public Insurance Only, 40.3. Un

Source: Agency for Healthcare Research and Quality, Medical Expenditure Panel Survey, pooled 2002-2007.
Note: Adjusted percentages are predicted marginals from a statistical model that includes the covariates race/ethnicity, family income, education, health insurance, and residence location. Chapter 1, Introduction and Methods, provides more information.

  • After adjustment, 36% of Hispanic adults with diabetes would have received all three recommended services for diabetes, which is significantly lower than the 42% of non-Hispanic Whites who would have received the three recommended services for diabetes (Figure 2.6).
  • After adjustment, among adults ages 40-64 with diagnosed diabetes, 33% of poor adults, 32% of low-income adults, and 38% of middle-income adults would have received the three recommended services for diabetes in the calendar year. These are all significantly lower than the 50% of high-income adults.
  • After adjustment, only 38% of adults with diabetes with less than a high school education and 39% of high school graduates would have received the three recommended services for diabetes. Both are significantly lower than the 44% of adults with some college education who would have received the three recommended services for diabetes.
  • After adjustment, only 30% of adults with diabetes who were uninsured all year would have received all three recommended services for diabetes, which is significantly lower than the 42% of adults who had any private insurance.
Outcome: Admissions for Short-Term Diabetes Complications

Individuals who do not achieve good control of their diabetes are more prone to short-term complications that can reduce quality of life, increase chances of death, and increase health care costs both directly and indirectly. The acute metabolic complications of diabetes consist of diabetic ketoacidosis (DKA), hyperosmolar nonketotic coma (HNC), lactic acidosis (LA), and hypoglycemia.10 Patients with DKA, HNC, and LA require hospitalization for treatment, which results in the use of significant health care resources with increased health care costs. Patients with hypoglycemia often do not require hospitalization but can still incur costs for treatment in an ambulatory setting, as well as loss of productivity. Prevention is an important component in reducing health care costs for these disorders10 and helping people with diabetes maintain optimal function.

 

Figure 2.7. Hospital admissions for diabetes with short-term complications per 100,000 population, age 18 and over, by race/ethnicity and area income, 2004-2007

Trend line chart, hospital admissions per 100,000, 2004 through 2007. Total, 2004, 57.4, 2005, 57.7, 2006, 57.8, 2007, 60.3. White, 2004, 46.0, 2005, 47.1, 2006, 46.8, 2007, 48.8. Black, 2004, 154.3, 2005, 145.4, 2006, 151.2, 2007, 159. API, 2004, 19.3, 2005, 16.4, 2006, 15.5, 2007, 17.4. Hispanic, 2004, 55.7, 2005, 56.8, 2006, 53.6, 2007, 52.5. 2008 Achievable Benchmark: 37.8 per 100,000 Population.  Trend line chart, hospital admissions per 100,000, 2004 through 2007. First Quartile, 2004, 83.9, 2005, 84.7, 2006, 91.6, 2007, 91.4. Second Quartile, 2004, 59.3, 2005, 61.8, 2006, 65.6, 2007, 65.3. Third Quartile, 2004, 46.5, 2005, 47.2, 2006, 49.4, 2007, 49.4. Fourth Quartile, 2004, 32.1, 2005, 33.9, 2006, 34.5, 2007, 33.3. 2008 Achievable Benchmark: 37.8 per 100,000 Population.

Key: API = Asian or Pacific Islander.
Source: Agency for Healthcare Research and Quality, Healthcare Cost and Utilization Project (HCUP), State Inpatient Databases (SID) disparities analysis file, 2004-2007.
Denominator: U.S. resident population age 18 and over.
Note: Short-term complications include ketoacidosis, hyperosmolarity, or coma and exclude obstetric admissions and transfers from other institutions. White, Black, and API are non-Hispanic. Data are adjusted for age, gender, and diagnosis-related group clusters.

