Disease Management For Diabetes

Disease Management for

Diabetes

DISCLAIMER: The information contained in this annotated bibliography was obtained from the publications listed. The National Pharmaceutical Council (NPC) has worked to ensure that the annotations accurately reflect the information contained in the publications, but cannot guarantee the accuracy of the annotations or the publications. There are articles available on the treatment of diabetes that are not included in this bibliography, which may include relevant information not covered herein. The inclusion of any publication in this bibliography does not constitute an endorsement of that publication by NPC or an endorsement of the services, programs, treatments, or other information contained in such publication. This bibliography is designed for informational purposes only, and should not be construed as professional advice on any specific set of facts and circumstances. This bibliography is not intended to be a comprehensive source of disease management services or programs in the treatment of diabetes, or a substitute for informed medical advice. If medical advice or other expert assistance is required, readers are urged to consult a qualified health care provider or other professional. NPC is not responsible for any claims or losses that may arise from any errors or omissions in the information contained in this bibliography or in the listed publications, whether caused by NPC or originating in any of the listed publications, or any reliance thereon, whether in a clinical or other setting. © February 2004 National Pharmaceutical Council, Inc.

Disease Management for Diabetes Introduction Faced with rising health care costs and limited resources, health care providers continually seek new ways to provide high-quality, cost-efficient care.1,2 Years ago, disease management emerged as a strategy with the potential to achieve this goal.3 The Boston Consulting Group first used the phrase “disease management” in its current sense in a 1993 report. Since that time, disease management programs, techniques, and models have been designed by the pharmaceutical industry, managed care, pharmacy benefit management (PBM) plans, and most recently by state Medicaid programs.1,4–8 The Centers for Medicare and Medicaid Services and the Disease Management Association of America define disease management as a system of coordinated health care interventions and communications for populations with conditions in which patient self-care efforts are significant.9,10 Disease management supports the clinician-patient relationship and plan of care, and emphasizes prevention of disease-related exacerbations and complications using evidence-based practice guidelines and patient empowerment tools.9,10 Disease management also evaluates clinical, humanistic, and economic outcomes on an ongoing basis with the goal of improving overall health.9–11 The goals of disease management include:9–11 • Improving patient self-care through patient education, monitoring, and communication with members of the health care team. • Improving physician performance through feedback and/or reports on patient progress in compliance with protocols. • Improving communication and coordination of services between patient, physician, disease management organization, and other providers. • Improving access to services, including prevention services and prescription drugs as needed.

• Identification of patient populations. • Use of evidence-based practice guidelines. • Support of adherence to evidence-based medical practice guidelines by providing practice guidelines to physicians and other providers, reporting on the patient's progress in compliance with protocols, and

• • •

Disease management programs are used widely for many chronic diseases, but the most common diseases include asthma, congestive heart failure, diabetes mellitus, and hypertension. Considerations in selecting a disease for disease management often include:11,12 • Availability of treatment guidelines with consensus about what constitutes appropriate and effective care. • Presence of generally recognized problems in therapy that are well documented in the medical literature. • Large practice variation and a range of drug treatment modalities. • Large number of patients with the disease whose therapy could be improved. • Preventable acute events that often are associated with the chronic disease (e.g., emergency department or urgent care visits). • Outcomes that can be defined and measured in standardized and objective ways and that can be modified by application of appropriate therapy (e.g., decreased number of emergency department visits or hospitalizations). • The potential for costs savings within a short period (e.g., less than 3 years). Three major not-for-profit organizations whose mission is to promote quality health care have recognized the contribution of disease management activities to quality health care by establishing disease management certification or accreditation programs. The Joint Commission on Accreditation of Healthcare Organizations, an independent, not-for-profit organization and the nation’s predominant standards-setting and accrediting body in health care, offers

Disease Management for Diabetes

The following functions are the main components of disease management:9–11

•

providing support services to assist the physician in monitoring the patient. Provision of services designed to enhance patient selfmanagement and adherence to the patient’s treatment plan. Routine reporting and feedback to the health care providers and to the patient. Communication and collaboration among providers and between the patient and the patient’s providers. Collection and analysis of process and outcomes measures along with a system to make necessary changes based on the findings of the process and outcomes measures.

[1]

disease-specific care program certification. Program certification is based on an assessment of compliance with consensus-based national standards, effective use of established clinical practice guidelines to manage and optimize care, and activities for performance measurement and improvement.13 The National Committee for Quality Assurance recently began accrediting disease management programs on the basis of standards that are patient oriented, practitioner oriented, or both. It also offers organizations certification for program design (i.e., content development), systems (i.e., clinical information and other support systems), or patient contact (e.g., for nurse call centers and other organizations without comprehensive activities).14 Many disease management organizations were so eager to undergo the accreditation or certification process that they volunteered to do so before the standards were finalized.15 The Utilization Review Accreditation Commission (URAC), also known as the American Accreditation HealthCare Commission, establishes standards for the health care and insurance industries. By establishing standards, education and communication programs, and a process of accreditation, URAC motivates purchasers, providers, and patients to achieve excellence, thus promoting continuous improvement in the quality and efficiency of health care delivery. URAC has accreditation programs for disease management as well as case management, claims processing, core accreditation, credential verification, health call centers, health networks, health plans, health provider credentialing, health utilization management, health Web sites, Health Insurance Portability and Accountability Act privacy and security, independent review organizations, vendor certification, and workers’ compensation utilization management. URAC also has goals for disease management accreditation and case management.16

Disease Management for Diabetes

Penetration and Trends

[2]

The ultimate goal of disease management is to produce optimal health outcomes for patients. Therefore, virtually all stakeholders in health care want to be involved. Disease management is of interest to providers, patients, managed care organizations, insurance companies, government agencies, PBM firms, and employer purchasing coalitions.7,11 Most disease management programs are implemented through health maintenance organizations (HMOs), PBM firms, or Medicaid agencies.9–12 Some organizations choose to hire a vendor and contract out disease management

services, whereas others choose to develop their own programs. Each method has advantages and disadvantages; success often depends on the organization and its level of resources and commitment.

Managed Care Organizations and Pharmacy Benefit Management Firms Managed care organizations and PBM firms were the first to implement disease management programs. PBM firms offer disease management programs and services to employers and managed care clients as part of their overall benefit management services.17 The 1998 Novartis Pharmacy Benefit Report indicated that 75% of PBM pharmacy directors were expending resources to develop disease management programs for conditions that respond to or depend on pharmaceutical products and services.17 HMOs reported that 16% of their disease management programs were provided through a PBM firm.17 Most employers reported using PBM firms to manage costs, and many employers used PBM firms to provide disease management services. The American Association of Health Plans (AAHP) represents more than 1,000 HMOs, preferred provider organizations, and other network-based plans. Members of the association provide health care to more than 200 million Americans nationwide. The AAHP 2002 annual survey of health plans found that 99.5% of health plans offer a disease management program.18

State Medicaid Programs In the rapidly changing environment of Medicaid managed care, it is essential for Medicaid directors and their top managed care staff to remain abreast of innovations in organization and payment that are occurring to serve the special needs of the Medicaid population. Traditionally, state Medicaid programs either have retained insurance risk and paid on a fee-for-service basis or have outsourced risk and contracted with Medicaid HMOs. Disease management represents a method of managed care in the middle between traditional fee-for-service and HMOs. Four types of models are emerging:19 1.

Medicaid health outcomes partnerships are usually applied to an existing fee-for-service primary care case management program. Medicaid programs focus on high-priority diseases, offering a number of support systems to help existing Medicaid providers better serve the patients assigned to them.20

2.

3.

4.

Disease management organizations are outside contractors who are retained by the state to address particular diseases, either by supplementing existing Medicaid providers and their case management activities or by taking over responsibility for targeted patients. Pay-for-performance approaches establish new rules for scope of practice or referrals and involve nontraditional providers in the care of patients with specific diseases. The nontraditional providers are paid a special fee contingent on improving health outcomes or lowering costs. Centers of excellence focus on particular disease episodes for high-cost, high-volume diseases and select a network of hospitals, physicians, and other providers who are already organized to receive a prospective, bundled payment per episode of care. To meet criteria for designation as a center of excellence, an organization must provide written documentation of the quality and outcomes of care for a selected disease.

Most states are actively involved in the disease management process. By far, the diseases most often focused on in these programs are asthma and diabetes. Other diseases and conditions included in state disease management programs are arthritis, congestive heart failure, depression, gastrointestinal disease, hemophilia, HIV infection/AIDS, hyperkinetic activity, dyslipidemia, mental health, otitis media, pregnancy, smoking, ulcer, and upper respiratory infections.19 Current information about state disease/case management activities is available on the Web at http://www.dmnow.org/state_activities/. Concerns have been raised about inadequate monitoring of clinical outcomes in patients with diabetes by state Medicaid agencies when implementing cost-containment strategies for the Medicaid pharmaceutical benefit.21 Additional research is needed to determine whether costcontainment strategies adversely affect patient outcomes.

In the 1990s, managed care organizations began an intense utilization review process to identify areas in which cost control measures would be appropriate. Diabetes was one of the first diseases selected because there is great opportunity to treat this disease more effectively and to develop programs that will help payers and plans manage the high costs associated with it (see “Economic Impact”).

Economic Impact In 2002, the total cost of diabetes was approximately $132 billion, of which $92 billion was for direct medical costs and $40 billion was for indirect costs associated with lost productivity due to disability and death.24 Inpatient hospital care and nursing home care accounted for approximately $40 billion and $14 billion, respectively, of the direct medical costs.24 Nearly 17 million days of hospitalization and more than 82 million nursing home days were attributed to diabetes in 2002.24 The total per capita health care expenditure for patients with diabetes (including health care costs not associated with diabetes) amounted to more than $13,000 in 2002.24 In contrast, the total per capita health care expenditure for people without diabetes was only about $2,500.24

Epidemiology Approximately 6.2% of the U.S. population—an estimated 17 million Americans—have diabetes mellitus.22 However, only about two in three of these patients are aware that they have the disease. Diabetes is the sixth leading cause of

Disease Management for Diabetes

Why Focus on Diabetes?

Diabetes disease management programs continue to gain in popularity as providers try to contain health care costs. Diabetes affects roughly 1 in 20 Americans, but the disease accounts for 1 of every 10 health care dollars spent in the United States.22 Diabetes is among the leading health indicators identified in Healthy People 2010, a set of health objectives for the nation to achieve in the first decade of the new century.23 Diabetes often is selected as a disease for intervention for the following reasons: • High-cost patients can be identified easily based on the patterns of drug use (e.g., chronic refills of insulin or antidiabetic agents). • Consistent clinical practice guidelines are available with core recommendations that apply to both children and adults. • Validated outcome measures (e.g., hospital admissions, emergency department admissions) that can help measure the effectiveness of the interventions are available. • Communication programs are available and have been shown to work in the treatment of diabetes. • Patient education materials are plentiful. • Feedback and information necessary for behavior modification in health care practitioners can be generated easily.

[3]

death and the leading cause of both treated end-stage renal disease and new cases of blindness in the United States. Diabetes increases the risk for heart disease and stroke twoto fourfold. Diabetes also causes nerve damage and increases the risk for lower-limb amputation.22 Certain racial and ethnic groups (e.g., African Americans, Hispanics/Latinos, Asian Americans, Pacific Islanders, Native Americans) are at increased risk for diabetes.22 Other risk factors for diabetes mellitus include a family history of the disease, age greater than 45 years, obesity, lack of regular exercise, dyslipidemia (high triglyceride or low high-density lipoprotein cholesterol levels), a history of gestational diabetes, and delivery of a baby weighing nine pounds or more at birth.

What Is Diabetes?

Disease Management for Diabetes

Diabetes mellitus is a heterogeneous group of chronic metabolic disorders that result in hyperglycemia—excess blood levels of glucose, which provides energy to cells. The various forms of the disease are caused by defects in insulin secretion, insulin action, or both.25,26 The hormone insulin, which is produced by the pancreas, is vital for the transport of glucose to body cells. When glucose levels in the blood are too high, the body’s cells become starved for energy. Over time, blood vessels may be damaged, leading to eye problems (retinopathy), kidney problems (nephropathy), nerve problems (neuropathy), and cardiovascular, cerebrovascular, and peripheral vascular disease.25 Hyperglycemia also causes excessive urination (polyuria) and excessive thirst (polydipsia).25,26 Approximately 10% of patients with diabetes have type 1, or insulin-dependent, diabetes, which is characterized by an absolute lack of insulin. Type 1 disease is caused by an autoimmune process that destroys the pancreatic beta cells that produce insulin.25,26 Because type 1 diabetes usually

[4]

develops in childhood or early adulthood, it was once referred to as juvenile diabetes, although it also occurs in adults.25,26 The great majority—about 90%—of patients with diabetes mellitus have type 2, or non–insulin-dependent, diabetes. Type 2 diabetes is characterized by insulin resistance (reduced sensitivity of cells to insulin), a relative insulin deficiency, or both.25 In people with type 2 diabetes, the pancreas may not produce enough insulin, the body’s cells may be unable to utilize the insulin that is produced, or both. Type 2 diabetes usually develops in adulthood, and most patients with type 2 disease are obese.25,26

Diabetes Health Goals The basic goals of diabetes therapy remained largely unchanged for many years. Many clinicians assumed that strict management of blood glucose levels was beneficial to the patient’s welfare, although few clinical data supported this idea. The report of the Diabetes Control and Complications Trial (DCCT) was made available in 1993.27 This long-term study evaluated (1) whether intensive management of blood glucose levels prevents the development of diabetes complications (primary prevention) and (2) whether intensive management prevents or limits the progression of diabetes complications (secondary prevention).27 The 1,441 patients in the trial had type 1 diabetes and were randomly assigned to intensive and conventional treatment groups (Table 1).27 The study results are shown in Table 2. The dramatic findings of the DCCT changed goals for the treatment of type 1 diabetes.27 The findings from the DCCT have led to plans to apply similar principles to managing type 2 diabetes because the cellular changes causing complications in both type 1 and type 2 disease are very similar.27

Table 1. Treatment Plan in Diabetes Control and Complications Trial Treatment Parameter Insulin Testing Diet and exercise instruction Follow-up exam Care contact Source: Reference 27.

Conventional 1 or 2 daily injections Daily monitoring, urine or blood Quarterly Quarterly As needed by patient

Intensive 3 daily injections or insulin pump Blood monitoring several times a day Monthly Monthly Weekly by nurse

Table 2. Results of Diabetes Control and Complications Trial: Risk Reductions in Intensive Cohorta Complication Reduction (%) Sustained retinopathy 63 Macular edema 26 Severe nonproliferative or proliferative retinopathy 47 Laser treatment 51 Urinary albumin excretion (>40 mg/24 hrs) 39 Urinary albumin excretion (>300 mg/24 hrs) 54 Clinical neuropathy at 5 yrs 60 a

Includes both primary prevention and secondary prevention.

Source: Reference 27.

Table 3 lists commonly used authoritative guidelines for managing diabetes. Up-to-date information on treatment guidelines from various sources also is available online from the National Guideline Clearinghouse (http://www.guideline.gov/). The management of diabetes, based on information in the guidelines, is discussed in Appendix A. Table 4 is a list of organizations with information about diabetes for patients.

Review of the Diabetes Disease Management Literature

Methodologies Reports on 67 diabetes disease management programs (Appendix B and Appendix C) met the inclusion criteria (other programs that did not meet the criteria are described in Appendix D). Various methods were used to identify patients to participate in the disease management programs. Nine programs were developed exclusively for children, an age group in whom diabetes must be carefully monitored to prevent future complications. Children were included with adults in another six programs. The size of the patient population in these studies ranged from 17 to 4,400. Most of the interventions were educational programs for patients. Sixteen were educational programs directed at health care providers (usually nurses or physicians). Twentyfive programs were based on guidelines widely accepted by the medical community (e.g., the American Diabetes Association). Most of the educational programs that were targeted to patients were administered by specially trained nurses or health educators (i.e., non-physicians). In some of the studies, educational programs were administered by a team of providers such as physicians, physician assistants, nurses, diabetes nurse educators, nutritionists, pharmacists, social workers, and podiatrists. Various formats and settings were used for the educational programs. Individualized and small-group sessions supplemented by audiovisual and printed information were common. Program content typically included diabetes selfmanagement principles, concepts about the disease, and information about diet, exercise, and weight loss. Newsletters, written in easy-to-understand language, often were used. Most programs were conducted in an outpatient clinic setting, although a few programs were provided in hospitals (i.e., to inpatients), physicians’ offices, and pharmacies. Thirty-five of the studies that evaluated disease management programs were randomized, controlled trials. In eight studies, patients served as their own controls. In 14 studies, outcomes were assessed over a 1-year period after the intervention.

