What Is the Current State of Quality of Care in Diabetes?

According to the National Diabetes Information Clearinghouse (a service of the National Institutes of Health), 13 million people in the United States had been diagnosed with diabetes in 2002. It is estimated that over 5 million remain undiagnosed, bringing the total number of people with diabetes to over 18 million, or 6.3% of the population.^[1]

The Rising Trend

During the past decade, we have seen a 33% increase in the incidence and prevalence of diabetes in the United States,^[2] and this trend is reflected worldwide.^[3] The lifetime risk of being diagnosed with diabetes in the United States is one in three for men, and two in five for women. About one in three Americans born in 2000 will develop diabetes; Hispanic females are at especially high risk, with a 50/50 chance of developing diabetes during their lifetimes.^[4] Of particular concern is the increase of type 2 diabetes in children and younger adults. Historically, type 2 diabetes mainly occurred in those over 45 years of age.^[5] However, in the United States, 8% to 45% (depending on geographic area) of children newly diagnosed with diabetes have type 2. These cases occur mainly in African American, Mexican American, Native American, and Asian American children and young adults.^[4]

The rise in diabetes has been linked to the rise of obesity and overweight; therefore, addressing overweight and obesity could help control this “new epidemic.”^[2,6,7] Studies have shown that even modest weight loss and increased physical activity (i.e., 7% weight reduction and walking for two and a half hours each week) can halve the incidence of diabetes.^[7,8]

The Ethnic Disparity

Hispanic/Latin Americans, African Americans, and Native Americans are more likely to be diagnosed with diabetes than non-Hispanic whites (Figure 1). Asian Americans are also at higher risk, although the U.S. data are limited to Native Hawaiians and Japanese and Filipinos who live in Hawaii.^[1]

Figure 1. Age-Adjusted Prevalence of Diabetes in U.S. Adults by Ethnicity, 2002[1]

Examples of the relationship of diabetes with racial and ethnic minority populations include:

• One third of African Americans with diabetes do not know they have it.^[11]


• About 24% of Mexican Americans, 26% of Puerto Ricans, and 16% of Cuban Americans between the ages of 45 and 74 years have diabetes.^[12]


• Complications from diabetes are major causes of death and health problems in most Native American populations; in one tribe, about half of the adults between 30 and 64 years of age have diabetes.^[13]


• The prevalence of diabetes among African American, Hispanic/Latin American, Native American, and Asian/Pacific Islander women is at least two to four times higher than among white women.^[14]

NCQA’s HEDIS^® measures assess the quality of diabetes management in the following six areas, under the overall measure of Comprehensive Diabetes Care^[15]:

• Testing Blood Glucose Level (hemoglobin A1c [HbA1c] test)
• Controlling HbA1c Level
• Screening for Serum Cholesterol Level (LDL-C Screening)
• Controlling Serum Cholesterol Level
• Examining Eyes for Retinal Disease
• Monitoring for Kidney Disease

Two additional HEDIS^® measures are also critical to the care of persons with diabetes:

• Controlling High Blood Pressure
• Medical Assistance With Smoking Cessation

In 2003, the rates for four of the six diabetes management components improved in commercial health care organizations. However, health plans scored better than 80% in only two of the six components: HbA1c testing and cholesterol screening. The rates for controlling high blood pressure and advising smokers to quit also improved.^[15]The HEDIS^® data demonstrate that although health plans are making progress, the need to pursue better management of diabetes and its complications continues.

Testing for and control of blood glucose and low-density lipoprotein (LDL) cholesterol levels improved in 2003. In contrast, the rate of eye exams and testing for kidney disease declined, but the reductions were likely due to changes in the specifications for those measures.^[15]

HbA1c Testing

The HbA1c testing measure (now called A1c test) determines if testing for this critical information occurred one or more times during the measurement year. The proportion of commercial plans’ members who received at least one A1c test during the year continued to rise, to 84.6% (Figure 2).^[15]

Figure 2. Rates of Testing Blood Glucose Levels (Commercial Plan) This component of the Comprehensive Diabetes Care measure estimates the percentage of health plan members with type 1 or type 2 diabetes who are aged 18 to 75 years and have had one or more A1c tests during the year.