  • In all years, the rate of hospital admissions for short-term complications was significantly higher for Blacks than for Whites. Blacks had an admission rate more than three times the rate of Whites in all years (Figure 2.7).
  • With the exception of 2007, Hispanics had significantly higher rates of hospital admissions for short-term complications compared with Whites.
  • In all years, the rate of hospital admissions for short-term complications was significantly higher for adults living in communities with median household incomes in the first quartile than for people living in communities with median household incomes in the fourth quartile. In all years, the rates of admission were about 2.5 times as high for adults living in communities with median household incomes in the first quartile compared with adults living in communities with median household incomes in the fourth quartile.
  • The 2007 top 4 State achievable benchmark was 37.8 per 100,000 population.viii At the current annual rate of increase of 1.7, there is no overall progress toward the benchmark.
  • Adults living in communities with a median income in the fourth quartile and APIs have already achieved the benchmark. Whites, Blacks, Hispanics, and adults living in communities with median incomes in the first, second, and third quartiles indicate no progress toward the benchmark.

Also, in the NHQR:

  • In all years, residents of large fringe metropolitan areas had significantly lower hospital admissions for short-term complications than residents of micropolitan areas. Residents of large fringe metropolitan areas also had significantly lower hospital admissions than residents of large central metropolitan areas in 3 of 4 years.
  • In all years, adults age 65 and over had significantly lower rates of admissions for short-term complications than adults ages 18-64.
Outcome: Controlled Hemoglobin, Cholesterol, and Blood Pressure

People diagnosed with diabetes are often at higher risk for other cardiovascular risk factors, such as high blood pressure and high cholesterol. Having these conditions in combination with diagnosed diabetes increases the likelihood of complications, such as heart and kidney diseases, blindness, nerve damage, and stroke. Patients who manage their diabetes and maintain an HbA1c level of <7%, total cholesterol of <200 mg/dL, and blood pressure of <140/80 mm Hgix can decrease these risks.

 

Figure 2.8. Adults age 40 and over with diagnosed diabetes with hemoglobin A1c, total cholesterol, and blood pressure under control, by race/ethnicity, 2001-2004 and 2005-2008

Trend line chart, percentage, adults 40 and over, with diabetes, A1c under control, 2001 through 2004 and 2005 through 2008. Non-Hispanic White, 2001-2004, 59.8, 2005-2008, 56.3. Non-Hispanic Black, 2001-2004, 41.7, 2005-2008, 47.6. Mexican American, 2001-2004, 36.8, 2005-2008, 43.9. HBA1c less than 7.0%.          Trend line chart, percentage, adults 40 and over, with diabetes, total cholesterol under control, 2001 through 2004 and 2005 through 2008. Non-Hispanic White, 2001-2004, 47.7, 2005-2008, 69.3. Non-Hispanic Black, 2001-2004, 52.7, 2005-2008, 61.4. Mexican American, 2001-2004, 51.9, 2005-2008, 56.5. Total cholesterol less than 200mg/dL.

Trend line chart, percentage, adults 40 and over, with diabetes, blood pressure under control, 2001 through 2004 and 2005 through 2008. Non-Hispanic White, 2001-2004, 63.6, 2005-2008, 57.1. Non-Hispanic Black, 2001-2004, 47.8, 2005-2008, 58.2. Mexican American, 2001-2004, 62.7, 2005-2008, 64.5. Blood pressure less than 140/80 mm Hg.

Source: Centers for Disease Control and Prevention, National Center for Health Statistics, National Health and Nutrition Examination Survey, 2001-2008.
Denominator: Civilian noninstitutionalized population with diagnosed diabetes, age 40 and over.
Note: Age adjusted to the 2000 standard population using two age groups: 40-59 and 60 and over.

  • In 2005-2008, 56.3% of non-Hispanic White adults with diabetes age 40 and over had their HbA1c level under control (Figure 2.8). This figure is significantly higher than for Mexican Americans; only 43.9% had their HbA1c under control. Comparing these results to those for 2001-2004 shows a similar pattern. Almost 60% of non-Hispanic Whites with diabetes had their HbA1c under control compared with only 36.8% of Mexican Americans.
  • In 2001-2004, 47.7% of non-Hispanic Whites had their cholesterol at optimal levels compared with 51.9% of Mexican Americans. In 2005-2008, the percentage of non-Hispanic Whites who had their cholesterol at optimal levels had increased to 69.3% and the percentage of Mexican Americans with optimal control had increased to only 56.5%, significantly less than non-Hispanic Whites.
  • In 2001-2004, 63.6% of non-Hispanic Whites had their blood pressure under optimal control, which was significantly higher than the percentage of non-Hispanic Blacks (47.8%). However, in 2005-2008, the percentage of non-Hispanic Whites who had their blood pressure under optimal control had decreased to 57.1% and the percentage of non-Hispanic Blacks with optimal control had increased to 58.2%. There was no statistically significant difference between the two groups.

