Improving Care for Diabetes Patients Through Intensive Therapy and a Team Approach

Research in Action, Issue 2

Diabetes is the sixth leading cause of death in the United States and proper treatment can be complex. This synthesis provides good evidence to show that intensive therapy using a team approach is an effective way to reach the major goals of diabetes therapy: lowering glucose (blood sugar) to appropriate levels and avoiding or postponing the onset of serious complications.


Biomedical Research Shows Positive Outcomes Under Ideal Conditions
Health Services Research Studies Show Varied Success in Controlling Blood Sugar
Providers and Patients Experience Barriers to Meeting Treatment Goals
Combination of Intensive Therapy and Team Approach Promotes Good Outcomes
Preventive Care Can Help To Postpone or Avoid Complications
Provider Questionnaires Improve Adherence to Treatment Protocols
Barriers to Patient Adherence Can Be Reduced
Research Is Currently Underway to Improve Diabetes Care
For More Information


Diabetes, the sixth leading cause of death in the United States, is a chronic disease characterized by persistent hyperglycemia (high blood glucose levels). Left untreated, diabetes can cause serious complications affecting the circulatory and nervous systems, kidneys, eyes, and feet.

Even though the proper treatment of diabetes can be complex, biomedical research has demonstrated in model settings what can be achieved clinically. Therefore, health services research funded by the Agency for Healthcare Research and Quality (AHRQ) has built on such results to learn more about what can be achieved when treating diabetic patients in a typical office practice. However, there are limits beyond which the data from biomedical trials cannot be extrapolated to the typical office practice because of differences between sample populations selected for such trials and the general patient population.

This synthesis provides good evidence to show that intensive therapy using a team approach is an effective way to reach the major goals of diabetes therapy: lowering glucose (blood sugar) to appropriate levels and avoiding or postponing the onset of serious complications.

The following discussion is intended to provide health care professionals with information they can use in their practices to assist them in providing better care and helping their patients to become better self-managers.

Return to Contents


An increasing number of Americans have diabetes—approximately 16 million as of 1999—and diabetes costs society close to $100 billion per year in medical and nonmedical costs. In 2001, almost 800,000 people will be diagnosed with the disease, and close to 200,000 will die from its complications. Furthermore, the burden of diabetes is considerably higher for the elderly and minorities than the general population.1

In a healthy person, blood sugar levels, which fluctuate based on food intake, exercise, and other factors, are kept within an acceptable range by insulin. Insulin, a hormone produced by the pancreas, helps the body absorb excess sugar from the bloodstream. In a person with diabetes, blood sugar levels are not adequately controlled by insulin.

Physicians monitor glycemic levels by using the hemoglobin (Hb) A1c test, which shows the average amount of sugar in the blood over the preceding 3 months. In healthy persons, glycemic values as measured by HbA1c levels are in the 4-6 percent range, according to the American Diabetes Association (ADA).a The ADA guideline for diabetic patients is <7 percent.2

There are two major types of diabetes:b

  • Type 1 is an autoimmune disease in which the ability of the pancreas to make insulin has been destroyed. Type 1 diabetes usually develops in children or adults under age 30, but it can occur in older individuals.
  • Type 2 diabetes is a disease in which the pancreas produces some insulin, sometimes even large amounts; however, either the pancreas does not produce enough insulin or the body's cells are resistant to the action of insulin. Between 90 and 95 percent of diabetes patients have type 2 diabetes, which mostly affects adults over age 40. The incidence of this type of diabetes is rising rapidly; increasingly type 2 diabetes is appearing in patients in their 30s and younger.3,4

a. The percentages for normal values may vary somewhat, depending on the authority being cited.
b. There are also two other types of diabetes: "gestational diabetes," which affects pregnant women, and "other specific types," resulting from specific genetic syndromes, surgery, drugs, malnutrition, infections, and other illnesses.