Disease Management for Diabetes

A comprehensive search of the diabetes disease management literature was conducted to identify reports on the impact of educational interventions to improve diabetes treatment and management. These reports discuss the impact of specific educational interventions or programs on diabetes treatment and management. These may not be comprehensive disease management programs. The database searched was MEDLINE, the National Library of Medicine’s premier database of more than 12 million bibliographic citations and abstracts from more than 4,600 journals in the fields of medicine, nursing, dentistry, veterinary medicine, health care systems, and preclinical sciences. The decision to index a journal in MEDLINE is based on scientific policy and quality considerations. The MEDLINE database was searched from January 1990 to October 2003, using the terms “disease management”

AND “diabetes.” Because disease management is a relatively new concept, articles were manually culled from journals that are known to publish reports about disease management programs but that had not been indexed by MEDLINE (e.g., Disease Management and Health Outcomes). The primary criteria for inclusion of a study report in the analysis were (1) an educational intervention to improve treatment and management of diabetes and (2) measurement of the impact of that intervention.

[5]

Table 3. Authoritative Guidelines for Managing Diabetesa 1. American Association of Clinical Endocrinologists/American College of Endocrinology Medical guidelines for the management of diabetes mellitus. Available in print (Endocr Pract. 2002;8[suppl 1]:41–82) and online at: http://www.aace.com/clin/guidelines/ diabetes_2002.pdf. Position statement on insulin resistance syndrome. Available in print (Endocr Pract. 2003;9:240–252) and online at: http://www.aace.com/clin/guidelines/ACEIRSPositionStatement .pdf. 2. American College of Physicians The evidence base for tight blood pressure control in the management of type 2 diabetes mellitus. Available in print (Ann Intern Med. 2003;138:587–592) and online at: http://www.annals.org/cgi/reprint/138/7/587.pdf. Treatment of hypertension in type 2 diabetes mellitus: blood pressure goals, choice of agents, and setting priorities in diabetes care. Available in print (Ann Intern Med. 2003;138:593–602) and online at: http://www.annals.org/cgi/content/full/138/7/593 3. American Diabetes Association Clinical practice recommendations, including position statements, technical reviews, and consensus statements. Available in print annually in a supplement to the January issue of Diabetes Care, which is published by the American Diabetes Association, and online at: http://care.diabetesjournals.org/

4. American Dietetic Association Nutrition practice guidelines for type 1 and type 2 diabetes mellitus (December 2001). Available in print and on CD-ROM at www.eatright.org and 120 South Riverside Plaza, Suite 2000, Chicago, IL 60606-6995 (1-800-877-1600, extension 5000). 5. American Society of Health-System Pharmacists Therapeutic position statement on strict glycemic control in patients with diabetes. Available in print (Am J Health Syst Pharm. 2003;60:2357–2362) and online at: http://www.ashp.org/bestpractices/tps/Therapeutic%20Positio n%20Statement%20Strict%20Glycemic%20Control%20in%20 Patients%20with%20Diabetes.pdf. 6. Centers for Disease Control and Prevention Recommendations for health care system and selfmanagement education interventions to reduce morbidity and mortality from diabetes. Available in print (Am J Prev Med. 2002;22[suppl 4]:10–4) and online at: http://www. thecommunityguide.org/diabetes/dm-AJPM-recs.pdf. 7. U.S. Preventive Services Task Force Screening for type 2 diabetes mellitus in adults: recommendations and rationale. Available in print (Ann Intern Med. 2003;138:212–214) and online at: http://www.ahrq.gov/ clinic/3rduspstf/diabscr/diabetrr.htm.

a Clinical practice is subject to constant change, and the guidelines in this list may become outdated or be superceded by newer ones. The reader is encouraged to consult the National Guideline Clearinghouse (http://www.guideline.gov/), a public resource for evidence-based clinical practice guidelines sponsored by the Agency for Healthcare Research and Quality (formerly the Agency for Health Care Policy and Research) in partnership with the American Medical Association and the American Association of Health Plans, for the most current guidelines.

Disease Management for Diabetes

Outcomes

[6]

Clinical values (e.g., glycosylated hemoglobin, fasting blood glucose levels, medication requirements, body weight, blood pressure, serum cholesterol, triglyceride levels) were used as outcome measures in most of the studies. Foot examinations also were used as an outcome measure in many of the studies. The number of emergency department visits or hospital admissions was used as an outcome measure in six studies. In three of these six studies, the intervention significantly reduced the number of emergency department visits or hospital admissions. Three studies found a significant reduction in physician office visits as a result of the intervention. Diabetes knowledge increased significantly as a result of

the intervention in all 20 of the studies in which cognitive outcomes were measured. Nine studies demonstrated improved understanding and self-management practices as a result of the intervention. All nine of the studies in which patient adherence to the drug regimen or education program was evaluated demonstrated a positive influence of the intervention on adherence. The use of medications was an outcome measure in 14 studies. The intervention led to improved medication use or understanding in eight of these studies. The dispensing of drugs recommended for the treatment of diabetes was not measured in any of the studies. Cost-effectiveness was measured in 10 of the 65 studies, and 2 studies determined that the cost savings were

attributable to the intervention (e.g., a reduced number of physician office visits). The studies that measured costeffectiveness portend that the cost-benefit ratio of providing a diabetes disease management program will decrease (i.e., become more favorable) as the intervention is applied to a larger patient population. Over time, the fixed costs associated with developing a disease management program will be spread over larger numbers of patients.

The Future of Disease Management Disease management can improve patient outcomes and quality of life while potentially reducing overall costs. It is an important approach to integrated care. As health care payers incorporate disease management principles into the delivery of care, they need to become more sophisticated in contracting with outside vendors for these services. The Disease Management Association of America works with potential customers to address issues associated with contracting, such as data contracting and risk sharing. Currently, the Disease Management Association of America has more than 100 members that provide disease management services. Disease management vendors have begun using the Internet to reach out to target populations. The Internet allows two-way communication between clinicians and patients, as well as immediate and free access to educational materials. Compared with traditional office visits and postal mailings, the Internet may save time and money. Initially the Internet may be used to educate Medicaid physicians, nurses, pharmacists, and other providers about disease management. As more people gain access to personal computers and enter the “information superhighway,” the Internet will become an increasingly powerful tool. Disease management is a useful, efficient approach to health care. It will continue to gain widespread acceptance among health plans that provide care for patients with chronic disease.

Table 4. Organizations with Information About Diabetes for Patients American Association of Diabetes Educators 100 West Monroe, Suite 400 Chicago, IL 60603 1-800-338-3633 http://www.aadenet.org American Diabetes Association Attn: National Call Center 1701 North Beauregard Street Alexandria, VA 22311 1-800-DIABETES (1-800-342-2383) E-mail: [email protected] http://www.diabetes.org Centers for Disease Control and Prevention National Center for Chronic Disease Prevention and Health Promotion Diabetes Public Health Resource P.O. Box 8728 Silver Spring, MD 20910 1-877-CDC-DIAB http://www.cdc.gov/diabetes National Diabetes Education Program (a partnership of the National Institutes of Health, the Centers for Disease Control and Prevention, and more than 200 public and private organizations) http://ndep.nih.gov National Diabetes Information Clearinghouse (a service of the National Institute of Diabetes & Digestive & Kidney Diseases) 1 Information Way Bethesda, MD 20892-3560 1-800-860-8747 or 301-654-3327 Fax: 301-907-8906 E-mail: [email protected] http://diabetes.niddk.nih.gov

Disease Management for Diabetes [7]

Appendix A. Management of Diabetes The goals of treatment for diabetes are to reduce and control blood glucose levels, relieve the symptoms of the disease, and prevent complications.28 Numerous studies have demonstrated that intensive treatment and careful control of blood glucose levels can reduce the risk of complications from diabetes.27,29–32 The American Diabetes Association recommends the formulation of an individualized diabetes management plan in collaboration with the patient.28 A high degree of patient involvement in self-management should be part of this plan, including frequent self-monitoring of blood glucose.28 Long-term control of blood glucose levels can be assessed by measuring the extent to which glucose is bound to the A1c component of hemoglobin. This glycosylated hemoglobin value is expressed as a percentage and should be less than 7% in patients with diabetes.28 A proven method to curb the escalating costs of diabetes-related morbidity and mortality is through education and disease management.27–29 The successful management of diabetes requires lifestyle changes for the patient (e.g., diet, exercise, self-monitoring of blood and possibly urine). The patient must be involved in the decisionmaking process and must learn as much as possible about diabetes, including why the symptoms of diabetes occur and how they can be alleviated through control of blood glucose levels. Education is essential for motivating patients to manage their disease, encouraging changes in lifestyle, and improving patient outcomes. Disease management programs provide an excellent way to integrate education into the management of the disease.

Nonpharmacologic Interventions Medical nutrition therapy (i.e., diet) and exercise are important aspects of nonpharmacologic treatment for diabetes.33,34 Weight loss is a vital part of treatment for type 2 diabetes because it can help improve the sensitivity of cells to insulin and the uptake of glucose by cells.

Disease Management for Diabetes

A goal of medical nutrition therapy is to attain and maintain blood glucose levels in the normal range or as close to normal as possible.33 Patients with type 2 diabetes typically have dyslipidemia, and another goal for these patients is to improve the lipid profile.33 Blood pressure levels that reduce the risk for vascular disease are another goal for patients with type 2 disease.33 Caloric restriction and weight loss are recommended for patients who are overweight, because weight loss may help normalize blood glucose levels.33

[8]

Physical activity can improve insulin sensitivity (i.e., help cells take in more glucose), which can lower elevated blood glucose levels.34 Exercise also provides many other health benefits, including improved energy levels and increased self-esteem.

Pharmacologic Interventions All patients with type 1 diabetes require insulin injections. Patients with type 2 disease who have multiple symptoms, are severely hyperglycemic, are pregnant, or have ketosis (increased ketone production due to the breakdown of fat for energy) also should use

insulin injections.35 Other patients with type 2 diabetes may attempt to control their diabetes through diet and exercise for a 3-month trial period.35 However, only 10% of patients with type 2 disease can control their blood glucose levels with diet and exercise alone.35 Various oral antidiabetic agents (or insulin) may be used in conjunction with diet and exercise to control blood glucose levels in these patients. Insulin. Injections of insulin help cells take in glucose, thereby reducing blood glucose levels. Insulin is usually given as an intermediate-acting insulin or a mixture of an intermediate-acting insulin plus a short-acting or rapid-acting insulin. Injections should be administered at least twice daily.35 Some patients may require three or more daily insulin injections or continuous subcutaneous infusions with an insulin pump.35 Oral Antidiabetic Agents. Classes of oral antidiabetic agents include sulfonylureas, biguanides, alpha-glucosidase inhibitors, meglitinides, and thiazolidinediones. Each class has a unique mechanism of action, pharmacologic effects, and adverse-effect profile. If monotherapy with an oral antidiabetic agent (plus diet and exercise) is ineffective in controlling blood glucose levels, adding a second antidiabetic agent (preferably with a different mechanism of action) or insulin is recommended.35 Using two or more oral antidiabetic agents with different mechanisms of action may provide an additive blood glucose–lowering effect and minimize adverse effects.35 If combination therapy is not effective, insulin monotherapy is recommended.35 Sulfonylureas. Sulfonylureas are the best-established oral antidiabetic agents and are a rational choice for initial drug therapy.35 They act primarily to increase insulin secretion by pancreatic beta cells and can cause hypoglycemia.25 Weight gain also can occur with sulfonylurea therapy.25 Biguanide. Currently, metformin is the only drug in the biguanide class. Metformin reduces blood glucose concentrations by increasing glucose uptake in the peripheral muscles and decreasing the amount of glucose produced and released in the liver.25 It also improves the lipid profile.35 The most common adverse effects from metformin are diarrhea and other gastrointestinal disorders.25 Metformin is contraindicated in patients with renal impairment because of the risk of lactic acidosis, a potentially serious complication.25 Alpha-Glucosidase Inhibitors. The alpha-glucosidase inhibitors acarbose and miglitol inhibit the action of intestinal enzymes that break down carbohydrates.25 These oral antidiabetic agents delay glucose absorption and are particularly useful for patients with postprandial hyperglycemia (high blood sugar levels after eating). However, they are not as effective as sulfonylureas and biguanides in providing long-term control of blood glucose levels.35 Alphaglucosidase inhibitors do not cause hypoglycemia or weight gain. However, they can cause flatulence, diarrhea, and abdominal cramps.25,36

Appendix A. Management of Diabetes

(continued)

Repaglinide. Repaglinide is a meglitinide that stimulates insulin release by pancreatic beta cells, although it is not a sulfonylurea. Repaglinide is indicated as monotherapy or, if blood glucose control cannot be achieved with repaglinide or metformin alone, in combination with metformin.36 Repaglinide and metformin together act synergistically to reduce glucose levels to lower levels than can be achieved with either drug alone.36 Repaglinide usually is taken two to four times a day, within 30 minutes before meals.36 Adverse effects include hypoglycemia and weight gain. Patients who are elderly, debilitated, or malnourished or who have adrenal, pituitary, or hepatic insufficiency are particularly susceptible to hypoglycemic reactions.36 Nateglinide. Nateglinide is a non-sulfonylurea with a mechanism of action that is similar to that of repaglinide; it stimulates the release of insulin by pancreatic beta cells.36 Like repaglinide, nateglinide is indicated for use as monotherapy or in combination with metformin.36 Nateglinide and metformin have a synergistic blood glucose–lowering effect.36

Nateglinide is taken three times daily, within 30 minutes before meals.36 Hypoglycemia and weight gain can occur during nateglinide therapy.36 Thiazolidinediones. Thiazolidinediones are the newest class of antidiabetic agents. They increase insulin sensitivity and action in liver, muscle, and fatty tissues. Thiazolidinediones reduce blood glucose concentrations and improve the lipid profile.36 These agents do not stimulate insulin secretion or cause hypoglycemia.36 However, they can cause weight gain.36 Troglitazone, the first thiazolidinedione to become available, has been withdrawn from the market because of reports of severe liver toxicity resulting in death or requiring transplantation.37 The risk of severe liver toxicity appears to be lower with the newer thiazolidinediones rosiglitazone and pioglitazone.36 Nevertheless, periodic monitoring of liver enzymes is recommended during treatment with these agents.36

Disease Management for Diabetes [9]

Appendix B. Reports of the Impact of Disease Management Interventions on Treatment of Diabetes The impact of a diabetes nurse educator on nurses’ knowledge of diabetes and nursing interventions in a home care setting. Adams CE, Cook DL. The Diabetes Educator. 1994;20:49–53. The impact of a diabetes nurse educator on the diabetes knowledge of and patient care provided by registered nurses at home health care agencies was assessed by comparing a home health care agency that had a diabetes nurse educator with an agency without a diabetes nurse educator. The responsibilities of the diabetes nurse educator included completing needs assessments for both patients and staff; developing nursing standards of care, staff education programs, and specific methods for implementing and evaluating care; designing diabetes self-care content that is user friendly; and providing ongoing support for staff nurses. The average score on a test of basic diabetes knowledge was 77.3% at the agency with a diabetes nurse educator (N=45 registered nurses) and 70.0% at the agency without the educator (N=17 registered nurses), a difference that is significant. Patient care was assessed using a retrospective chart review of 31 patient charts at each agency and a list of 33 items addressing patient admission, care plan, ongoing care (assessment, patient education, nursing care treatments, and evaluation), and discharge. This list was based on guidelines of the American Diabetes Association and American Association of Diabetes Educators and federal and state regulations governing home health care. Patient care scores were significantly higher at the agency with the diabetes nurse educator than at the agency without the educator. Patient care outcomes (e.g., glycosylated hemoglobin) were not studied. The investigators concluded that the role of the diabetes nurse educator should include educating staff nurses and setting up standards of patient care as well as educating patients.