Also, in the NHQR:

  • In 2005-2008, only 54.1% of adults age 40 and over with diabetes had achieved control of their HbA1c level, 65.2% had control over their cholesterol level, and 58.6% had their blood pressure under control. Although the percentage of adults with controlled HbA1c and blood pressure does not differ markedly from that in the 2001-2004 period, a significant increase in the percentage who had their cholesterol levels under control was observed over time, from 48.5% in 2001-2004 to 65.2% in 2005-2008.
Prevention: Lower Extremity Amputations

People living with diabetes represent more than 60% of nontraumatic lower extremity amputations7 even though amputations can be avoided through proper care on the part of patients and providers. Hospital admissions for lower extremity amputations for patients with diagnosed diabetes reflect poorly controlled diabetes. Better management of diabetes would prevent the need for lower extremity amputations. Differences in rates may also vary across racial and ethnic groups due to differences in prevalence.

 

Figure 2.9. Hospital admissions for lower extremity amputations per 1,000 adult patients with diagnosed diabetes, by race and gender, 1999-2007

Trend line chart, hospital admissions per 1,000 with diagnosed diabetes, by race, 1997 through 2007. Black, 1999-2001, 5.6, 2002-2004, 5.9, 2005-2007, 4.9. White, 1999-2001, 4.4, 2002-2004, 3.1, 2005-2007, 2.4. Trend line chart, hospital admissions per 1,000 with diagnosed diabetes, by gender, 1997 through 2007. Male, 1999-2001, 7.4, 2002-2004, 6.0, 2005-2007, 4.8. Female, 1999-2001, 3.9, 2002-2004, 2.9, 2005-2007, 2.2.

Source: Centers for Disease Control and Prevention, National Center for Health Statistics, National Hospital Discharge Survey and National Health Interview Survey.
Denominator: Civilian noninstitutionalized population 18 years and over who report they have ever been told they have diabetes.
Note: Data are age adjusted to the 2000 standard population using three age groups: 0-64, 65-74, and 75 and over.

  • From 1999-2001 to 2005-2007, Whites, males, and females all had significant decreases in the hospitalization rate for lower extremity amputation (Figure 2.9).
  • In 2002-2004 and 2005-2007, Blacks had significantly higher rates of hospitalization for lower extremity amputation compared with Whites.
  • In all years, males had significantly higher rates of admission, about twice the rate of females.

Also, in the NHQR:

  • In all years, adults ages 18-44 had significantly lower rates of hospital admission for lower extremity amputation than the overall population and adults ages 45 and over. The rate of admission for adults age 65 and over was twice the rate of adults ages 18-44 in all years.
Indian Health Service Facilities

Nationwide, many AI/ANs who are members of a federally recognized Tribe rely on the Indian Health Service (IHS) to provide access to health care in the counties on or near reservations.11,12, x Due to low numbers and lack of data, information about AI/AN hospitalizations is difficult to obtain in most Federal and State hospital utilization data sources. The NHQR and NHDR address this gap by examining utilization data from IHS, Tribal, and contract hospitals.

Diabetes is one of the leading causes of morbidity and mortality among AI/AN populations. Its prevention and control are a major focus of the IHS Director's Chronic Disease Initiative and the IHS Health Promotion/Disease Prevention Initiative. Addressing barriers to health care is a large part of the overall IHS goal of ensuring that comprehensive, culturally acceptable personal and public health services are available and accessible to AI/ANs.