Providers say that diabetes is a difficult disease to treat because:

  • It is closely related to patient behaviors that require complex interventions and persistent efforts to change (sedentary lifestyle, poor diet, and obesity).
  • Patients may have the disease for years without knowing it, and even after diagnosis may remain asymptomatic, with the result that they are less motivated to make needed lifestyle changes.
  • Patients need to actively manage the disease through diet, exercise, self-monitoring, and medicating, even in the absence of symptoms.
  • Providers may need to coordinate interventions with different specialists.

Providers evaluate a patient's blood sugar levels both by glucose measurements and by the HbA1c test. In contrast, the patient's self-monitoring of blood glucose shows the level of blood sugar only at the time of measurement.


  • A study of over 8,000 patients with type 2 diabetes found that, 2 years after the initiation of insulin, more than 60 percent of the patients still had blood sugar levels of over 8 percent (HbA1c).
  • A survey of providers found that there are many barriers to achieving treatment goals, including the frequently asymptomatic character of diabetes, the involvement of many body systems, and difficulties in altering lifestyle.
  • Another study found that, after 12 months, 87 percent of patients (80 percent with type 2 diabetes) achieved good control of blood sugar (8 percent or less HbA1c) by the use of complex treatment regimens and a team approach, with many patients receiving either two oral hypoglycemics or one oral hypoglycemic plus insulin injections.
  • Patients can achieve good diabetic control if providers recommend intensive therapies, use a team approach, furnish appropriate preventive care, and put into practice proven strategies that help patients better manage their care.

Important Facts About Diabetes

Persons Affected:

  • 15.7 million people (5.9 percent of the U.S. population).
  • 6.3 million (18.4 percent ) of the elderly (65 and over).
  • 8.2 percent of people 20 and over.
  • 10.8 percent of non-Hispanic blacks, 10.6 percent of Mexican-Americans, and 9.0 percent of American Indians.
  • From 1990 to 1998, diabetes increased by 70 percent for people ages 30 to 39, by 40 percent for people ages 40 to 49, and by 31 percent for people ages 50 to 59.

Human Costs:

  • 798,000 new cases diagnosed per year.
  • 193,000 deaths (1996).
  • 12,000 to 24,000 new cases of blindness each year due to diabetic retinopathy.
  • 27,000 cases of end stage renal disease (1995).
  • 67,000 amputations per year (1993-95).

Financial Costs:

  • $98.2 billion per year:
    • $44.1 billion in medical costs.
    • $54.1 billion in indirect costs (work loss, disability, and premature death).

Sources: National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Disorders. Diabetes Statistics of the United States, Nov. 1, 1998; Centers for Disease Control and Prevention Web site:; American Diabetes Association. Direct and Indirect Costs of Diabetes:; Mokdad AH, Ford ES, Bowman BA, et al. Diabetes trends in the U.S.: 1990-1998, Diabetes Care 2001; 23(9):1278-83.

Return to Contents

Biomedical Research Shows Positive Outcomes Under Ideal Conditions

Until the last decade, there was some controversy in the medical community regarding the relationship between careful control of blood sugar and delaying or preventing complications such as eye and kidney disease and vascular complications for people with diabetes. However, the United Kingdom Prospective Diabetes Study (UKPDS) for type 2 diabetes and, in the United States, the Diabetes Control and Complications Trial (DCCT) for type 1 diabetes have demonstrated that any reduction in blood sugar (as measured by HbA1c) is likely to reduce the risk of complications.5,6 These long-term trials, conducted under closely supervised clinical conditions and using newly diagnosed volunteer subjects, found that intensive therapy could achieve effective glycemic control and postponement of major complications over a period of many years.

In the UKPDS, 5,000 newly diagnosed type 2 diabetes patients first attended several sessions of dietary education given by physicians and dietitians and were treated with diet therapy for 3 months before receiving other treatment.7,8 They were then randomly assigned to four groups. Each of the three "intensive therapy" groups received a single medication (chlorpropamide, glyburide, or insulin); the fourth group, a control group, received a special diet only. However, the single medication therapy in the intensive therapy groups failed to achieve tight glycemic control, making it necessary to institute combination drug therapy (using insulin or metformin along with one of the other two drugs).