Disease Management for Diabetes

A national diabetes care and education programme: the Ghana model. Amoah AG, Owusu SK, Acheampong JW, Agyenim-Boateng K, Asare HR, Owusu AA, Mensah-Poku MF, Adamu FC, Amegashie RA, Saunders JT, Fang WL, Pastors JG, Sanborn C, Barrett EJ, Woode MK. Diabetes Res Clin Pract. 2000 Aug;49(2-3):149–57.

[10]

A national diabetes care and education program was created in a developing country (Ghana) through international collaboration between medical schools, industry, and governmental health care institutions. Ghana’s population of 18 million (85% of whom are less than 45 years old) is served by 2 teaching hospitals, 9 regional hospitals, and 100 subregional hospitals. Two core teams comprising a physician, a nurse, and a dietitian underwent intensive training in a multidisciplinary approach to diabetes care and education. The core teams subsequently developed a patient education booklet and a professional diabetes education curriculum for training health care providers at regional and subregional hospitals and

health care facilities. Culturally appropriate illustrations were developed for teaching illiterate patients. After 3 years, trained diabetes health care teams were in place and diabetes services were provided in all regional and 63% of subregional facilities. The number of patients with diabetes registered before the program was 4,719, and after the program, the number was 13,643. Prohibitively high prices of antidiabetic medications and supplies were a problem. The experience implementing this program can serve as a model for other developing countries. Evaluation of an activated patient diabetes education newsletter. Anderson RM, Fitzgerald JT, Funnell MM, Barr PA, Stepien CJ, Hiss RG, Armbruster BA. The Diabetes Educator. 1994;20:29–34. In a nonrandomized study, 1,863 patients with type 1 or 2 diabetes were recruited from physician offices, pharmacies, clinic waiting rooms, and diabetes patient education programs in nine Michigan communities. The patients were asked to return a postage-paid postcard if they wished to receive a free monthly newsletter about diabetes. The newsletter was approximately one page long and provided readable, concise, and action-oriented information about diabetes care. The newsletter also encouraged patients to become assertive consumers of diabetes care who are prepared to interact with physicians, nurses, and dietitians during diabetes care visits. Each of the 21 newsletters mailed during the 2-year study period addressed a specific topic (e.g., diabetic retinopathy) and concluded with a specific recommended action for patients to take (e.g., ask the physician for a referral for an eye examination). The cost of providing the newsletters was about 25 cents per person per month. Among 720 patients who received and read most or all of the newsletters, the newsletters were rated “Helpful” or “Very Helpful” by 77% of patients with type 2 diabetes who were not using insulin, 70% of patients with type 2 diabetes who were using insulin, and 57% of patients with type 1 diabetes. Patients who found the newsletters “Helpful” or “Very Helpful” were older and in poorer overall health and had more complications, lower incomes, and a lesser understanding of diabetes than patients who rated the newsletters “Not Helpful” or “Somewhat Helpful.” Patients were more likely to discuss newsletter topics with family or friends than with physicians, nurses, or dietitians and were least likely to discuss topics with nurses and dietitians. The topic most frequently discussed with physicians was the importance of annual eye examinations. Few of the newsletters resulted in changes in diabetes care; on average, 7.8% of patients made changes in response to a newsletter. The topic resulting in change in the largest percentage of patients (13.1%) was the importance of daily self-testing of blood glucose levels. Patients who found the newsletters “Helpful” or “Very Helpful” were significantly more likely to make changes in diabetes care than patients who rated the newsletters “Not Helpful” or “Somewhat Helpful.”

Appendix B. Reports of the Impact of Disease Management Interventions on Treatment of Diabetes (continued) Learning to empower patients: results of professional education program for diabetes educators. Anderson RM, Funnell MM, Barr PA, Dedrick RF, David WK. Diabetes Care. 1991;14:584–9. A counseling and educational skills training program was conducted for 23 diabetes educators (including 20 nurses and 2 dietitians) who responded to a flyer mailed to diabetes educators in Michigan. The program was designed to equip the educators to use a patient empowerment approach to diabetes patient education, which requires health care providers to help patients increase their selfawareness of their values, needs, and goals for diabetes care; assume greater autonomy; and make informed decisions about their own care. The program for diabetes educators involved a 3-day simulated diabetes care regimen (two daily saline injections, self-monitoring of blood glucose four times daily, a 1,200-kcal/day diet, daily foot care and aerobic exercise, and extensive record keeping) to give the educators a greater sense of the challenges faced by patients. The simulated diabetes care regimen was followed by a 3-day intensive skills-based workshop, using a fourphase learning sequence with information (lectures, handouts, and reprints) on educational and counseling skills, demonstration of counseling skills, videotaped small-group practice in counseling, and review of videotaped counseling sessions. Values clarification, the impact of communication style on approach to diabetes education, and effective and ineffective teaching also were addressed in the program. Counseling skills in simulated and actual patient encounters improved significantly as a result of the program. A questionnaire completed before and after the program revealed that the attitudes of the diabetes educators toward patient autonomy also improved significantly. Attitudes toward the need for special training for diabetes educators did not change significantly. Participants considered the 3-day simulated diabetes care regimen a valuable learning experience that increased their appreciation of patients’ concerns. Boost glycemic control in teen diabetics through “familyfocused teamwork.” Anon. Dis Manag Advis. 2003;9:120–2.

Diabetes QI program cuts complications by 60%. Anon. Hosp Case Manag. 2003;11(7):103–4. A quality improvement initiative was undertaken to improve the provision of diabetes disease management in an Iowa health system that provided care to 58,000 patients with diabetes. The health system had 11 hospitals, each of which had a diabetic teaching center (clinic). The initiative involved developing a strategic plan and standardizing care. A standardized patient education curriculum was adopted, and staff training was provided in patient education and data collection. Group patient education sessions were used instead of individual sessions to improve clinic operational efficiency. Reminder calls to patients for return appointments also were used. The average glycosylated hemoglobin decreased from 9.2% to 7.0% (i.e., glycemic control improved) over a 6-month period. Hourly clinic operational costs decreased by 20% as a result of the switch from individual to group patient education sessions. A comparison of learning activity packages and classroom instruction for diet management of patients with non–insulindependent diabetes mellitus. Arsenau DL, Mason AC, Wood OB, Schwab E, Green D. The Diabetes Educator. 1994;20:509–14. Forty patients attending a diabetes education program at a hospital-based diabetes center volunteered to participate in a randomized 5-month study comparing the effectiveness of diabetes education using an individualized method of learning with that of formal classroom sessions. The individualized learning program was based on goals of the American Diabetes Association and American Dietetic Association. It comprised pretest and posttest materials, goals and learning objectives, factual information (using inexpensive existing materials), and activities. Program content was reviewed by two registered dietitians, a nutrition specialist, and an education specialist. Follow-up visits were conducted 2 and 5 months after randomization. There were no significant differences between the two groups in fasting plasma glucose, glycosylated hemoglobin, knowledge, behavior (a measure of compliance), or body weight at the 2-month follow-up visit. The behavior score in the individualized group was significantly increased (indicating improved compliance) and body weight was significantly decreased at the 2-month follow-up visit compared with baseline. However, behavior scores decreased and fasting plasma glucose concentrations increased significantly in the

Disease Management for Diabetes

Family-focused teamwork is an approach to improving glycemic control in teenagers with type 1 diabetes that was developed by researchers at Joslin Diabetes Center in Boston, Massachusetts. Sharing of responsibility by the teenager and his or her parents is a key element in the approach. A series of educational modules was developed and used over the course of a year at routine visits. To make the program cost-effective, college graduates without a health care background administered the modules instead of more costly health care professionals. A randomized controlled study was conducted of 105 children 8–17 years of age to compare the intervention with usual care. After 1 year, glycemic control was significantly better in the intervention group (8.2%) than in the usual

care group (8.7%), although there was room for improvement to reach the goal of 7% in both groups. Participants in the intervention group were twice as likely as participants in the usual care group to maintain or increase family involvement with diabetesrelated tasks.

[11]

Appendix B. Reports of the Impact of Disease Management Interventions on Treatment of Diabetes (continued) subsequent 3 months. Knowledge and body weight improved over the course of the 5-month study in the individualized group. At the time of the 2-month follow-up in the classroom group, there were significant decreases in fasting plasma glucose and glycosylated hemoglobin levels and an increase in behavior scores compared with baseline. Behavior correlated with glycemic control. Glycosylated hemoglobin and behavior scores improved over the course of the 5-month study. Change in knowledge was the only variable that differed significantly between the two groups. Increases in knowledge did not correlate with improved glycemic control or body weight. The results of the study suggest that an individualized method of education is as effective as traditional classroom instruction in educating patients with diabetes. The individualized method may be more costeffective, particularly for patients who live a long distance from the classroom site or for whom the classroom schedule is inconvenient (e.g., employed patients with inflexible work schedules). Choice of an education method for patients with diabetes might take into consideration various patient factors.

Disease Management for Diabetes

Nurse case management to improve glycemic control in diabetic patients in a health maintenance organization. Aubert RE, Herman WH, Waters J, Moore W, Sutton D, Peterson BL, Bailey CM, Koplan JP. Ann Intern Med. 1998;129:605–12.

[12]

A 12-month randomized, controlled trial was conducted to compare diabetes control in patients receiving nurse case management with that in patients receiving usual care at primary care clinics in a group-model health maintenance organization. Patients were identified from pharmacy records and a database of patients who had visited a physician for diabetes care, been hospitalized for diabetes, been seen by a utilization management nurse, or been referred to an ophthalmologist for a diabetic retinal examination. Seventeen patients with type 1 diabetes and 121 patients with type 2 disease were randomized to the nurse case management group or the usual care group. The nurse case manager (a registered nurse and certified diabetes educator) was trained to follow detailed management algorithms under the direction of a board-certified family medicine physician and an endocrinologist. The algorithms were designed to improve glycemic control by adjusting drug therapy, planning meals, and reinforcing exercise recommendations. Patients assigned to the nurse case management group were instructed about blood glucose monitoring at an initial visit with the nurse case manager and returned for a follow-up visit 2 weeks later for reevaluation and adjustment of the treatment plan. Patients also were referred to a 5-week, 12-hour diabetes education program that addressed diet and exercise among other topics. Patients received weekly or biweekly follow-up telephone calls to discuss blood glucose values, adjust drug therapy, and reinforce dietary and exercise recommendations. Patients in the usual care group received diabetes care from their primary care physicians.

They were provided with blood glucose monitoring equipment and encouraged to participate in the 5-week diabetes education program attended by the nurse case management group. The nurse case management group had a greater decrease in glycosylated hemoglobin values than did the usual care group (1.7% versus 0.6%, respectively) over the course of the 12-month study. The difference between the two groups was evident after 6 months and was maintained for the subsequent 6 months. The reduction in glycosylated hemoglobin was greater with nurse case management than with usual care in the subset of patients with type 1 diabetes and in the subset of patients with type 2 disease. There was no significant difference between the two groups in blood pressure, serum cholesterol and triglyceride levels, body weight, insulin requirements, hospital admissions, emergency department visits, or frequency of severe hypoglycemic episodes. Self-reported health status improved significantly in the nurse case management group. A diabetes control program in a public health care setting. Baker SB, Vallbona C, Pavlik V, Fasser CE, Armbruster M, McCray R, Baker RL. Public Health Rep. 1993;108(5):595–604. Protocols for the prevention and care of diabetes-related complications of the eyes, lower extremities, and cardiovascular system were developed and implemented in nine community health centers. These facilities were located in urban low-income neighborhoods in Houston, Texas, and served approximately 4,300 patients. A culturally sensitive patient education curriculum was provided in four 2-hour sessions by a diabetes nurse educator and a nutritionist. Health care professionals received continuing medical education, and there was financial incentive for nurses to become certified diabetes educators. These efforts are part of a program that is ongoing, although some results are available after 5 years of experience. There was an increase in eye examinations from 8% to 26% of the patient population. A reduction in the incidence of legal blindness from 9.5 to 2.7 per 1,000 patients was observed after 4 years. A cost-benefit analysis of providing 12 months of screening and preventive treatment for diabetic eye disease suggests a benefit-tocost ratio of at least 2.77 for such a program. The percentage of the population with annual foot examinations increased from 18% to 44%. Control of blood pressure was achieved in 77% of patients. However, there were no significant improvements in body weight or blood glucose control. The average percentage of correct responses to test questions about diabetes increased from 65% to 85% in patients. The rate at which proliferative retinopathy and nonproliferative retinopathy (or fundus abnormalities) were correctly identified by physicians in slides or photographs was 79% and 77%, respectively, in 1990 and 95% and 86%, respectively, in 1991. In retrospective chart

Appendix B. Reports of the Impact of Disease Management Interventions on Treatment of Diabetes (continued) reviews, 83% of patients with hypertension and 92% of patients with ischemic heart disease received appropriate medications and dosages. Intensive education improves knowledge, compliance, and foot problems in type 2 diabetes. Barth R, Campbell LV, Allen S, Jupp JJ, Chisholm DJ. Diabetic Med. 1991;8:111–17. The effectiveness of an intensive foot care intervention program and a conventional one were compared in 62 patients with type 2 diabetes. Patients were recruited through an extensive radio and newspaper campaign, from referrals by general practitioners and from people attending the Diabetes Center, St. Vincent’s Hospital and other diabetes centers in Sydney, Australia. There were two types of programs (conventional and intensive). The conventional program included information on diabetes education (what is diabetes, diet and exercise), complications of diabetes, and a 1 hour lecture on foot care (washing, drying, and inspecting the feet; cutting toenails, treating minor foot problems, appropriate footwear, and dealing with extreme temperatures). The intensive program also had comprehensive diabetes education and foot care information but patients were encouraged to perform daily foot care procedures so as to avoid foot problems and reduce the risk of diabetic complications. Detailed foot care recommendations and demonstrations were given. Practice sessions on foot care were performed during the study. The intensive group showed significantly greater improvement than the conventional group in foot care knowledge (p<0.001), compliance with the recommended foot care routine (p=0.012), and compliance with the initial advice to consult a podiatrist (other than the project podiatrist) for further treatment (p=0.008). At the first follow-up visit, the intensive group also showed a significantly greater reduction in the number of foot problems requiring treatment than the conventional group. It is not possible to investigate whether the intensive group’s significant improvement in knowledge, compliance, and foot problems will reduce the number of amputations.