 

Figure 2.10. Hospital admissions for uncontrolled diabetes per 100,000 population age 18 and over in IHS, Tribal, and contract hospitals, 2004-2007 (left), and community hospitals (right), by race and ethnicity, 2004-2007

Trend line per 100,000 in IHS, Tribal and contract hospitals, by race/ethnicity, 2004 through 2007. AI/AN, 2004, 31.4, 2005, 29.3, 2006, 26.3, 2007, 23.8.          Trend line per 100,000 in community hospitals (HCUP SID), by race/ethnicity, 2003 through 2006. Total, 2004, 22, 2005, 22.1, 2006, 21, 2007, 20.3. White, 2004, 13.6, 2005, 12.9, 2006, 13.1, 2007, 12.2. Black, 2004, 67.5, 2005, 70.7, 2006, 63.2, 2007, 65.7. API, 2004, 9.4, 2005, 10.8, 2006, 8, 2007, 7.2. Hispanic, 2004, 48.2, 2005, 51, 2006, 40.7, 2007, 37.8.

Key: AI/AN = American Indian or Alaska Native; API = Asian or Pacific Islander.
Source: Indian Health Service (IHS), Tribal, and contract hospitals: IHS, Office of Information Technology/National Patient Information Reporting System, National Data Warehouse, Workload and Population Data Mart; community hospitals: Agency for Healthcare Research and Quality, Healthcare Cost and Utilization Project (HCUP) State Inpatient Databases (SID) disparities analysis file, 2004-2007.
Note: White, Black, and API are non-Hispanic populations. Data are adjusted for age and gender. The HCUP SID disparities analysis file is designed to provide national estimates using weighted records from a sample of hospitals from 25 States that have 66% of the U.S. resident population.

  • From 2004 to 2007, the age-adjusted rate of total hospitalizations for uncontrolled diabetes significantly decreased for AI/AN patients in IHS, Tribal, and contract hospitals (from 31.4 per 100,000 to 23.8 per 100,000; Figure 2.10).
  • For all years, White patients in community hospitals had significantly fewer admissions for uncontrolled diabetes compared with AI/AN patients in IHS hospitals. Compared with AI/AN patients in IHS hospitals, Blacks had significantly more admissions in all years and Hispanics had significantly more admissions in 3 of 4 years.

i Throughout this report, total cost equals cost of medical care (direct cost) and economic costs of morbidity and mortality (indirect cost).
ii Direct costs are defined as "personal health care expenditures for hospital and nursing home care, drugs, home care, and physician and other professional services."
iii Cost-effectiveness is measured here by the average net cost of each quality-adjusted life year (QALY) that is saved by the provision of a particular health intervention. QALYs are a measure of survival adjusted for its value: 1 year in perfect health is equal to 1.0 QALY, while a year in poor health would be something less than 1.0. A lower cost per QALY saved indicates a greater degree of cost-effectiveness.
iv The top 5 States contributing to the achievable benchmark are Delaware, Maine, Maryland, Massachusetts, and New Hampshire.
v HbA1c or glycosylated hemoglobin, is a measure of average levels of glucose in the blood.
vi More information on the PQIs is available at: http://www.qualityindicators.ahrq.gov/downloads/pqi/word/pqi_guide_v31.doc.
vii The top 4 States contributing to the achievable benchmark are Alaska, New Hampshire, Vermont, and the District of Columbia.
viii The top 4 States contributing to the achievable benchmark are Hawaii, Nebraska, Utah, and Vermont.
ix Blood pressure control guidelines were updated in 2005. Previously, having a blood pressure reading of <140/90 mm Hg was considered under control. For this measure, the new threshold of <140/80 mm Hg has been applied to historic data for the sake of consistency and comparability.
x Of potentially eligible AI/ANs, 74% sought health care in 2004 at an IHS or tribally contracted facility, according to IHS estimates published by the Office of Public Health Support, Division of Program Statistics.
 



Proceed to Next Section

Page last reviewed October 2014
Internet Citation: Chapter 2. Effectiveness: National Healthcare Disparities Report, 2010. October 2014. Agency for Healthcare Research and Quality, Rockville, MD. http://archive.ahrq.gov/research/findings/nhqrdr/nhdr10/Chap2.html