Median HbA1c levels over a 10-year period were 7.0 percent for the intensive therapy groups compared with 7.9 percent for those receiving conventional treatment. Additionally, complication rates were significantly lower for the intensive therapy groups compared to the conventional therapy group—33 percent lower for end stage renal disease and 17 percent lower for retinopathy.

In the DCCT, 1,441 type 1 diabetes patients were assigned to one of two groups.6 The first group received intensive insulin therapy, either by external insulin pump or by receiving three or more daily insulin injections. The second (control) group received conventional therapy (one or two daily insulin injections). Patients in the first group monitored their blood sugar four times each day and, when necessary, made daily adjustments to their dosage levels. As part of their therapy, they made monthly visits to a health care team composed of a physician, nurse educator, dietitian, and behavioral therapist. Each patient was given a diet and exercise plan and was contacted weekly by a member of the team to review and adjust the treatment regimen. Patients in the control group did daily self-monitoring of urine or blood but did not usually make daily adjustments to their dosage levels. These patients also received education about diet and exercise but were seen only every 3 months. Over a period of 6.5 years, the intensive therapy group had average HbA1c levels (at 7.2 percent) and rates of diabetic complications significantly lower than the control group.c

c. The mean value for all glucose profiles was 155 + 30 mg per deciliter in the intensive therapy group vs. 231 + 55 mg per deciliter in the control group. For the intensive therapy group, the rate of complications was 70 percent lower for the onset of retinopathy, 54 percent lower for albuminuria, and 60 percent lower for the appearance of neuropathy.

A third study found that patients not only avoid complications, but also feel better, even over the short term, when their glucose levels are closer to normal. In this study by Harvard University researchers, patients received intensive drug therapy over a 12-week period. During the study period, patients were closely monitored at nine different times using clinical and laboratory evaluations. The patients received glipizide (an oral medication) in doses that were adjusted upward during the first part of the study in order to achieve appropriate glycemic control. The Harvard study found that the patients with mild to moderate type 2 diabetes receiving glipizide plus diet modification, when compared to a control group receiving a placebo plus diet modification, had better glycemic control (7.5 percent vs. 9.3 percent HbA1c) and enjoyed substantial short-term symptomatic and quality of life (QOL) benefits.9 Patients showed significant improvement on four of five QOL scales (symptom distress, general perceived health, cognitive functioning, overall visual acuity scale). On the fifth scale, mental and emotional health, there were no statistically significant changes.

Return to Contents

Health Services Research Studies Show Varied Success in Controlling Blood Sugar

Studies funded by AHRQ have used more broadly representative groups of patients with diabetes than either the DCCT or the UKPDS. The median patient age in the largest AHRQ study was 65, compared to a median age of 53 in the UKPDS; older patients are more likely to have other comorbidities or complications from diabetes. These AHRQ-funded studies have shown that providers and patients may achieve only partial success in bringing blood sugar levels down to recommended levels and performing recommended examinations. For example, three AHRQ-funded studies evaluating diabetes care found that glycemic control was achieved for no more than 50 percent of the patients. They are summarized in Table 1 (Text Version).

The largest study examined what happened to the blood sugar levels of a large group of diabetic patients after primary care physicians started to treat them with insulin.10 Most of these patients had previously been taking oral hypoglycemic medications. After 2 years, more than 60 percent of the patients in the study still had blood sugar levels of over 8 percent (HbA1c).

A second AHRQ-funded study evaluated internal medicine residents' treatment of African-American diabetic patients in a hospital ambulatory care setting.11 This study found that 51 percent of patients on oral medications and 47 percent of patients on insulin did not reach the recommended treatment goal for the study (an HbA1c level of less than 8 percent). It also found that on five tests recommended by the American Diabetes Association (dilated eye exam, lipid exam, home glucose monitoring, foot exam, and urine protein screen), practice levels fell short of national standards.

The third AHRQ-funded study looked at the outcomes of 170 mild-to-moderately ill diabetic patients treated by endocrinologists, family practitioners, and general internists.12 In general, after 2 years of treatment, researchers found no meaningful differences in HbA1c levels or other health outcomes between patients of endocrinologists and patients of family practitioners and general internists. It is unclear whether similar results would occur if severely ill patients were studied.