The use of adaptive self-care skills (e.g., the use of adaptive equipment such as syringe magnifiers, syringe loading devices, and glucose monitoring systems with speech capability and tactile aids for proper blood sample placement) was studied in 163 visuallyimpaired patients with type 1 or type 2 diabetes over a 10-year period at a university-based hospital clinic. The mean age was 56 years. Diabetes educators specializing in adaptive diabetes tech-

Success was achieved in 72% of participants. The adaptive diabetes education program was more successful in restoring independent diabetes self-care practices in patients who were young or had early visual impairment (legal blindness with some residual functional vision) than in patients who were older or had late visual impairment (light perception only or total blindness). The success rate was 57% in patients with late visual impairment compared with 81% in patients with early visual impairment, possibly because residual vision may facilitate the acquisition of self-care skills. The impact of a staged management approach to diabetes foot care in the Louisiana public hospital system. Birke JA, Horswell R, Patout CA Jr, Chen SL. J La State Med Soc. 2003;155(1):37–42. The impact of a statewide (Louisiana), public hospital-based, disease management initiative (DMI) for diabetes and a citywide (Baton Rouge) staged-management diabetes foot program (DFP) on (1) diabetes foot-related hospitalizations and (2) lower extremity amputations was assessed in a retrospective, nonrandomized study of a low-income patient population. The DMI included goals for the medical management of diabetes consistent with American Diabetes Association frequencies for performing various tests and examinations and for counseling patients on nutrition and selfmanagement. The Lower Extremity Amputation Prevention Program, which includes annual foot screening, patient education about foot care, assistance in selection of proper footwear, daily foot self-inspection, and management of simple foot problems, was implemented as part of the DMI. The staged management DFP was established to provide referral care for high-risk foot problems so that patients receive prompt treatment. Patients with diabetes were stratified by risk for foot injury based on various factors (e.g., loss of protective sensation, foot deformities, poor circulation, history of ulceration), and interventions and follow-up were provided in accordance with the degree of risk. The diabetes foot-related hospitalization rate during the year after program implementation was significantly lower than during the year before implementation both in the five facilities outside Baton Rouge where only the DMI was used and in the three facilities in Baton Rouge where the DMI and DFP programs were used. The reduction in hospitalization rate over time was greater with DMI plus DFP (44%) than with DMI alone (15%). The lower extremity amputation rate during the year after program implementation was significantly lower than during the year before

Disease Management for Diabetes

Effectiveness of a diabetes education program adapted for people with vision impairment. Bernbaum M, Wittry S, Stich T, Brusca S, Albert SG. Diabetes Care. 2000 Sep;23(9):1430–2.

niques assessed visual function and diabetes knowledge, provided instruction in basic diabetes knowledge and self-care, and assisted in the selection of self-management devices. Demonstration of proficient self-care technique was considered a successful outcome.

[13]

Appendix B. Reports of the Impact of Disease Management Interventions on Treatment of Diabetes (continued) implementation both in facilities outside Baton Rouge where only the DMI was used and in facilities in Baton Rouge where the DMI and DFP programs were used. There was no difference between groups in the reduction in lower extremity amputation rate over time (29% with DMI alone and 33% with DMI plus DFP).

Evaluation of a holistic treatment and teaching programme for patients with type 1 diabetes who failed to achieve their therapeutic goals under intensified insulin therapy. Bott U, Bott S, Hemmann D, Berger M. Diabet Med. 2000 Sep;17(9):635–43.

Randomized, controlled trial of diabetic patient education: improved knowledge without improved metabolic status. Bloomgarden ZT, Karmally W, Metzger MJ, Brothers M, Nechemias C, Bookman J, Faierman D, Ginsberg-Fellner F, Rayfiled E, Brown WV. Diabetes Care. 1987;10(3):263–72.

The impact of a specialized educational program was studied in 83 experienced patients with type 1 diabetes who failed to achieve their treatment goals for glycemic control or quality of life despite the use of intensified insulin therapy and participation in standard educational programs. Patients were referred to the program because of frequent hypoglycemic events, motivational problems, a need to refresh their diabetes knowledge, or a need for greater flexibility in the insulin treatment regimen. The specialized program was provided for small groups (4–6 patients) over a 5-day period (20 hours of education) on an inpatient basis. Insulin therapy, hypoglycemia, diet, and patient-specific concerns were addressed. Psychosocial problems (especially lack of motivation) and coping strategies were the focus of intensive group discussions.

Disease Management for Diabetes

In a randomized, controlled trial, the effect of a traditional diabetes patient education program on metabolic control (HbA1c) and other variables was assessed in 266 patients at a large, hospital-based diabetes clinic. All 749 insulin-treated, “predominantly type II” diabetic patients on the clinic roster were randomly assigned to an education group or a control group. Of these patients, 345 agreed to participate in the program and 266 (education group=127; control group=139) completed the study. Patients in both groups had contact at each clinic visit with their physician and nurse, who reviewed medications and specific problems. Patients in the education group were also offered nine monthly education sessions, which included sessions and audiovisual materials in Spanish for Hispanic patients. Material reviewed in these sessions included basic physiology; foot, skin, and dental care; insulin administration and emergencies; complications and risk factors, and nutritional information, which was presented via card games, films, filmstrips, and slides. Sixty-two percent (N=79) of the patients in the education group attended seven or more sessions and were considered “graduates.”

[14]

During the 1.5-year observation period, HbA1c fell from 6.8% to 6.1% in the education group and 6.6% to 6.3% in the control group, representing an insignificant difference (p=0.1995). Postintervention comparison of fasting blood glucose, triglycerides, cholesterol, insulin dosages, blood pressure, and foot lesion scores also failed to show significant intergroup variation. Whereas women in the education group gained less weight than those in the control group, this effect was not attributed to the intervention. However, post-intervention comparison of the groups did reveal a significant increase in cognitive scores in the education group, but not the control group, that was attributed to the program (p=0.0073). Subgroup analysis also revealed that program graduates showed significantly greater improvement in behavior scores than either nongraduates or controls. The authors noted that, as the group population sizes were sufficient to detect a significant difference in the primary outcome measure—mean HbA1c, between the education and control groups, patient education may not be an efficacious therapeutic intervention in most adults with type 2 diabetes.

Face-to-face or telephone interviews were conducted an average of 17.5 months (range 9–31 months) after the specialized program. Compared with baseline values, there was no change in the mean glycosylated hemoglobin A1c at follow-up. However, there was a significant decrease in the incidence of severe hypoglycemia (need for glucose i.v. or glucagon injection) from 0.62 episodes per patient/year at baseline to 0.16 episodes per patient/year at followup. Hospital admission decreased from 9.3 days per patient/year to 6.2 days per patient/year, a difference that is not significant. The number of sick leave days per patient/year decreased significantly from 17.0 at baseline to 7.7 at follow-up. Thus, the specialized educational program exploring motivation and other psychosocial aspects of self-management improved outcomes in experienced patients with poor glycemic control. A community-based, culturally sensitive education and group-support intervention for Mexican Americans with NIDDM: a pilot study of efficacy. Brown SA, Hanis CL. The Diabetes Educator. 1995;21:203–10. A 9-week pilot study was conducted to determine the feasibility of providing culturally sensitive diabetes patient education and group support to Mexican Americans in a rural community setting. Seven subjects were randomly selected from a cohort of 353 patients with type 2 diabetes who had been followed in another study for 12 years. Eight weekly 2-hour diabetes education sessions (culturally relevant videotape presentations, discussions, demonstrations, and a visit to a grocery store to learn to read and interpret food labels) and one 2-hour support group session were conducted in the Spanish language. A family member or friend participated in all sessions to provide support for the patient. Sessions were conducted by a clinical nurse specialist, registered dietitian, and community lay worker, all of whom were Mexican American. Program

Appendix B. Reports of the Impact of Disease Management Interventions on Treatment of Diabetes (continued) content was based on a curriculum guide from the American Association of Diabetes Educators and included nutrition (especially reduction in fat, sodium, and calorie intake), exercise, selfmonitoring of blood glucose, drug therapy, and disease complications. The sessions reflected cultural preferences (e.g., food preparation methods) and addressed religious influences (e.g., diabetes as punishment for wrongdoing, conflict from thinking that assuming responsibility for health interferes with God’s will). The sessions were conducted in the context of social activities (e.g., cooking demonstrations) at a county agricultural extension office (i.e., a nonthreatening site with an informal, social atmosphere). The use of medical terminology was minimized, and tools for assessing knowledge of diabetes were designed to obviate the need for a high reading level. Two subjects were dropped from the analysis because of serious family illness and coronary artery bypass surgery, respectively. Three months after the intervention, glycosylated hemoglobin and fasting blood glucose levels and knowledge of diabetes selfmanagement were significantly improved compared with baseline in the remaining five subjects. The subjects found the presence of family members or other support persons helpful. The group size was judged appropriate. The subjects needed additional time to learn the nutritional content of foods. An additional four weekly educational sessions were requested by the subjects. Subjects particularly appreciated the grocery store visit, use of culturally relevant videotapes, and sharing of healthy recipes. Effects of behavior-modifying education in the metabolic profile of the type 2 diabetes mellitus patient. Cabrera-Pivaral CE, Gonzalez-Perez G, Vega-Lopez G, GonzalezHita M, Centeno-Lopez M, Gonzalez-Ortiz M, Martinez-Abundis E, Gonzalez Ojeda A. J Diabetes Complications. 2000 Nov-Dec;14(6):322–6.

The average serum glucose level decreased significantly from 210 mg/dL at baseline to 147 mg/dL at the end of the study in the experimental group. Total cholesterol and triglyceride levels also decreased significantly from baseline. There were no significant changes in metabolic profile in the control group. Thus, behaviormodifying education was more effective for providing metabolic control than traditional education.

The impact of a nurse-led, face-to-face, individually structured intervention to help smokers quit smoking was assessed in a 6month, randomized, controlled trial of 280 patients with type 1 or 2 diabetes at two urban hospitals and 15 urban primary care centers. The intervention involved a 40-minute interview, optional transdermal nicotine replacement therapy, and a follow-up support program. In the interview, the advantages of smoking cessation were emphasized, a cessation date was negotiated with the patient, and self-help materials were provided. The intervention was based on the protocols established in How to Help Your Patients Stop Smoking: A National Cancer Institute Manual for Physicians. Subjects assigned to the control group received routine care, including advice to quit smoking. After 6 months, the incidence of smoking cessation (validated by urine cotinine concentrations) was 7.5 times higher in the intervention group (17.0%) than in the control group (2.3%). In patients who failed to quit smoking, the mean number of cigarettes smoked daily decreased significantly from baseline in both groups but the reduction was greater in the intervention group (4.6 cigarettes/day) than in the control group (1.6 cigarettes/day). Thus, the nurse-led intervention was effective in modifying smoking behavior in patients with diabetes. Diabetes care organization, process, and patient outcomes: effects of a diabetes control program. Carlson A, Rosenqvist U. The Diabetes Educator. 1991;17:42–8. The effects of a diabetes control program with continuing medical education for health care providers and organizational change on several organization, process, and outcome measures in patients with diabetes were evaluated in an 18-month, randomized, controlled study. Thirty-four primary health care centers in Sweden were randomized to the intervention (the diabetes control program) group or a control group. A random sample of 20% of the patients in a registry of patients with diabetes at each center was selected for study. Routine measurement of glycosylated hemoglobin and performance of eye examinations were significantly higher in the intervention group than in the control group. There were no significant differences between the two groups in access to or continuity of care, patient education, or dietary knowledge. Self-monitoring of blood glucose was significantly more common in the intervention group than in the control group, although the degree of metabolic control was similar in the two groups. Multivariate analysis by multiple

Disease Management for Diabetes

The effects of a behavior-modifying educational program on the metabolic profile (serum glucose, total cholesterol, and triglyceride levels) were assessed in a randomized, controlled trial of 49 patients with type 2 diabetes who were not using insulin. Both the experimental group and control group attended classes once a week for 9 months. The behavior-modifying educational program was designed to improve nutritional habits. Classes addressed basic and applied nutrition concepts and involved workshops, teamwork, and discussions. A traditional educational intervention (i.e., didactic instruction in anatomy and physiology and memorization of nutritional facts) was used for the control group.

Intervention study for smoking cessation in diabetic patients: a randomized controlled trial in both clinical and primary care settings. Canga N, De Irala J, Vara E, Duaso MJ, Ferrer A, MartinezGonzalez MA. Diabetes Care. 2000 Oct;23(10):1455–60.

[15]

Appendix B. Reports of the Impact of Disease Management Interventions on Treatment of Diabetes (continued) regression revealed that routine diabetes care and patient-rated quality of care were related to patient education, which in turn was related to dietary knowledge, self-care practices, and metabolic control. The link between metabolic control and organization and process variables was weak. Group visits in medically and economically disadvantaged patients with type 2 diabetes and their relationships to clinical outcomes. Clancy DE, Brown SB, Magruder KM, Huang P. Top Health Inf Manage. 2003;24(1):8–14. The use of group visits (the intervention) was compared with usual care in a randomized controlled study of 120 adults with poorly controlled type 2 diabetes (glycosylated hemoglobin >8.5%) at a university-based clinic. Most patients were female and African American, with a mean self-reported educational level of 10.6 years and an average health literacy level of grade 7.5. Most patients lacked insurance (30%) or were covered by Medicaid (44%) or Medicare (19%). In the intervention group, groups of 19 or 20 patients met monthly for 6 months. Each 2-hour group visit included a presentation on a health-related topic (e.g., nutrition, exercise, sick day management, stress management), a question-and-answer session, and individual attention to patient needs (e.g., immunizations). A control group received usual care.

Disease Management for Diabetes

After 6 months, glycemic control and lipid profiles had improved in both groups, with no significant differences between groups. Patients in the intervention group had significantly higher scores in the Trust in Physician Scale, a measure of patients’ sense of trust in their health care providers, than patients in the control group. The use of the Patient Care Assessment Tool (a validated measure of patient satisfaction with various aspects of primary care) revealed trends toward improved coordination of care, community orientation, and cultural competence in the intervention group, although the differences between the two groups were not significant. Significantly greater compliance with American Diabetes Association standards of medical care process indicators (e.g., annual performance of eye examinations) was observed in the intervention group than in the control group at the end of the study.

[16]

Management of patients with type 2 diabetes by pharmacists in primary care clinics. Coast-Senior EA, Kroner BA, Kelley CL, Trilli LE. Ann Pharmacother. 1998;32:636–41. The impact of clinical pharmacists working as part of a multidisciplinary team on glycemic control in patients with type 2 diabetes who require insulin was evaluated. The study was conducted at two primary care clinics at a university-affiliated Veterans Affairs medical center. Primary care providers referred 23 patients with inadequate glycemic control to the pharmacists, who provided patient educa-

tion, medication counseling, monitoring, and insulin initiation or adjustment. Goals for glycemic control were in accordance with American Diabetes Association standards. Patients met with a pharmacist for an initial 30- to 60-minute visit during which the patient’s knowledge of diabetes, eating patterns, and glycemic control were assessed and the patient’s medical history and medications were reviewed. Follow-up visits were scheduled on an individualized basis at approximately 1-month intervals. Glycemic control and symptoms were assessed and insulin was adjusted as needed at these visits. Patients also were monitored by telephone at 1- to 4-week intervals, depending on glycemic control. Appointments were scheduled with the primary care provider at least every 6 months. Pharmacists also arranged for visits with physicians, dietitians, social workers, and psychologists as needed. Patients were followed for up to 45 weeks (27.1 weeks on average). The average glycosylated hemoglobin decreased from 11.1% at baseline to 8.9% at the last follow-up visit, a change that is significant. Six (26%) patients achieved a value in the target range (less than 8%). Average fasting blood glucose concentrations and average random blood glucose concentrations also decreased significantly from baseline. Ten (43%) patients achieved a fasting blood glucose value in the target range (140 mg/dL or less), and 19 (83%) patients achieved a random blood glucose value in the target range (180 mg/dL or less). Eight (35%) patients experienced 85 episodes of symptomatic hypoglycemia (one patient who failed to adhere to recommendations for diet, alcohol use, and exercise accounted for 48 of these episodes). None of the patients was seen at the emergency department or hospitalized. Thus, improved glycemic control was associated with pharmacists’ efforts as part of a multidisciplinary team. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulindependent diabetes mellitus. The Diabetes Control and Complications Trial Research Group. N Engl J Med. 1993;329:977–86. In what was a landmark study, 1,441 patients with type 1 diabetes were followed for an average of 6.5 years to compare the impact of intensive therapy (an external insulin pump or three or more daily insulin injections and frequent blood glucose monitoring) with that of conventional therapy (one or two daily insulin injections) on microvascular complications (retinopathy and nephropathy) and neuropathy. Glycosylated hemoglobin and blood glucose levels were significantly lower in the intensive therapy group than in the conventional therapy group. Intensive therapy reduced the risk for developing retinopathy by 76% (compared with conventional therapy) in the 726 patients with no retinopathy at baseline. Intensive therapy slowed the progression of retinopathy by 54% and reduced the development of proliferative or severe nonproliferative retinopathy by 47% in the 715 patients with mild retinopathy at baseline. Intensive therapy also reduced the occurrence of microalbuminuria and albuminuria (two measures of nephropathy) by 39% and 54%,

Appendix B. Reports of the Impact of Disease Management Interventions on Treatment of Diabetes (continued) respectively, and reduced the occurrence of clinical neuropathy by 60%. The development of hypercholesterolemia (elevated serum concentrations of low-density lipoprotein cholesterol) was significantly lower in the intensive therapy group compared with the conventional therapy group. The risk of macrovascular disease (i.e., major cardiovascular and peripheral vascular events) was reduced by 41% by intensive therapy, although the difference between the intensive therapy group and the conventional therapy group was not significant. The risk of severe hypoglycemia was increased approximately three-fold by intensive therapy compared with conventional therapy, a difference that is significant. Diabetes managed care and clinical outcomes: the Harbor City, California, Kaiser Permanente Diabetes Care System. Domurat ES. Am J Manag Care. 1999;5:1299–1307. The impact of Diabetes Care System (DCS), a computer-supported diabetes care management program, on clinical outcomes in a California health maintenance organization (HMO) was assessed using data from a 3-year period. Approximately 30% of HMO members with diabetes were targeted by DCS because of factors that placed them at high risk for disease complications or problems. Care for these patients was provided by a team of health care professionals, and data from individual visits, group encounters, and telephone contacts were tracked in a computer database. The other 70% of patients with diabetes received usual care. Goals for blood glucose and blood pressure control were established in accordance with American Diabetes Association clinical practice recommendations.