A1c Control

Blood sugar control is critical to reducing the development and progression of diabetes microvascular complications. Studies have shown that reducing A1c levels by just 1% can reduce the risk of developing eye, kidney, and nerve disease by 40%.^[1] Some analyses suggest that an A1c of less than 7% may reduce the risk of complications further; individual health care practitioners may want to consider a lesser goal for certain patients at risk for hypoglycemia. ^[6]

In this HEDIS^® component, which measures the rates of patients with poor glycemic control (A1c greater than 9%), lower percentages are better. Health plans improved slightly from 2002 (Figure 3), even though the specifications became more stringent by defining poor control as an A1c level of 9% or greater, rather than the 9.5% used earlier.^[15] NCQA and the National Alliance for Diabetes Quality Improvement are working together to develop a measure of “good control” in diabetes for future use.

Figure 3. Rates of Poor Control of Blood Glucose Level (Commercial Plans) This component of the Comprehensive Diabetes Management measure estimates the percentage of health plan members with type 1 or type 2 diabetes who are aged 18 to 75 years and have had an A1c level >9% during the year.

Meeting the ADA definition of glycemic control (7% or less) is likely to be an ongoing challenge. The 1999-2000 National Health and Nutrition Examination Survey (NHANES) assessed performance in the general population using the ADA’s more stringent definition, and only 37% of diabetic patients met the goal of an A1c of 7% or less. ^[16]

LDL-C Screening

Given the known effect of elevated cholesterol on development of heart disease in general, and with diabetes in particular, measuring lipid levels is a critical step to managing cardiovascular disease. For the third consecutive year, health plans continued to improve their LDL-C screening rate to 88.4% (Figure 4). ^[15]

Figure 4. Rates of Screening for Serum Cholesterol (Commercial Plans) This component of the Comprehensive Diabetes Management measure estimates the percentage of health plan members with type 1 or type 2 diabetes who are aged 18 to 75 years and have had an LDL-C screening during the year.

LDL-C Control

Controlling lipids can reduce cardiovascular complications by 20% to 50%.^[1] HEDIS^® now assesses two rates for the LDL-C level in diabetes: below 130 mg/dL and, to align with changes recommended by the National Cholesterol Education Program (NCEP) guidelines, below 100 mg/dL. Reducing the rate for LDL-C control to <130 mg/dL improved several percentage points from 2002 (Figure 5). The assessment for an LDL-C of <100 mg/dL was new for 2003, so there is no comparative value. Commercial plans reported a rate of 34.7% control here, leaving plenty of room for improvement against the newer NCEP guidelines.^[15]

Figure 5. Rates of Controlling Cholesterol Levels This component of the Comprehensive Diabetes Management measure estimates the percentage of health plan members with type 1 or type 2 diabetes who are aged 18 to 75 years and have had an LDL-C <130 mg/dL during the year.

Eye Examinations

Detecting and treating diabetic retinopathy with laser therapy can prevent severe vision loss in 50% to 60% of patients.^[1] The rates of eye examinations fell 2.9 percentage points (Figure 6), likely due to changes in the measure specifications that required more frequent screenings for certain patients.^[15] The retinal eye exam rate has remained about the same for the last four years; health plans and providers will need to apply innovative strategies to increase it further.

Figure 6. Rates of Eye Examinations (Commercial Plans) This component of the Comprehensive Diabetes Management measure estimates the percentage of health plan members with type 1 or type 2 diabetes who are aged 18 to 75 years and have had an eye exam during the year.

Monitoring for Kidney Disease

Detecting early diabetic kidney disease and treating it by working with patients to lower blood pressure can reduce the decline in kidney function by 30% to 70%.^[1] HEDIS^® scores for health plans in the area of screening for kidney disease fell 3.6 percentage points (Figure 7), likely due to changes in the measure specifications that required more frequent screenings for certain patients.^[15] Similar to the retinal eye exam rate, the diabetic nephropathy screening rate has also remained about the same for the last four years, calling for particular focus on improvement.^[15]

Figure 7. Rates of Screening for Diabetic Nephropathy (Commercial Rates) This component of the Comprehensive Diabetes Management measure estimates the percentage of health plan members with type 1 or type 2 diabetes who are aged 18 to 75 years and were screened for kidney disease during the year.