AHRQ-Funded Research on Diabetes Care

  • Variations in the Management and Outcomes of Diabetes, 1990-97. New England Medical Center. Measured the effectiveness of existing clinical care in controlling blood sugar and avoiding or postponing complications. Showed that help from family members increases elderly diabetics' adherence to treatment and dietary regimens.
  • Center for Medical Treatment Effectiveness Programs, 1997-98. Case Western Reserve University Henry Ford Health System. Developed, evaluated, and implemented culturally appropriate partnerships among patients, health care providers, and the community. Showed that social support increases African-American diabetic patients' adherence to treatment and dietary regimens.
  • Mexican-American Treatment Research Center, 1992-98. University of Texas Health Sciences Center. Studied the effectiveness of existing treatments for chronic disabling conditions in Mexican Americans. Identified the existence of substantial barriers for both providers and patients in implementing accepted treatment guidelines.
  • Improving Primary Care of African-Americans with NIDDM, 1998-2002. Emory University. Evaluates provider support strategies for the management of non-insulin-dependent diabetes mellitus (NIDDM). Identified barriers to provider adherence to management goals. Showed how an intensive approach to therapy (including patients taking more than one medication, a team of providers, and more frequent interactions between patients and clinicians) achieved therapeutic success.
  • Automated Assessments and the Quality of Diabetes Care, 1999-2003. Palo Alto Institute for Research and Education. Evaluates whether innovations in information technology can increase patient/provider interactions and assist patients in self-management. Will examine variation in outcomes for patients with diabetes by using an automated telephone disease-management system to assess patients weekly for 1 year.

Return to Contents

Providers and Patients Experience Barriers to Meeting Treatment Goals

Both providers and patients encounter many barriers that prevent them from achieving established treatment goals. AHRQ has funded studies to understand the barriers that physicians cite for difficulties in achieving therapeutic goals. These barriers, summarized in Table 2, frequently were related to the complexity of the patients' condition and the difficulties in ensuring patient adherence, and they may have additional implications.13,14 For example, when diabetic complications begin to affect different body systems and a variety of specialists become involved in treating the disease, primary care providers may find it difficult to ensure that treatment protocols are followed.

Underscoring provider perceptions of patient barriers to care, an AHRQ-funded study looking at how diabetic patients respond to physician recommendations found that patients were much less likely to follow recommendations about diet (69 percent) and exercise (19 percent) than they were to follow medication regimens (91 percent).15 The researchers suggested that the reasons for poor adherence to lifestyle recommendations could include public awareness of controversy over the validity of recommendations, belief that physicians can sometimes be wrong, belief that physician instructions can be ignored, and the difficulty of incorporating new behavior into their daily lives.

Table 2. Summary of barriers to treatment goals

Physician barriers:

  • Diabetes affects many body systems.
  • Controversies about proper diagnosis and treatment.
  • Uncertainty of prognosis.
  • Lack of time.
  • Clinical inertia.

Patient barriers:

  • The frequently asymptomatic character of diabetes (removing an important incentive for self-care).
  • Necessity of daily patient interventions.
  • Need to alter lifestyle (primarily diet and exercise patterns).
  • Acute illness.
  • Hypoglycemia due to treatment.
  • Patient refusal to follow provider recommendations.
  • Lack of patient educational materials.

Sources: Larme AC, Pugh JA. Attitudes of primary care providers toward diabetes. Barriers to guideline implementation. Diabetes Care 1998; 21(9):1391-6; El-Kebbi IM, Ziemer DC, Gallina, DI, et al. Diabetes in urban African-Americans. XV. Identification of barriers to provider adherence to management protocols. Diabetes Care 1999; 22(10):1617-20.

Return to Contents
Proceed to Next Section

Current as of November 2001
Internet Citation: Improving Care for Diabetes Patients Through Intensive Therapy and a Team Approach: Research in Action, Issue 2. November 2001. Agency for Healthcare Research and Quality, Rockville, MD.