The impact of implementation of a diabetes disease management program at a rural not-for-profit health plan was evaluated by performing a chart review. A subset of 3,118 of 6,799 health plan enrollees with diabetes participated in the disease management program; participants were referred by a physician or self-referred. The program involved one to four annual visits with a nurse in the primary care setting (depending on the presence of diabetes complications) and patient education by nurse educators about selfmanagement techniques and preventive care. Treatment guidelines were developed by physicians using Health Plan Employer Data and Information Set (HEDIS) measures that establish frequencies for various tests (e.g., glycosylated hemoglobin, lipids, kidney function, eye exams). Nurse educators promoted guideline use in the course of providing patient care. Participants in the disease management program had higher rates of testing (i.e., higher HEDIS scores); better glycemic control (after only 3 months); and fewer emergency department visits and inpatient days than nonparticipants. An annual cost savings of $4 million (nearly $1,300 per participant) was realized, which more than offsets the program cost of $1.8 million. A project to reduce the burden of diabetes in the AfricanAmerican Community: Project DIRECT. Engelgau MM, Narayan KMV, Geiss LS, et al. J Natl Med Assoc. 1998;90:605–13. Project DIRECT (Diabetes Interventions Reaching and Educating Communities Together) is a 5-year community-based demonstration project that targets African Americans because of the high prevalence of diabetes and risk for complications from the disease in this patient population. It will be a collaborative effort involving community leaders, health care providers, and the Centers for Disease Control and Prevention. The project comprises (1) health promotion initiatives to improve diet and physical activity in patients with diabetes; (2) outreach efforts to increase community awareness of diabetes risk factors, screening for patients with undiagnosed diabetes, and integration of previously diagnosed patients into the health care system; and (3) efforts to improve diabetes selfcare practices, increase access to care, and quality of care provided by primary care providers. The interventions are not based on specific practice guidelines. Approximately 4,400 households will be contacted in each of two communities in North Carolina with a large population of African Americans and similar socioeconomic profiles and health care resources. One community (Raleigh) will participate in the project, and the other community (Greensboro) will serve as a control. Participation in programs, screening rates, changes in diabetes practice, glycosylated hemoglobin concentration (i.e., long-term blood glucose control), and patient knowledge and skills are among the process and outcome measures that will

Disease Management for Diabetes

Screening rates for glycosylated hemoglobin (a measure of longterm glycemic control), urinary protein, and serum lipids were significantly higher in the 2,617 patients managed by DCS compared with the 5,993 patients receiving usual care. Similarly, the rate of follow-up testing in patients with an elevated glycosylated hemoglobin was significantly higher in the DCS patients than in the usual care patients (77% versus 44%, respectively). Significant reductions in glycosylated hemoglobin levels were observed in patients in both groups with initial elevations. There was a significant reduction in blood pressure in the DCS patients with an elevated blood pressure but no change in blood pressure in usual care patients with elevated blood pressure. Inpatient utilization decreased significantly between 1995 and 1997 in a subset of 386 patients managed through the DCS program during that period and in 1997 was significantly lower than in 287 patients who were discontinued from the DCS program because of patient factors and resource limitations. Screening rates for glycosylated hemoglobin, urinary protein, and serum lipids in patients who were no longer managed through DCS were significantly lower than rates in patients managed through DCS on a long-term basis. Thus, computerized tracking of patient care improved clinical outcomes.

Disease management proves itself at Geisinger. Edlin M. Healthplan. 2003;44(4):55–8.

[17]

Appendix B. Reports of the Impact of Disease Management Interventions on Treatment of Diabetes (continued) be evaluated. Many of the data will be collected by questionnaire. In a 1993 pilot study, 45% of 902 participating individuals were African American. Half of these African Americans were physically inactive (light or no physical activity during most weeks) and half were overweight. The prevalence of diagnosed and undiagnosed diabetes was higher in African Americans than in members of other races. African Americans also were more likely to smoke and have uncontrolled hypertension and less likely to have a primary health care provider. In patients of all races with diabetes, the level of preventive care was low; only 42% had undergone a diabetes eye examination and 50% had their feet examined by a health care provider within the previous year. Diabetes support groups improve health care of older diabetic patients. Gilden JL, Hendryx MS, Clar S, Casia C, Singh SP. J Am Geriatr Soc. 1992;40:147–50.

Disease Management for Diabetes

A study of 32 elderly (68 years of age on average) male patients with diabetes was conducted to determine whether diabetes knowledge, quality of life, and glycemic control are affected by providing support group sessions in conjunction with a diabetes education program. The diabetes education program comprised six weekly sessions on the disease, nutrition, drug management, psychosocial adjustment, stress management skills, and self-care (e.g., self-monitoring of blood glucose, foot care). The 11 patients in Group A participated in the diabetes education program followed by an 18-month support group program with monthly sessions consisting of continuing education, coping skills training, group discussions, and structured social activities. Support group sessions were self-directed by patients and supervised by a social worker, with continuing education provided by members of the diabetes care team (a diabetologist, nurse educator, dietitian, social worker, psychologist, podiatrist, and pharmacist) as appropriate. The 13 patients in Group B participated only in the diabetes education program (they were on a waiting list for the support group program). Eight patients in Group C participated in neither the diabetes education program nor the support group program.

[18]

Patients who received both the support group intervention and diabetes education program (Group A) had significantly better scores in measures of diabetes knowledge, quality of life, and depression than patients in the other two groups at the end of the study. Glycemic control (fasting blood glucose and glycosylated hemoglobin) was significantly worse in Group C (patients who did not participate in either program) than in patients in the other two groups but there was no significant difference between Group A and Group B. At the 2-year follow-up, the patients who participated in the support group program (Group A) were more successful in maintaining diabetes knowledge and had a better quality of life than patients in Group B who received diabetes education without group support.

A community-based case management model for hypertension and diabetes. Ginn M, Frate DA, Keys L. J Miss State Med Assc. 1999 July; 40(7):226–8. The effect of a community-based case management approach on disease knowledge, self-perceived health status, and actual health status in 754 patients with diabetes, hypertension, or both was assessed in a 24-month uncontrolled study. A case management program involving client assessment, care plan development, implementation, client service monitoring, and periodic assessment was used at 12 ambulatory clinics and one hospital. These facilities were located in five rural Mississippi counties with barriers to health care, including a lack of financial resources and insurance, inadequate distribution of primary care providers, transportation limitations, and low educational levels. Case managers were hired from the local community and trained in methods for educating and managing patients. Knowledge about diabetes and hypertension and self-reported health status improved significantly over the course of the study. The percentage of patients with good blood pressure control (systolic blood pressure <140 mm Hg and diastolic blood pressure <90 mm Hg) increased from 33.5% at baseline to 47% after 24 months, a difference that is significant. The percentage of patients with good blood glucose control (60–120 mg/dL) also increased (from 13.9% to 18.9%) over this period, although the change from baseline was not significant. Brief, computer-assisted diabetes dietary self-management counseling: effects on behavior, physiologic outcomes, and quality of life. Glasgow RE, Toobert DJ. Med Care. 2000 Nov;38(11):1062–73. The reach (percentage of eligible patients participating), adoption (willingness of primary care physicians to participate), and effectiveness of a behavioral dietary intervention with or without diabetes self-management support were assessed in 320 patients with type 2 diabetes in a 6-month study. The behavior intervention involved assessment and analysis of dietary habits and barriers to dietary self-management and establishment of goals for reducing dietary fat intake using a multimedia touch-screen computer. Diabetes self-management support involved follow-up telephone calls and community resources enhancement (e.g., newsletters, information on dining out and shopping). Seventy-six percent of eligible patients and 40 of 42 primary care physicians agreed to participate. On average there was a 50% reduction in dietary fat intake. However, there were small reductions in glycosylated hemoglobin A1c and the ratio of total cholesterol to high-density lipoprotein cholesterol (a predictor of heart disease) that were not significant. There was relatively little change in illness-related quality of life. Follow-up phone calls and community

Appendix B. Reports of the Impact of Disease Management Interventions on Treatment of Diabetes (continued) resources enhancement did not add to efficacy of the behavioral intervention. A brief office-based intervention to facilitate diabetes dietary self-management. Glasgow RE, Toobert DJ, Hampson SE. Diabetes Care. 1996 Aug; 19(8):833–42. There is a pressing need for brief practical interventions that address diabetes management. Using a randomized design, a medical office-based intervention focused on behavioral issues relevant to dietary self-management was evaluated. There were 206 adult diabetes patients randomized to usual care or brief intervention, which consisted of touch-screen computer-assisted assessment to provide immediate feedback on key barriers to dietary selfmanagement, and goal-setting and problem-solving counseling for patients. Follow-up components to the single session intervention included phone calls and interactive video or videotape instruction as needed. Results using multivariate analyses of covariance revealed that the brief intervention produced greater improvements than usual care on a number of measures of dietary behavior (e.g., fewer calories from saturated fat, fewer high-fat eating habits and behaviors) at the 3-month follow-up. There were also significant differences favoring intervention on changes in serum cholesterol levels and patient satisfaction but not on glycosylated hemoglobin. The intervention effects were relatively robust across a variety of characteristics, the two participating physicians, and intervention staff members. If the results of this study were to be generalized to other settings, this intervention could provide a prototype for a feasible cost-effective way to integrate patient views and behavioral management into office-based care for diabetes. A brief office-based intervention to facilitate diabetes dietary self-management. Glasgow RE, Toobert DJ, Hampson SE, Noell JW. Health Educ Res. 1995;10:467–78.

The intervention used computer-assisted and interactive video procedures that assess patient views and behaviors related to diabetes care and self-management; assistance for patients in setting dietary goals and problem-solving strategies; and follow-up phone

Follow-up components include phone calls and videotape or interactive video instruction as needed. Initial process results suggest success in producing modest, targeted behavior changes among a broad cross section of patients. If long-term results are equally positive, this intervention could provide a prototype for a feasible, cost-effective way to integrate patient views and behavioral management into office-based care for diabetes. Outcomes measured include weight, eating patterns (fat content), serum cholesterol, and glycemic control (glycosylated hemoglobin) and will be measured but are not yet available. Coping skills training for youths with diabetes on intensive therapy. Grey M, Boland EA, Davidson M, Yu C, Tamborlane WV. Applied Nursing Research. 1999;12:3–12. A 6-month randomized, controlled study was conducted to evaluate the impact of coping and skills training (CST) on metabolic control in adolescents with type 1 diabetes who were initiating intensive insulin therapy. Seventy-seven patients selected from a universitybased pediatric diabetes service were randomized to the intervention group, which received CST and intensive insulin therapy, or the control group, which received intensive insulin therapy alone. Patients were admitted for 1 day to a research center to obtain baseline data and review treatment goals and methods. The health care providers were blinded to the study group assignment of patients. Intensive insulin therapy involved three or more daily insulin injections or the use of an external insulin pump and selfmonitoring of blood glucose at least four times daily. CST was designed to teach problem-solving skills and communications. Four to eight 1- to 1.5-hour weekly small-group sessions followed by monthly sessions were led by a master’s-prepared nurse practitioner with experience in pediatric psychiatry and diabetes. Roleplaying in various social situations (e.g., making food choices with friends and decisions about drugs and alcohol) was used in these sessions. Both the intervention group (intensive insulin therapy plus CST) and the control group (intensive insulin therapy alone) had a significant reduction in glycosylated hemoglobin after 6 months, although the reduction was faster and greater in the intervention group. CST neither reduced nor increased the rate of acute complications of intensive therapy (i.e., severe hypoglycemic episodes). Patients in the intervention group had significantly better general self-efficacy (personal competence, power, resourcefulness) and diabetesrelated quality of life, reported that diabetes had less of a negative impact on quality of life, and had fewer worries about diabetes in relation to quality of life than patients in the control group.

Disease Management for Diabetes

An office-based intervention to prompt both patients and providers to focus on behavioral issues relevant to dietary self-management was developed for approximately 200 adult patients with type 1 or 2 diabetes who are at least 40 years of age (data for the initial 95 are presented in this article). Patients were identified via letter about the study, with informed consent statements sent to patients who have diabetes and a scheduled appointment at an officebased practice, followed by a phone call from an investigator.

contact and office visits for support. Diet was chosen as a focus because it is an objective of Healthy People 2010. An endocrinologist and internist who specialize in diabetes were the managers of the intervention.

[19]

Appendix B. Reports of the Impact of Disease Management Interventions on Treatment of Diabetes (continued) Effect of multiple patient reminders in improving diabetic retinopathy screening. Halbert RJ, Leung KM, Nichol JM, Legorreta AP. Diabetes Care. 1992 May;22(5):752–5. The impact of multiple patient reminders on improving the rate of diabetic retinopathy screening was evaluated in 19,523 diabetic members of a large, network-based, health maintenance organization (HMO). Claims and pharmacy databases were used to identify all HMO members ≥18 years with diabetes and their diabetic retinal examination (DRE) status. Patients were then randomized to single (N=9614) or multiple (N=9909) reminder intervention groups. Both groups received mailed educational materials and a reminder for DRE. Their treating physician groups also received a letter explaining the program, current American Diabetes Association (ADA) guidelines for retinal examinations, and a list of their diabetic patients, indicating DRE status. The single intervention group (the control group) received no additional reminders, while the multiple intervention group (the study group) received additional reminders 3, 6, and 9 months after baseline if the claims database indicated no record of DRE. Monthly DRE rates were determined for the 7 months preceding the first intervention (patient reminder) and subsequent 12 months.

Disease Management for Diabetes

There was a short-lived (~ 2-month) increase in monthly DRE rates in both intervention groups following the initial intervention (first reminder). After a second reminder was sent to the multiple intervention group, the percentage of patients who underwent DRE was higher than that in the single intervention group. However, DRE rates between groups did not significantly differ after the third reminder, nor did monthly differences in rates exist. There was a significant difference in overall annual DRE rates between groups (p=0.023). Analysis of direct costs of reminders (only printing and postage) in the multiple reminders group suggested an incremental cost of $80 per additional eye examination.