Rates for Other Measures

Controlling High Blood Pressure Measure

Hypertension is common in people with diabetes and is itself a major risk factor for cardiovascular disease (CVD) and microvascular complications.^[6] In general, each reduction of systolic blood pressure by 10 mm Hg can reduce the risk for any complication related to diabetes by 12%.^[1]

This HEDIS^® measure tracks adults aged 46 to 85 years, with or without diabetes, who have been diagnosed with hypertension.15 Even though it includes persons without diabetes, improvements in this measure can have a positive effect on the 73% of persons with diabetes who also have hypertension.^[7] The proportion of members whose blood pressure was controlled increased, continuing a trend for this measure. However, more improvement is needed, as only 62.2% of adults were controlled (Figure 8).^[15] Moreover, 140/90 mm Hg is not optimal blood pressure control for patients with diabetes. Both the NCEP and ADA guidelines specify that blood pressure in patients with diabetes should be controlled at 130/80 mm Hg.

Medical Assistance With Smoking Cessation Measure

Figure 8. Rates of High Blood Pressure Control (Commercial Rates) This component of the Comprehensive Diabetes Management measure estimates the percentage of health plan members with type 1 or type 2 diabetes who are aged 46 to 85 years and with diagnosed hypertension, whose blood pressure was 140/90 mm Hg or lower.

Several studies have shown that smoking increases the risk for developing type 2 diabetes.17 Moreover, people who have diabetes and smoke are more likely to die prematurely from heart disease and stroke than people who have diabetes and do not smoke. They may also experience microvascular complications earlier than nonsmokers.^[6]

The HEDIS^® Medical Assistance With Smoking Cessation measure queries all members who receive the Consumer Assessment of Health Plans (CAHPS^®) survey about whether their health care practitioner encouraged them to quit smoking and discussed strategies for quitting. Although this measure includes both patients with and without diabetes, improvements in this measure may help decrease the risk for developing type 2 diabetes. Of current smokers and recent quitters, 68.6% were advised to quit by their health care practitioners, a slight increase over previous years (Figure 9). However, only 37.6% discussed cessation medication with their health care practitioner, and only 36% discussed cessation strategies.

Figure 9. Rates of Advising Smokers to Quit (Commercial Plans) This measure estimates the percentage of smokers or recent quitters who were advised to quite by their physician during the year. *First year in calculating "rolling average," therefore no published rates.

The following summarizes one health plan’s QIA that addresses the LDL-C Screening and Control HEDIS^® measurements, as well as the rates for eye examinations.

Quality Improvement Activity: Diverse Approaches for Improving Diabetes Care

The care of diabetes is challenging due to its complexity and the fact that patients often present with other diagnoses or health concerns, including obesity, smoking, hypertension, and hyperlipidemia. The Centers for Disease Control and Prevention (CDC) has stated that the diabetes community has three challenges in response to the growing health burden: prevent diabetes, cure diabetes, and improve the quality of care of people with diabetes to avoid devastating complications.18 Managed care organizations (MCOs) are in a position to work on two of these goals: prevention of diabetes and reduction of complications

A health plan took on these goals in 1999 by implementing a registry and identifying over 10,000 members with diabetes (4.5% of their member population). The organization’s goal was to develop an approach to diabetes management that would stress patient empowerment (including self-management skills and knowledge of preventive services) to reduce complications and maintain quality of life.

In 2000, the health plan utilized HEDIS^® measures to assess the care of members with diabetes in both its commercial and Medicare products. The health plan focused on the results that presented the greatest opportunity for improvement as compared with national benchmarks. These measurements would also serve as an indicator for assessing the effectiveness of prevention efforts aimed at reducing complications. Table 1 summarizes the baseline results.

A Diabetes Task Force convened to review these results. This task force consisted of a medical director, QI staff, diabetes nurse case managers, and a research analyst. In addition, staff members from customer service, marketing, and provider relations were involved. This diverse team approach enabled the health plan to draw on input from staff involved at all levels of medical management and service for members. Clinical input was also provided through a committee that included five practitioners with varying specialties, all with knowledge or expertise in diabetes.

The Diabetes Task Force assessed barriers based on a review of literature and past experience with diabetes activities. In 1999, the health plan had modified the eye care benefit to allow members with diabetes easier access to dilated retinal examinations. The task force questioned whether this intervention impacted results and what other barriers to eye examinations existed from the members’ perspectives. The committee decided that a survey would help the organization ascertain whether members with diabetes were knowledgeable about their eye care benefits and could also help identify other issues preventing members from receiving services. A random sample of 1,000 commercial and 1,000 Medicare members with diabetes who did not receive an eye examination during the baseline measurement period were selected to be surveyed by telephone (see Appendix 1).

The barrier survey identified the top reasons that members did not have an eye exam were related to:

• Cost
• Difficulty scheduling appointments
• Dislike of the procedure

The task force concluded that the barriers to eye care they could influence fell into two categories: administrative and financial. Since the health plan had already waived the co- payment for eye examinations the prior year, the task force believed the root cause to be a lack of knowledge about the benefit change. Additional barriers to diabetes care identified by the team included:

• Member lack of self-management skills
• Member lack of disease knowledge, including secondary prevention of diabetes
• Physician practice variation

The health plan began enhancements in earnest in 2000 to improve diabetes care. To address the identified barriers, it implemented a series of strategies aimed at soliciting member involvement in education and enhancing physician practice patterns.

The interventions targeted to members in the diabetes registry included the following:

• Physicians were given member profiles based on information tracked through the registry. The profiles included A1c and LDL-C testing and results, as well as evidence of eye exams.
• Multiple educational modules were mailed throughout the year to 4,653 commercial and 3,447 Medicare members with diabetes. Modules included information on testing for lipids, A1c, and retinal eye examinations, along with information about diabetes complications. The material focused on lifestyle information to help prevent complications and the importance of regular screening by practitioners.
• Telephone assessment calls were made to 1,900 members identified with moderate disease risk based on stratification methodology. These outreach calls were made by diabetes nurses trained to educate and make necessary referrals based on identified member needs.
• A diabetic eye benefit packet, including instructions on co-payment waiver, self-referral, and a listing of all participating eye care professionals, was mailed to all members with diabetes.
• A diabetic eye benefit packet mailing was sent to 900 primary care physicians (PCPs) and all eye care specialists reminding them of the upgraded benefit and ease of arranging for eye examinations.

The first remeasurement was reported in 2001 based on data from January 1, 2000, through December 31, 2000. Statistically significant gains were achieved in every measure for both the commercial and Medicare products, except for the commercial rate for lipid screening (Table 2).

The Diabetes Task Force observed that 80% to 85% of members had their lipid screening, yet a large percentage of members with diabetes did not have LDL-C control. This fact was very important in light of the findings in the 2001 Adult Treatment Panel (ATP) III guidelines that diabetes is considered a coronary heart disease (CHD) risk equivalent, meaning that a person with diabetes has the same risk of having a cardiovascular (CV) event as someone with CHD. As a result, the task force embarked on an initiative to analyze the LDL-C levels of the entire membership with diabetes.

It decided to target members with LDL-C levels beyond the 130 mg/dL control level, since these members were at increased risk for heart disease.

The task force was pleased with the improvements achieved for retinal eye examinations and credited this to the education of members and practitioners about the benefit enhancements and importance of this service. The task force decided to continue existing efforts to improve eye examination rates.

Several new strategies based on initial results and the continuing barriers to care were developed. New interventions were:

• A continuing medical education (CME) monograph on diabetes as a disease of beta cell dysfunction was sent to the top 140 primary care groups on the basis of the volume of members with diabetes. This education was designed to provide a better understanding of the disease process and the opportunities to prevent complications. CME credits were available as an incentive to study the material.
• Case management was restructured from a reactive model (identify and assist members once they demonstrate complications from diabetes) to a proactive model that identified members at risk of exacerbation of their diabetes. The case managers focused on working with these members on glycemic control, lipid management, and interventions to promote self-management. This intervention was implemented in direct response to members with poor LDL-C control rates.
• Telephone assessment calls were made to over 3,400 members with moderate disease risk. These outreach calls were made by diabetes nurses trained to educate and make necessary referrals based on identified member needs.
• Revised hypercholesterolemia guidelines based on the 2001 ATP III guidelines were distributed to practitioners. Information was published in the form of a quick desk reference for ease of use.
• Test results were tracked down, and practitioners and members were educated about this issue

Performance indicators for lipid screening and control improved in 2002 based on data from January 1, 2001, through December 31, 2001. The eye examination rates for members in both products dropped slightly (Table 3).