[20]

Multiple patient reminders are more effective than single reminders in improving rates of diabetic retinopathy screening in a managed care setting. However, incremental improvement is small and does not extend beyond a second reminder. Future research should explore other avenues of improving DRE rates (e.g., telephone follow-up of nonresponders, increased provider involvement, provider reminders). Given the costs of multiple reminders, financial resources may be better utilized on other approaches for reducing diabetic complications. Evaluation of a clinical pharmacist in caring for hypertensive and diabetic patients. Hawkins DW, Fiedler FP, Douglas HL, Eschbach RC. Am J Hosp Pharm. 1979;36:1321–5. A 29-month randomized, controlled study of 1,722 patients with diabetes, hypertension, or both was conducted to determine whether the outcomes of care provided by a clinical pharmacist (an

individual with a doctor of pharmacy degree and 2 years of clinical training in general medicine) differ from the outcomes of care provided by a physician. The pharmacist-managed care was closely monitored by a physician to ensure that care would be adequate, although care plans written by the pharmacist were rarely altered. The attrition rate was significantly lower and the kept-clinic-appointment rate was significantly higher in the experimental group (pharmacist-managed care) than in the control group (physician-managed care). These differences may reflect greater patient satisfaction with pharmacist-managed care. There was no significant difference between the two groups in compliance with drug therapy, emergency department visits, hospital admissions, fasting blood glucose level, or diastolic blood pressure. Systolic blood pressure was significantly lower in the control group compared with the experimental group, possibly in part because the control group had a larger percentage of patients with both diabetes and hypertension than the experimental group (diabetes contributes to vascular changes and hypertension). Care provided by the clinical pharmacist was safe and effective. Is the quality of diabetes care better in a diabetes clinic or in a general medicine clinic? Ho M, Marger J, Beart J, Yip I, Shekelle P. Diabetes Care. 1997;20:472–5. A retrospective chart review was conducted to compare the quality of ambulatory diabetes care delivered by specialists (faculty diabetologists, endocrine fellows, medical residents, a diabetic nurse educator, a podiatrist, and an optometrist) at a diabetes mellitus clinic (DMC) with that provided by generalists (faculty internists, medical fellows, medical residents, and nurse practitioners) at a general medicine clinic (GMC) in 1993 and 1994. Computerized medication profiles and clinic enrollment lists were used to identify 112 patient medical records, half of which documented care provided in the DMC and half of which reflected care delivered at the GMC. Process criteria for “good” quality of diabetes care were developed using American Diabetes Association guidelines. A subset of minimally acceptable criteria was chosen because of a strong link with good patient outcomes in clinical trials (e.g., blood pressure and glycosylated hemoglobin measurements, urinalysis) or because of their importance for continuity of care. Compliance with criteria for self-monitoring of blood glucose levels, foot examination, comprehensive eye examination, and referral for diabetes education when glycemic control was poor was significantly greater in the DMC patient medical records than in the GMC records. Compliance with the criterion for inquiry about cardiac symptoms was significantly higher in the GMC than in the DMC. None of the medical records from either clinic met all the criteria for “good” quality of care. Significantly more DMC patient medical records than GMC records met the minimally acceptable criteria (73% versus 52%, respectively). The most common shortcoming for both clinics was failure to perform a comprehensive eye examination. Differences between specialists and generalists in knowledge about diabetes management could explain these results.

Appendix B. Reports of the Impact of Disease Management Interventions on Treatment of Diabetes (continued) Empowering diabetes out-patients with structured education: short-term and long-term effects of functional insulin treatment on perceived control over diabetes. Howorka K, Pumprla J, Wagner-Nosiska D, Grillmayr H, Schlusche C, Schabmann A. J Psychosom Res. 2000 Jan;48(1):37–44. The short- and long-term effects of structured education about functional insulin treatment (insulin dosing according to self-monitoring of blood glucose levels and food intake, thereby avoiding the need for adherence to a prescribed schedule of doses and meals) were studied in outpatients with type 1 diabetes. Study 1 was a 4week randomized, controlled study of 32 patients. Both the intervention group and the control group received general diabetes education but only the intervention group received training in functional insulin treatment (FIT). Patients undergoing FIT training learned criteria for adapting insulin dosing before meals and for the correction of hyperglycemia. Study 2 was a 3-year uncontrolled pilot study of 68 patients. The effect of training in FIT on perceived control over diabetes and diabetes-related health beliefs was assessed in both studies using questionnaires. In study 1, FIT training induced feelings of independence from situational control while self-managing diabetes. There was no change in the control group over the 4-week period. After 3 years, FIT training improved perceived self-efficacy, treatment satisfaction, and glycemic control, resulting in a feeling of empowerment. Patients felt greater freedom from the control of physicians and treatment restrictions. Perceived treatment cost-effectiveness (the difference between measures of benefits from treatment and barriers to treatment) improved significantly due to decreases in barriers to treatment. Pharmacist-managed diabetes education service. Huff PS, Ives TJ, Almond SN, Griffin NW. Am J Hosp Pharm. 1983;40:991–4.

Pharmacists were able to provide more instructional time than typically is provided by physicians, thereby improving patient understanding. Patients were grateful to have ready access to pharmacists for information or help solving problems. Physicians had more time available to spend with other patients once pharmacists assumed the patient education responsibility. Communication between pharmacists and physicians improved. Impact of community-based diabetes education on program attenders and nonattenders. Irvine AA, Mitchell CM. The Diabetes Educator. 1992;18:29–33. The impact of community-based diabetes education on patients with diabetes who dropped out of the program was compared with that on patients who continued in the program. In three Appalachian communities that were comparable in size and character, adults with diabetes were referred by physicians and sent a letter inviting them to participate in the study. Sixty-one patients attended three initial evening meetings at a local church over the course of 1 week. The diabetes education program was designed to increase knowledge, self-care, and metabolic control and was conducted by a nurse educator, nutritionist, psychologist, and physical therapist. Classes were held on alternate Thursdays over a 3-month period on various topics, including the disease, diet, stress, medication, blood glucose testing, complications, and exercise. The 24 subjects who attended fewer than half (two or fewer) of the education classes were considered the Attrition group. The other 19 subjects, who attended three or more of the education sessions, were considered the Education group. A control group of 18 patients received no special education. This control group was further divided on the basis of whether subjects attended all three testing sessions. Data for members of the control group who did not attend all three testing sessions (pretest, posttest, and 3month follow-up) were combined with data for the Attrition group to create a Nonattenders group, and data for members of the control group who attended all three testing sessions were combined with data for the Education group to form an Attenders group. Diabetes knowledge increased significantly over the course of the study in the Attenders group but not in the Nonattenders group. There was no significant difference between the two groups in change in attitude toward living with diabetes (self-esteem, denial, locus of control, degree to which regimen intrudes on lifestyle, attitude toward physician). Attitudes improved over the course of the study in both groups. There was no significant change over time in any group in self-reported adherence to the treatment regimen for diet, exercise, medication use, glucose testing, or number of pounds overweight. Foot care improved significantly in Attenders but not in Nonattenders. Glycosylated hemoglobin improved significantly in the Education group but not in the Control group.

Disease Management for Diabetes

Ten years of experience with a diabetes education service provided by pharmacists to ambulatory patients was described in 1983. Funding was provided by the U.S. Public Health Service of the Department of Health and Human Services, and pharmacists were reimbursed for the service. The education program was developed in accordance with American Diabetes Association patient education guidelines. Patients were referred to the service by a primary care provider after diagnosis. A diabetes questionnaire was administered to assess the patient’s knowledge of diabetes. The questionnaire was used to establish a teaching plan and short- and long-term treatment goals. An initial 1-hour counseling session was conducted to address the pathophysiology of the disease, complications, importance of patient compliance, role of diet and drug therapy, administration technique, adverse effects, and selfmonitoring. Patient progress in the educational process was documented. Additional 15-minute sessions at follow-up visits were provided as needed. Patients were assessed a fee of $15.00 for

the initial 1-hour session and $7.50 for additional sessions. Approximately 10 new patients joined the service each month.

[21]

Appendix B. Reports of the Impact of Disease Management Interventions on Treatment of Diabetes (continued) Nonattenders tended to be less educated, have a lower income, be younger, have had diabetes for twice as long, report more barriers to self-care, and have poorer health than Attenders. The results of this study suggest that Attenders may have characteristics that allow them to benefit from minimal intervention and that Nonattenders may require special efforts to overcome barriers to participating in programs. Evaluation of pharmaceutical care model on diabetes management. Jaber LA, Halapy H, Fernet M, Tummalapalli S, Diwakaran H. Ann Pharmacother. 1996;30:238–43. This study assessed the effectiveness of a comprehensive pharmaceutical care model in improving management of type 2 diabetes in 39 African American patients in a clinic setting. Potential subjects consisted of urban African American patients with type 2 diabetes who were currently attending a university-affiliated, internal medicine outpatient clinic. Medical chart review was used to exclude patients with certain medical conditions or a recent history of clinic visit noncompliance; the remaining 156 (eligible) patients were contacted by letter followed by a telephone call. The 45 patients who agreed to participate were assigned to an intervention group or control group in a random, parallel design fashion. These patients were followed up over a period of 4 months, and 39 patients (17 intervention-group patients; 22 controls) completed the study.

Disease Management for Diabetes

Patients in the intervention group received pharmaceutical care consistent with, but broader than, that described by Helper and Strand’s disease-specific model of pharmaceutical care. This pharmacist-administered care consisted of (1) disease-specific pharmacotherapeutic evaluation and dosage adjustments, (2) comprehensive, individualized patient education about diabetes and its complications, (3) training in the recognition and treatment of hypoglycemia and hyperglycemia, (4) medication counseling, (5) specific instructions about diet and exercise, and (6) training in the self-monitoring of blood glucose (SMBG). Follow-up visits were conducted weekly until targeted glycemic control was reached and then every 2 to 4 weeks for the study duration. Patients in the control group attended an initial assessment at the clinic and an exit interview at the study’s conclusion; they were instructed to continue with standard care, which usually involved clinic visits every 3 to 4 months.

[22]

Clinical, laboratory, and quality of life assessments were performed at baseline and at the end of the study (i.e., at 4 months). Primary outcome measures consisted of fasting plasma glucose (FPG) and glycosylated hemoglobin (GHb). Secondary outcome endpoints included blood pressure, lipid levels, and measures of renal function as well as quality of life. Information about compliance with medications, diet, and exercise, as well as the frequency of hypo- and hyperglycemic episodes, was obtained at each clinic visit. In addition, SMBG records were reviewed and FBG, weight, and blood pressure were measured.

Post-intervention evaluation revealed that glycemic control significantly improved in the intervention group and remained unchanged in the control subjects. Final GHb and FPG concentrations in the intervention group were significantly decreased compared with baseline, and significant intergroup differences existed for final GHb, FPG, and the mean absolute change in GHb. No significant changes in blood pressure control, lipid profiles, renal function parameters, body weight, or quality-of-life measures were documented within or between groups, except for a decrease in the final serum creatinine in controls. There were 17 reported hypoglycemic events in the intervention group versus two in the control group; all were mild to moderate, accompanied by classic symptoms, and recognized and successfully treated by patients. These results suggest that this disease-specific model of pharmaceutical care effectively improved glycemic control in a clinic-based population of urban African American, diabetic patients. Given the lack of change in other parameters (e.g., body weight, lipids, blood pressure), this improvement was primarily attributed to optimization of glycemic control and enhancement of patient knowledge versus instruction about diet and exercise. Cost analysis of this program was not performed. Effects of diet and exercise interventions on control and quality of life in non–insulin-dependent diabetes mellitus. Kaplan RM, Hartwell SL, Wilson DK, Wallace JP. J Gen Intern Med. 1987;2:220–7. An 18-month randomized trial evaluated the effects of diet, exercise, diet plus exercise, or education on glycosylated hemoglobin, weight, and quality of life in 70 diabetic patients. Following recruitment by radio announcements, newspaper notices, and physicians, adult volunteers with a confirmed diagnosis of type 2 diabetes and a fasting plasma glucose of > 3.62 mmol/L were randomly assigned to one of four 10-week programs: (1) diet, (2) exercise, (3) diet plus exercise, or (4) education (control). Participants in all groups were given both the American Diabetes Association-recommended exchange diet (approximately 1,200 calories/day) and an test-based exercise prescription and asked to attend weekly 2-hour meetings for 10 consecutive weeks. Volunteers in the education group, designed to be a control group, attended lectures by health care specialists that provided information about diabetes care but not specific instructions for behavioral changes. Subjects in the diet, exercise, and diet-plus-exercise intervention groups attended weekly sessions that incorporated behavioral modification methods (e.g., diaries to monitor eating/exercise; identification of reinforcers and contingencies; use of distractors) and other strategies (e.g., instruction in stretching and foot care) to increase compliance with diet and/or exercise plans. Baseline parameters were similar among groups. Compared with controls, the diet group demonstrated weight loss at 3 and 6 months, but had regained this weight by the 18-month evaluation. The greatest reduction in mean glycosylated hemoglobin at 18

Appendix B. Reports of the Impact of Disease Management Interventions on Treatment of Diabetes (continued) months was found in the diet-plus-exercise group (p<0.05), whose mean weight remained essentially unchanged throughout the study. Significant differences in quality of life were noted between the dietplus-exercise and control groups (p<0.01) and the diet and control groups (p<0.05), but not the exercise and control groups, at 18 months. The estimated cost of the diet and exercise program was $1,000 per patient. Quality of life-based cost/utility analysis suggested the cost of a single “well year” produced by this program would be $10,870. Effects of a formalized diabetes education. Karlander SG, Kindstedt K. Acta Med Scand. 1983;213:41–3. The effects of a formalized diabetes education program on diabetes and nutrition knowledge and metabolic control were assessed in a 1-year controlled study of ambulatory patients with diabetes. All patients attending a diabetes clinic were invited to participate, although another 32 patients matched for age were not asked to participate and served as controls. Two daily lessons about the disease and nutrition, respectively, were conducted on each day of the 5-day program. Oral presentations were supplemented with color slides, and patients were encouraged to ask questions and participate in discussions. An illustrated text that summarized each lesson was provided.

Effect of a comprehensive nurse-managed diabetes program: an HMO prospective study. Legorreta AP, Peters AL, Ossorio RC, Lopez RJ, Jatulis D, Davidson MB. Am J Manag Care. 1996;2:1024–30. This 18-month prospective, nonrandomized, control trial assessed whether a diabetes management program involving special protocols linked to a computer system could provide cost-effective out-

A provider team at each experimental site, consisting of a nurse or physician assistant (PA), an endocrinologist, and a staff assistant, received a training manual, specially designed protocols and forms, and a computer information system. Protocols were designed to improve diabetes and lipid management as well as compliance with referrals and laboratory studies. The computer system was designed to (1) track patient clinical and laboratory data, (2) track patient appointments and generate reminders, and (3) provide algorithms to adjust insulin dosages based on results of blood glucose self-monitoring. Training in the application of the program lasted 2 weeks for nurses/PAs and one day for physicians. These experimental teams also attended quarterly meetings with diabetologists to discuss program progress. Providers at control sites were not informed that a study was being carried out. As part of the program, all patients treated at experimental sites underwent an initial evaluation and were then seen at least quarterly. Telephone calls and weekly visits were used for patients requiring more frequent monitoring. The computer ordered baseline laboratory values (glycosolated hemoglobin, fasting plasma glucose, full lipid panel, serum creatinine, urinalysis); follow-up glycosolated hemoglobin and fasting plasma glucose were obtained every 2 months, and the remaining laboratory studies were repeated yearly. Data collection from controls was less complete to avoid alerting providers and patients that a study was being conducted. Only patients whose charts provided a glycosolated hemoglobin level within a one-year interval were selected as controls, and a single follow-up level was obtained at the end of the study. Provider team adherence to protocols was assessed by follow-up data obtained from 117 experimental patients at site A, 123 experimental patients at site B, 88 control patients for site A, and 62 control patients for site B (N=390). At site A, 12-month and baselineto-endpoint decreases in glycosolated hemoglobin were significantly greater in the experimental group than the control group. Lowdensity lipoprotein cholesterol levels also improved in the former, but were not measured in controls. Referral for yearly ophthalmologic examination was 100%. The same staff was employed for the

Disease Management for Diabetes

Eighty-three patients with diabetes participated in the formal education program. Their scores on tests of knowledge of diabetes and nutrition increased significantly immediately after the formal program and 1 year after the program compared with baseline. Patients in the control group tended to have lower scores than patients in the intervention group. Scores for the control group at the second visit (usually 2 months after the initial visit) were unchanged compared with the first visit. The only significant changes in metabolic control in the intervention group 1 year after the formal education program were a reduction in relative body weight from 133% to 126% in the subset of patients managed with diet alone and a decrease in serum cholesterol from 6.3 mmol/L to 5.8 mmol/L in the subset of patients receiving oral antidiabetic drug therapy. There were no significant changes in blood glucose level, urinary glucose excretion, glycosylated hemoglobin, or serum cholesterol or triglycerides. The investigators concluded that efforts to improve diabetes control must address patient attitudes toward the disease and not be limited to imparting facts.

patient diabetes care. Patients for this study were recruited from two large medical groups contracted to provide health care to HMO members. One of the groups (site A) was a typical participating medical group (PMG); the other site (site B) was an IPA. A single separate site was selected as a control for site A, and 13 nonexperimental physician office sites served as the control for site B. ICD codes were used to generate a list of patients with diabetes at each control and experimental site. Approximately 15 patients per month were randomly selected from this list to enter the program over the course of 6 months at experimental sites. Due to insufficient enrollment, physicians at each experimental facility were subsequently permitted to assign patients for inclusion. Patients in the control group were selected a few months after enrollment had begun at the experimental sites on the basis of an available glycosolated hemoglobin level.

[23]

Appendix B. Reports of the Impact of Disease Management Interventions on Treatment of Diabetes (continued) duration of the study, and nurse adherence to protocols was judged to be close. At site B, baseline-to-endpoint analysis, but not 12-month data, showed a significantly greater decrease in glycosolated hemoglobin levels in experimental group patients than controls. However, these levels remained above desirable levels. Improvement in low-density lipoprotein cholesterol levels was also found in the experimental group, but only 52% of patients at site B were referred for ophthalmologic examination. Considerable staff turnover was also found at site B, which was thought to have contributed to disruption of the program and the level of care provided. This disruption was associated with lower levels of protocol adherence and reduced proficiency with the computer systems compared with that for the staff at site A. Comparison of glycosolated hemoglobin data from experimental groups A and B also demonstrated a stronger treatment effect at site A. These results suggest that physician-supervised nurses who follow protocols linked to a computer system can provide high-quality outpatient diabetes care associated with improved glycemic control, lipid levels, and patient compliance with referrals. Differences in patient improvements between experimental groups as sites A and B were attributed to higher staff turnover at site B, resulting in program and care disruption. Impact of endocrine and diabetes team consultation on hospital length of stay for patients with diabetes. Levetan CS, Salas JR, Wilets IF, Zumoff B. Am J Med. 1995;99:22–8.

Disease Management for Diabetes

The length of hospital stay of consecutive patients with diabetes was compared when (1) consultation by a diabetes team (nurse educator, registered dietitian, and endocrinologist) was obtained, (2) a traditional consultation by an endocrinologist was obtained, and (3) no consultation was obtained. The average length of stay of patients in the diabetes team consultation group (3.6 days) was significantly shorter than values for the endocrinologist-consultation group (5.5 days) and the no-consultation group (8.2 days). Each day delay in consultation resulted in an increase of 1 day in length of stay.

[24]

Management of chronic pediatric diseases with interactive health games: theory and research findings. Lieberman DA. J Ambulatory Care Manage. 2001 Jan;24(1):26–38. A 6-month randomized controlled trial evaluated the effects of a diabetes self-management video game in 59 children and adolescents with diabetes. The diabetes video game had two main characters who were presented with challenges modeled after the challenges faced by patients with diabetes. Players were called upon to help their character monitor blood glucose, take appropriate insulin doses, and eat appropriate foods. Winning the game required success in performing these tasks. Multiple-choice questions posed by a video character were used to teach diabetes self-

care principles. Subjects were randomized to take home the diabetes video game or an entertainment pinball video game (subjects using the pinball game served as controls). Over the 6-month study period, the average total playing time for the diabetes video game was 34 hours, which was not significantly different from the playing time in the control group. There was a 77% decrease from baseline in urgent care and emergency medical visits in the diabetes video game group but there was no change in the control group (an average of 2.4 visits per year in the control group). There were improvements in diabetes-related self-efficacy (ability to affect outcomes), communication with parents about diabetes, and daily diabetes self-care in the diabetes video game group but not in the control group. Pharmacists’ interventions using an electronic medicationevent monitoring device’s adherence data versus pill counts. Matsuyama JR, Mason BJ, Jue SG. Ann Pharmacother. 1993;27:851–5. A randomized, double-blind study was conducted to compare the usefulness of adherence data from an electronic medication-event monitoring system (MEMS) with that of pill counts in assisting pharmacists in making recommendations for diabetes therapy (patient education or change in drug, dosage, or schedule). MEMS is a medication vial cap with a microprocessor that records each date and time that the vial is opened. Computer medication profiles and laboratory data for all patients with scheduled medicine clinic appointments at Veterans Affairs medical center ambulatory clinics in Boise, Idaho, were screened for patients receiving a sulfonylurea for 3 months or longer with a consistent dosage and poor to fair metabolic control between July 1990 and May 1991. Patients were randomized to the MEMS group or a control (pill count) group. Patients in both groups received a 35-day supply of medication in a vial with the MEMS cap and returned to the pharmacy for a refill after 30 and 60 days, at which time pill counts were performed. Initially, the data from the microprocessor were retrieved only for patients in the MEMS group, and pill count data were analyzed only for patients in the control group. There was no significant difference between the two groups in measured non-adherence rates (60% in the MEMS group and 35% in the control group). Seven (47%) of 15 pharmacist recommendations based on adherence data in the MEMS group and two (12%) of 17 recommendations in the control group were for patient education, a difference that is significant. To determine how consistent the two adherence measurement methods (MEMS and pill counts) were, MEMS data were later retrieved for the control group and compared with the pill count data, and pill count data were later analyzed for the MEMS group. The pill count data were inconsistent with the MEMS data for seven (47%) of the 15 patients in the MEMS group. The MEMS data were inconsistent with the pill count data for nine (53%) of the 17 patients in the control group. If the pharmacists had had access to the MEMS data for four of the patients in the control group, they would have recommended

Appendix B. Reports of the Impact of Disease Management Interventions on Treatment of Diabetes (continued) patient education. For two of these patients, this recommendation would have been made instead of a dosage increase because a pattern of inconsistent medication-taking behavior was evident with MEMS. These results demonstrate that MEMS allows pharmacists to individualize recommendations to a greater extent than the pill count method. Demonstrating the added value of community health nursing for clients with insulin-dependent diabetes. Mazzuca KB, Farris NA, Mendenhall J, Stoupa RA. J Community Health Nurs. 1997;14:211–24. A randomized, controlled study was performed to determine whether community health nursing (weekly or biweekly home visits to provide health teaching and guidance, health referrals, coordination of care, and client advocacy) improves the self-care competency and health status of adults with insulin-treated diabetes and poor glycemic control. Teaching addressed nutrition, exercise, foot care, and blood glucose monitoring. Patients who had received care at a university-based internal medicine clinic were contacted by telephone by investigators and invited to participate in the study. Community health nursing students who were in their senior year provided community health nursing to the experimental group over a 32-week period. The students were supervised by a nursing faculty investigator. Twenty-two of 29 subjects completed the study. Self-care behaviors improved significantly over the course of the study in the 11 subjects in the intervention group; managing complications (hypoglycemia and hyperglycemia), adhering to a meal and snack regimen, testing and recording blood glucose levels, and calling the physician about foot changes are aspects of self-care that improved most. Self-care behaviors decreased slightly over the course of the study in the control group. There was no significant difference between the two groups in outcomes for dietary adherence (nutritional content), foot care, blood glucose levels, overall diabetes knowledge, or functional health status and well-being. These results may reflect the limitations of the instruments used to measure outcomes, the small sample size, the use of novice nurses instead of experienced nurses, and the complexity of outcomes research.

glycemic control. The registry served as a reminder of the recommended elements of care (e.g., foot examinations). Delivery of care was redesigned to provide for patient group visits as well as individual patient visits and to establish a decentralized team of diabetes experts that sees patients jointly with primary care providers. The decentralized team traveled to each clinic several times a year and, along with the primary care team, saw each patient for approximately 30–40 minutes. The prevalence of testing for glycosylated hemoglobin and eye examinations both increased over the 3-year period after implementation of the program. Nearly two-thirds of patients with diabetes received annual eye examinations by the end of the 3-year period. In the first year of the program, half of all patients with diabetes had a foot examination compared with fewer than 20% before program implementation. Microalbuminuria screening also increased markedly after program implementation. The prevalence of smoking decreased from 14% in 1994 to 10% in 1996. Thirty randomly selected practices that collaborated with the decentralized diabetes expert team (group A) were compared with 30 randomly selected practices that did not work with this team (group B). The rates of eye examination and glycosylated hemoglobin testing increased in group A but did not change in group B between 1994 and 1996. In 1996, both rates were significantly higher in group A than in group B. However, the average glycosylated hemoglobin level in 1996 in group A was not significantly different from that in group B (7.7% and 7.8%, respectively). The rates of foot examination and microalbuminuria screening were more than threefold higher in group A compared with group B. The percentage of randomly selected primary care physicians who rated diabetes resources (e.g., access to certified diabetes educators) as good or excellent increased between 1992 and 1996. A controlled trial of Web-based diabetes disease management: the MGH diabetes primary care improvement project. Meigs JB, Cagliero E, Dubey A, Murphy-Sheehy P, Gildesgame C, Chueh H, Barry MJ, Singer DE, Nathan DM. Diabetes Care. 2003;26(3):750–7.

The effect of a program of support for primary care providers who care for patients with diabetes on provider satisfaction and health outcomes was assessed over a 3-year period beginning in 1994 at a not-for-profit staff-model health maintenance organization in Washington state. The support program comprised an online patient registry and the use of evidence-based clinical guidelines for eye and foot examinations, screening for microalbuminuria, and

The 1-year period after the intervention was compared with the 1-year period before the intervention. Testing of glycosylated hemoglobin and low-density lipoprotein cholesterol and foot screening increased significantly in the intervention group compared with the control group. Values of glycosylated hemoglobin (i.e., glycemic control) improved in the intervention group but not in the control group. Lipid and blood pressure control improved and

Disease Management for Diabetes

A population-based approach to diabetes management in a primary care setting: early results and lessons learned. McCulloch DK, Price MJ, Hindmarsh M, Wagner EH. Effective Clinical Practice. 1998;1:12–22.

The impact of a Web-based information management/decision support tool was evaluated in a randomized controlled trial of 598 adults with diabetes who attended a hospital-based internal medicine clinic. The intervention involved the use of the Web-based tool, which linked patient-specific information with evidence-based treatment recommendations.

[25]

Appendix B. Reports of the Impact of Disease Management Interventions on Treatment of Diabetes (continued) eye examination rates increased in both groups. The investigators concluded that the Web-based patient-specific decision support tool has the potential to improve evidence-based parameters of diabetes care.

Community-based chronic disease management program for African Americans. Nine SL, Lakies CL, Jarrett HK, Davis BA. Outcomes Manag. 2003;7(3):106–12.

A computer-generated reminder system improves physician compliance with diabetes preventive care guidelines. Nilasena DS, Lincoln MJ. Proc Annu Symp Comput Appl Med Care. 1995;640–5.

The impact of a chronic disease management program for African American persons with diabetes, hypertension, or both was evaluated over a 1-year period in a community setting in West Virginia. The chronic disease management program comprised exercise, medical nutrition therapy (dietary counseling and meal planning), support groups, cooking schools, and service coordination (goal setting and phone calls from and home visits by a service coordinator). Seventy-five patients, most of whom were African American, volunteered to participate.

Disease Management for Diabetes

A 6-month, randomized, controlled study was conducted to evaluate the impact of a computer-generated reminder system on physician compliance with guidelines for diabetes preventive care. Internal medicine resident physicians in their second or third year at the University of Utah who were providing care to patients with diabetes at outpatient general internal medicine clinics were randomized to an intervention group or control group. Residents in the intervention group were provided with a patient-specific report listing patient health data and upcoming or overdue preventive health activities (e.g., physical examinations, laboratory tests, referrals, patient education). Diabetes preventive care guidelines published by the American Diabetes Association addressing renal care, foot care, eye care, glycemic control, macrovascular care, and neurologic care were the basis for the computerized-reminder system. Residents in the control group received a generic patient report without specific recommendations. All patients with diabetes type 1 or 2 who were treated by the residents were included in the study; these patients were identified by reviewing clinic, pharmacy, and laboratory records. Encounter forms with patient clinical data were completed by residents in both groups after patient visits.

[26]

A score for compliance of the resident physician with diabetes preventive care guidelines was calculated for each patient visit. The average compliance score 1 month after implementation of the reminder system was significantly higher compared with baseline in the intervention group (54.9% after 1 month versus 38.0% at baseline). The average compliance score in the control group also increased significantly over the same period (51.0% after 1 month versus 34.6% at baseline); there was no significant difference between the two groups in the change in compliance. Possible explanations for these unexpected findings include the short duration of the study (many aspects of patient care are recommended annually), the Hawthorne effect (i.e., the altered behavior in the control group could have been the result of the fact that the subjects were under observation), or other factors. The highest compliance scores were for laboratory tests and referrals, and the lowest scores were for patient education. The use of encounter forms facilitated documentation of clinical data and compliance with guidelines for diabetes preventive care.

Significant reductions from baseline in systolic and diastolic blood pressure were observed after 1 year. Reductions in glycosylated hemoglobin also were observed but the reductions were significant only in the 12 patients with elevated baseline values (>7%). Quality of life improved but the change from baseline was not significant. The impact of an educational program on improving diabetes knowledge and changing behaviors of nurses in longterm care facilities. Parker MT, Keggett-Frazier N, Vincent PA, Swanson MS. The Diabetes Educator. 1995;21:541–5. The impact of a diabetes education program on knowledge and diabetes care practices of the nursing staff was ascertained in two long-term-care facilities (LTCFs) in eastern North Carolina. The nursing staff at another LTCF served as a control group. The education program comprised seven biweekly sessions conducted by diabetes educators over a 12-week period. A lecture format with slides and a question-and-answer period was used. Topics included medications, patient monitoring (blood glucose and ketones), managing hyperglycemia and hypoglycemia, diet, foot care, illness care, and exercise and patient-family education. Each educational session was 20 minutes long to facilitate nursing staff participation. A diabetes knowledge test was developed using guidelines of the American Association of Diabetes Educators and the American Diabetes Association. This test was administered at the start of the study (before the education program in the intervention group) and at the end of the study. Patient charts from before and after the intervention were reviewed to determine whether carbohydrates were provided to patients with hypoglycemia, ketones were tested and a physician was contacted when a patient exhibited hyperglycemia, insulin was administered in the abdomen (a preferred site because of the rapid rate of absorption), and toenails were cut by a podiatrist. Thirty-five registered nurses and licensed practical nurses participated in the study. They had 8 years of nursing experience on average. Only one nurse had diabetes, and another 14 had a family member with the disease. The intervention and control groups

Appendix B. Reports of the Impact of Disease Management Interventions on Treatment of Diabetes (continued) did not differ significantly in nursing education, nursing experience, experience with patients with diabetes, or participation in diabetes continuing education programs. The average pretest diabetes knowledge score was 67% for both groups. The average posttest score was significantly higher in the intervention group (73%) than in the control group (69%). The difference between the average pretest and posttest scores in the control group was not significant. The retrospective chart review revealed that the frequency of giving carbohydrates to patients with hypoglycemia, administering insulin in the abdomen, and having toenails cut by a podiatrist increased over the course of the study in the intervention group. However, the change was not significant. Thus, additional training as well as education is needed to improve diabetes care practices in the nursing staff at LTCFs. Do automated calls with nurse follow-up improve self-care and glycemic control among vulnerable patients with diabetes? Piette JD, Weinberger M, McPhee SJ, Mah CA, Kraemer FB, Crapo LM. Am J Med. 2000 Jan;108(1):20–7. The effects of automated telephone assessment and nurse followup on self-care (self-monitoring of blood glucose, foot inspection, weight monitoring, and medication adherence) and glycemic control were assessed in a 12-month randomized, controlled study of 248 English- or Spanish-speaking adult patients with diabetes managed by medication. The intervention involved biweekly, automated telephone assessment (patient use of the touch-tone telephone keypad to enter blood glucose readings and other data for review by a nurse) and self-care education calls with nurse followup. The self-care education component of the automated telephone assessment was a 3- to 5-minute interactive module on diet and weight control. The nurse follow-up involved telephone calls to patients to address problems identified in a review of data entered in the automated telephone assessment. The control group received usual care.