The Diabetes Task Force was pleased with improvements that were achieved for lipid control. It credited the improvements to efforts used to track down all test results and educate both practitioners and members about this issue. The task force members believed that by tracking down every result for each member with diabetes, physicians were prompted to intervene with members in need of treatment. At the same time, they were disappointed in the eye examination rate and felt the prior interventions had run their course and new efforts were in order.

Using Surveys to Perform Root Cause Analysis Data derived from administrative systems provides only a result. It explains neither why services were not provided, nor the cause of poor outcomes—information that is needed for adequate barrier analysis. In this case, the health plan decided to seek additional data for root cause analysis. The organization utilized member surveys to obtain information about barriers to care. These surveys were conducted by nurses, allowed for clinical interpretation of results, and provided an opportunity to educate members about diabetes care while the nurses had the members’ attention. Survey questions should be designed to capture information that will be helpful in identifying opportunities to improve services and allow for comparison from one period to the next. A comparative analysis permits identification of barriers and affords the ability to assess the strength of interventions previously implemented. Appendix 3 contains a survey implemented by this health plan to assess diabetes care and identify barriers.

To better understand barriers to eye care, case managers telephoned the 306 members identified as not receiving a dilated retinal examination. A scripted survey was used to identify barriers. With a response rate of 39%, the health plan identified two reasons for not getting eye examinations: 1) the member forgot to schedule an appointment (27%), and 2) the member did not think it was necessary (24%). The task force decided to employ more reminders to members with diabetes and education about the risk of blindness in patients with diabetes.

The task force also studied the demographics of the plan’s population with diabetes and determined that it would make a special effort to reach the pediatric population; education and prevention efforts to date had been geared more to the adult member.

Interventions discussed previously continued throughout 2002. In addition, the following new activities were put in place:

• An educational module on eye health for patients with diabetes was mailed to 9,250 members and 900 PCPs and eye care specialists. The module discussed risk factors associated with diabetes and its possible impact on eyesight. Reminders about the benefits of eye exams and the need for scheduling appointments were also included.
• An educational module was sent to parents of pediatric members with diabetes. This material was designed for children who were of the age to be dealing with diabetes in school. The focus was tracking compliance and communicating with school personnel.
• The case management program was enhanced by offering ongoing support to members who were no longer in need of intensive case management services. In the past, members with diabetes were assisted by case managers until their laboratory levels were in control and any complications were stabilized. However, beginning in January 2002, a licensed practical nurse continued to contact these members once a month to assess how well the member was doing with self-management and to determine if any exacerbations of the disease had occurred. The goal was to keep the patient with diabetes stable and prevent any further complications.
• The member Web site was enhanced to provide interactive tools for diabetic self-assessment and information about community resources, support groups, and social service agencies. Information about case management services offered by the health plan was also provided, along with links to other diabetes educational sites.
• The risk stratification process was revised to capture members with elevated LDL-C values for targeted interventions.
• Birthday card reminders were initiated to remind members to obtain annual eye examinations and A1c and lipid testing.
• Hyperlipidemia, hypertension, and nephropathy enhancements were added to assessments, treatment plans, and educational materials.
• Annual Diabetes Report Card initiated to tell members if the plan had information indicating that the following annual ADA recommended preventive services were completed: A1c, LDL-C, microalbumin, and eye exam. Members were encouraged to receive the missing services (see Appendix 2).
• Educational module on dyslipidemia was sent to 900 commercial and 627 Medicare members with an LDL-C value >130 mg/dL.

The health plan reviewed interventions and determined the most successful were those directed at members with established health care needs. By using the registry, the organization was able to pinpoint members who were missing tests and services. The plan then reached out to these members and their physicians with targeted education and case management to encourage treatment adherence.

The health plan has continued its efforts to improve diabetes care. Results reported in 2004 demonstrated improvements for lipid screening (91.7% commercial and 96.4% Medicare) and lipid control (LDL-C <130 mg/dL) for Medicare (84.9%). Other measures have maintained results achieved at the conclusion of this study.

The organization featured in this case study decided to focus on prevention of complications for heart disease and retinopathy. Its approach relied on drilling down through data to uncover services that were not rendered and/or had poor results. The health plan pursued medical records and conducted member surveys to identify additional data and barriers. From there, it provided intensive outreach and education to members and their physicians to encourage adherence. This health plan not only improved its HEDIS^® rates for select measurements, but improved the future health of its members by addressing prevention of complications.

1. National Institute of Diabetes and Digestive and Kidney Diseases. National diabetes statistics. Available at: http://diabetes.niddk.nih.gov/dm/pubs/statistics/index.htm. Accessed March 14, 2005.

2. Kaufman FR. Type 2 diabetes in children and young adults: a “new epidemic.” Clin Diabetes. 2002;20:217-218.

3. Alberti G, Zimmet P, Shaw J, Bloomgarden Z, Kaufman F, Silink M, for the Consensus Workshop Group. Type 2 diabetes in the young: the evolving epidemic. The International Diabetes Federation consensus workshop. Diabetes Care. 2004;27:1798-1811.

4. Ven Kat, Narayan KM, Boyle JP, Thompson TJ, Sorensen SW, Williamson DF. Lifetime risk for diabetes mellitus in the United States. JAMA. 2003;290:1884-1890.

5. American Diabetes Association. Diabetes and cardiovascular (heart) disease. Available at: http://www.diabetes.org/utils/printthispage.jsp?PageID=STATISTICS_233190. Accessed March 14, 2005.

6. American Diabetes Association. Standards of medical care in diabetes. Diabetes Care. 2005;28(suppl 1):S4-S36.

7. American Diabetes Association. National diabetes fact sheet. Available at: http://www.diabetes.org/utils/printthispage.jsp?PageID=STATISTICS_233193. Accessed March 14, 2005.

8. American Diabetes Association. Become a corporate health ambassador. Available at: http://www.diabetes.org/utils/printthispage.jsp?PageID=SUPPORTCORPFRIENDS_270940. Accessed March 14, 2005.

9. US Census Bureau Public Information Office. Baby boom brought biggest increases among people 45 to 54 years old [press release]. Available at: http://www.census.gov/Press-Release/www/2001/cb01cn184.html. Accessed March 14, 2005.

10. The multicultural report-brief article. American Demographics. November 1, 2001.

11. American Diabetes Association. Diabetes statistics for African-Americans. Available at: http://www.diabetes.org/utils/printthispage.jsp?PageID=STATISTICS_233186. Accessed March 14, 2005.

12. American Diabetes Association. Diabetes statistics for Latinos. Available at: http://www.diabetes.org/utils/printthispage.jsp?PageID=STATISTICS_233187. Accessed March 14, 2005

13. American Diabetes Association. Diabetes statistics for Native Americans. Available at: http://www.diabetes.org/utils/printthispage.jsp?PageID=STATISTICS_233188. Accessed March 14, 2005

14. American Diabetes Association. Diabetes statistics for women. Available at: http://www.diabetes.org/utils/printthispage.jsp?PageID=STATISTICS_233183. Accessed March 14, 2005

15. National Committee for Quality Assurance. The State of Health Care Quality 2004. Washington, DC: National Committee for Quality Assurance; 2004.

16. Saydah SH, Fradkin J, Cowie CC. Poor control of risk factors for vascular disease among adults with previously diagnosed diabetes. JAMA. 2004;291:335-342.

17. Eliasson B. Cigarette smoking and diabetes. Prog Cardiovasc Dis. 2003;45:405-413.

18. Centers for Disease Control and Prevention. Frequently asked questions. Available at: http://www.cdc.gov/diabetes/faq/basics.htm. Accessed March 17, 2005.