Patient-centered outcomes, including depression, anxiety, self-efficacy (patients’ confidence in participating in their own care), days in bed because of illness, days cut down on activities because of illness, diabetes-specific health-related quality of life, satisfaction with care, and general quality of life were assessed in the study of automated telephone assessment and nurse follow-up that is described above (Piette JD, Weinberger M, et al. Am J Med. 2000 Jan;108(1):20–7). Patients in the intervention group (i.e., participating in automated telephone assessment) had significantly fewer symptoms of depression and days in bed because of illness and significantly greater self-efficacy to perform self-care activities than patients in the control (usual care) group. Anxiety levels, diabetesspecific health-related quality of life, and general health-related quality of life were similar in the two groups. Impact of automated calls with nurse follow-up on diabetes treatment outcomes in a Department of Veterans Affairs Health Care System: a randomized controlled trial. Piette JD, Weinberger M, Kraemer FB, McPhee SJ. Diabetes Care. 2001 Feb;24(2):202–8. The effects of automated telephone assessment and nurse followup (see the summaries of the two reports by Piette JD, Weinberger M, et al., above) on patients’ self-care, symptoms, satisfaction with care, and glycemic control were assessed in 272 patients with diabetes at a Veterans Affairs clinic in a 12-month randomized controlled study. Only 44% of patients had a baseline glycosylated hemoglobin A1c level of 8% or higher, which is the level at which intervention is recommended (i.e., more than half of patients had good glycemic control). After 12 months, patients in the intervention group reported significantly more frequent self-monitoring of blood glucose and foot inspections and were more likely to be seen in podiatry clinics and diabetes specialty clinics than patients in the control (usual care) group. Among patients with a baseline glycosylated hemoglobin A1c of 8% or higher, the mean value after 12 months was significantly lower in the intervention group than in the control group (8.7% versus 9.2%, respectively). Significantly fewer symptoms of poor glycemic control and greater satisfaction with care were reported by patients in the intervention group than by patients in the control group. Adherence to protein restriction in patients with type 2 diabetes mellitus: a randomized trial. Pijls LT, de Vries H, van Eijk JT, Donker AJ. Eur J Clin Nutr. 2000 Apr;54(4):347–52. The effect of a dietary counseling on protein restriction was assessed in a 12-month randomized, controlled study of 125

Disease Management for Diabetes

Self-monitoring of blood glucose, foot inspection, and weight monitoring were more frequent and problems with medication adherence were less common in the intervention group than in the usual care group. The differences were significant. Mean glycosylated hemoglobin A1c values after 12 months were slightly lower in the intervention group than in the usual care group but the difference was not significant. More than twice as many patients in the intervention group as in the usual care group (17% versus 8%, respectively) had glycosylated hemoglobin A1c values in the normal range. Mean serum glucose levels were 41 mg/dL lower in the intervention group than in the usual care group, a difference that is significant. Symptoms of hyperglycemia and hypoglycemia were less common in the intervention group than in the usual care group. Similar percentages of subjects in the two groups were hospitalized or seen in the emergency department.

The effect of automated calls with telephone nurse follow-up on patient-centered outcomes of diabetes care: a randomized, controlled trial. Piette JD, Weinberger M, McPhee SJ. Med Care. 2000 Feb;38(2):218–30.

[27]

Appendix B. Reports of the Impact of Disease Management Interventions on Treatment of Diabetes (continued) patients with type 2 diabetes and microalbuminuria, relatively high albuminuria, or diabetes for at least 5 years and a high dietary protein intake. The experimental and control groups both received counseling on restricting dietary intake of saturated fats. The experimental group also received counseling on reducing dietary protein intake, with partial replacement of proteins by unsaturated fats and carbohydrates. Dietary protein intake (measured indirectly using urinary urea excretion) was significantly lower in the experimental group than in the control group after 6 months but the difference was smaller and not significant after 12 months. Factors predicting reduction in dietary protein intake included satisfaction with preexisting diet, low body mass index, and living alone. Dietary counseling resulted in only modest protein restriction. A randomized controlled trial of an intervention designed to improve the care given in general practice to type II diabetic patients: patient outcomes and professional ability to change behaviour. Pill R, Stott NCH, Rollnick SR, Rees M. Family Practice. 1998;15:229–35.

Disease Management for Diabetes

A 3-year randomized, controlled study was conducted to evaluate the effect of training for clinicians (general practitioners and practice nurses) on outcomes for patients with type 2 diabetes. The training addressed a patient-centered intervention, which was designed to (1) encourage clinicians to negotiate individual patient care plans based on patients’ perceptions of their disease and readiness to change their lifestyles and (2) to work toward realistic targets for behavior change. Thirty-three general practices with 252 patients with type 2 diabetes were randomized to the experimental group or the control group. Clinicians in the experimental group received at least 3 hours (divided into two sessions) of training in the patientcentered intervention. This training involved discussions, demonstrations, and role-playing. A research nurse provided continuing support, visited the practice sites, sent newsletters every 3 to 4 months, and held two group meetings during the course of the study.

[28]

There were no significant differences between the experimental and control groups in glycosylated hemoglobin level or the number of complications. Patient satisfaction with recent consultations and treatment improved significantly over the course of the study in the control group but not in the experimental group, although the latter group had little room for improvement. Physical functioning also improved in the control group. Most of this improvement was in female subjects (the control group had a larger percentage of women than did the experimental group). Analysis of audiotapes of patient consultation sessions revealed that nurses covered more topics and addressed diet, alcohol use, smoking, and body weight more often than did physicians. The differences were more marked in the experimental group than in the control group. Experimental group patients were more likely

than control group patients to initiate a discussion of behavioral change. Twenty-two (69%) of 32 clinicians were judged to have a moderate to good understanding of the methods and principles underlying the intervention. However, only six (19%) clinicians routinely applied these principles in practice. Improve control of type 2 diabetes mellitus: a practical education/behavior modification program in a primary care clinic. Ridgeway NA, Harvill DR, Harvill LM, et al. South Med J. 1999 Jul;92(7):667–72. This study evaluated the efficacy and ease of administration of education/behavior modification classes in improving disease control in 38 diabetic patients in a primary care clinic population. A computerized audit identified diabetic patients who visited the clinic in the preceding year, and physicians recommended participation to patients who met the following inclusion criteria: (1) a diagnosis of type 2 diabetes, (2) poor glycemic control, (3) ≥ 20% over ideal body weight, and (4) able to attend monthly clinic visits and comprehend presented material. Patients were randomized to an intervention group or control group, and metabolic parameters, body weight, health-related quality of life, and cognitive knowledge of diabetes were assessed at baseline and regular intervals. The intervention group attended 90-minute classes taught by a registered nurse and registered dietitian monthly for 6 months, with a follow-up session at 12 months. Didactic portions of the classes were based on the Life Skills program, featuring lectures, teaching slides, and handouts about diabetes, its treatment, and consequences. Diet and exercise were emphasized as important means of controlling diabetes. Behavior modification involved (1) individual sessions with instructors to individualize diets, prescribe exercise, and set goals; (2) use of worksheets and contracts to emphasize need for patient responsibility; (3) class participation, encouragement, and praise; and (4) brief classroom appearances by physicians. Statistically significant reductions in mean fasting blood glucose (FBG), glycosylated hemoglobin (GHb), total cholesterol, and lowdensity lipoprotein cholesterol (LDL-C) values were observed at the end of the intervention in the intervention group (N=18; p< 0.05). These values remained decreased, but not significantly different from baseline, at the 12-month follow-up, when a significant decrease in mean body weight was noted (p=0.0166). Improvement in the control group (N=20) was limited to small decreases in GHb and body weight at 6 months (P<0.02), which did not persist at significant levels at 12 months. The Diabetes Education Society’s Life Skills test revealed that patients in the intervention group improved and had significantly greater knowledge about diabetes than those in the control group post-intervention. Results from quality life assessments (i.e., Medical Outcomes Study 36-item, short-form survey; diabetes-related problems questionnaire) showed no significant intergroup differences.

Appendix B. Reports of the Impact of Disease Management Interventions on Treatment of Diabetes (continued) The short-lived clinical improvements in the intervention group were consistent with those seen with different techniques, as was the lack of difference in quality of life between groups. The cost of the program was estimated to be $95 per patient for educational materials and teaching salaries only. Thus, the class was considered inexpensive as well as easy to administer, well suited to primary care clinics, and clinically worthwhile. Pharmacist-led, primary care-based disease management improves hemoglobin A1c in high-risk patients with diabetes. Rothman R, Malone R, Bryant B, Horlen C, Pignone M. Am J Med Qual. 2003;18(2):51–8. The impact of a pharmacist-based diabetes care program was evaluated retrospectively in 138 patients with type 2 diabetes and poor glycemic control (glycosylated hemoglobin 8% or higher). The intervention involved diabetes education, limited physical examination (e.g., foot exams), initial treatment recommendations, the use of medication algorithms, and frequent patient follow-up with the support of a computer database. Patient education addressed glucose control and monitoring, management of hypoglycemia, nutrition and exercise, proper foot and eye care, and medication management. After 6 months, a significant reduction in mean glycosylated hemoglobin was observed. A recent diagnosis of diabetes (within the previous 4 months) and a high baseline glycosylated hemoglobin value were associated with improvement in glycosylated hemoglobin. Age, race, gender, and educational status did not predict outcome. Evaluation of a quality improvement intervention for diabetes management. Schmidt SO, Burns C, Feller DB, Chang KL, Hernandez B, McCarthy J, Burg MA. J Healthc Qual. 2003;25(3):26–32.

Compared with baseline, both interventions significantly improved

In the flow sheet-only clinic, the percentage of patients with poor glycemic control (glycosylated hemoglobin >9.5%) decreased significantly, and ADA targets for this variable were met. No other significant changes in patient outcomes (lipid profile and blood pressure) were observed in either intervention group or in the control groups. Although the impact of the diabetes management flow sheet on provider compliance with ADA standards and patient outcomes was minimal, providers participating in the study believed that use of the flow sheet plus feedback increased their awareness of diabetes management guidelines. Motivational interviewing to improve adherence to a behavioral weight-control program for older obese women with NIDDM. Smith DE, Heckemeyer CM, Kraty PP, Mason DA. Diabetes Care. 1997;20:52–4. The impact of adding motivational interviewing strategies to a behavioral obesity program on treatment adherence, glycemic control, and weight loss was explored in a randomized pilot study. Twenty-two obese women who were 50 years of age or older and had type 2 diabetes were recruited by advertisements and a patient letter. All patients participated in a 16-session group behavioral weight-control program with moderate calorie restriction (1,200–1,500 kcal/day), fat intake restriction (30–40 g/day), increased physical activity, and home monitoring of blood glucose. Weekly group meetings provided nutritional information and training in how to modify eating and exercise behavior. Patients maintained and periodically submitted diaries in which calorie consumption and physical activity were noted daily and fasting blood glucose levels were recorded three times weekly. Patients randomized to the motivational group also participated in three motivational interviewing sessions (one at the beginning of the study and two during the study) conducted by a psychologist experienced in the technique. Motivational interviews assessed the patient’s ambivalence about behavior change, elicited and formulated personal goals, and solved problems that presented barriers to change. Objective data on health and behavior were reviewed with the patient to delineate discrepancies between current status and goals. The interviewer used open-ended questions and a reflective listening approach, avoiding confrontation. Realistic and objective goals were developed collaboratively with the patient.

Disease Management for Diabetes

The impact of two quality improvement interventions on physicians’ diabetes management behaviors was assessed in six outpatient clinics associated with a university-affiliated family practice department. One intervention was the use of a diabetes management flow sheet to be inserted in the patient chart where it would serve as a prompt for the health care provider as well as a tool to educate the provider on American Diabetes Association (ADA) guidelines for treating patients with diabetes (ADA recommended frequencies for various tests and examinations). The second intervention involved the use of the diabetes management flow sheet plus quarterly provider feedback on the results of chart audits of performance in treating patients (i.e., compliance with ADA recommendations). One clinic served as the “flow sheet-only” site, another clinic served as the “flow sheet-plus-provider-feedback” site, and the other four clinics served as control sites.

the performance of foot examinations over a 12-month period, with a greater improvement in the flow sheet-plus-provider-feedback clinic than in the flow sheet-only clinic. There was little change in this variable in the control groups. The percentages of patients receiving annual testing for glycosylated hemoglobin, eye examination, and assessment for diabetic nephropathy did not improve significantly in either intervention group. These percentages decreased in the control groups.

[29]

Appendix B. Reports of the Impact of Disease Management Interventions on Treatment of Diabetes (continued) The motivational group demonstrated significantly better adherence to the behavioral obesity program than did the standard (control) group. The group session attendance, number of diaries submitted, and blood glucose monitoring frequency were greater in the motivational group. Blood glucose control also was better in the motivational group than in the standard group; the glycosylated hemoglobin at the end of the 16-week study was 9.8% in the motivational group and 10.8% in the standard group. Both groups lost a significant amount of weight over the course of the study (on average, 5.5 kg in the motivational group and 4.5 kg in the standard group). However, there was no significant difference between the two groups in weight loss, possibly because of the small sample size (16 patients due to attrition). Telecommunication support for rural women with diabetes. Smith L, Weinert C. Diabetes Educ. 2000 Jul-Aug;26:645–55.

Disease Management for Diabetes

Thirty women with diabetes who lived in rural areas far from the closest source of health care participated in a 10-month randomized, controlled trial of the use of computer-based telecommunications technology to provide support, information, and education. A control group received usual modes of support and communication. Subjects in the computer group were not required to be computer literate to participate. The computer software had four components, including conversation among patients (i.e., a support group function), private dialogue between a patient and a nurse or between two patients, a “health chat” guided by a certified diabetes educator with diabetes information, and a bulletin board with postings of pertinent information by the investigators. The computer software compiled data on computer access frequency and duration. Telephone interviews and mail questionnaires were used to assess social support, quality of life, psychosocial adaptation to illness (i.e., well-being), and attitudes about the impact of the computer intervention.

[30]

The average time spent on the computer each month was highest the first month (129 minutes) and decreased in subsequent months, possibly because of the novelty or the need to learn to use the software during the first month. Quality of life scores were higher for women in both groups who had more education and improving health. The computer group had slightly better psychosocial adjustment to illness scores than the control group. Attitudes toward the use of telecommunications technology were positive and 77% of subjects in the computer group recommended its use “very highly.” Conversation among patients (i.e., the support group function) was the most frequently used software component. The program provided a great deal of support to 77% of women in the computer group.

Professional responses to innovative in clinical method: diabetes care and negotiating skills. Stott NC, Rees M, Rollnick S, Pill RM, Hackett P. Patient Educ Couns. 1996 Oct;29(1):67–73. This randomized controlled trial evaluated the responses of family doctors and nurses to application of an innovative clinical technique (visual agenda-setting technology) designed to facilitate patientclinician interaction in routine diabetes care and promote patient lifestyle changes. Doctors and nurses at 30 family practices identified as being “interested and active in diabetes care” (i.e., recent professional participation in continuing education sessions) were invited to participate. Twenty-nine practices agreed to participate, and doctors and nurses in these practices were randomly assigned to an experimental group (doctors=16, nurses=18) or a control group (doctors=14, nurses=15). Over the 3-year course of the study, these professionals provided care to 200 patients with type 2 diabetes within their practices. The intervention began with a visit to each practice to explore how the doctor(s) and nurse(s) felt about care of patients with diabetes and their intra-practice organization of such care. The contents of an “Agenda Setting Pack” were then discussed and demonstrated in various ways, including role-playing, to members of the intervention group. This Pack consisted of a visual agenda-setting chart, a readiness-to-change ruler, a diary, and a balance chart (to weigh pros and cons of a given change). The former chart appeared as a series of pictures (e.g., food, exercise, cigarettes, medication) with the words “To Change or Not To Change” at the top. Next, each intervention, family practice-based team was encouraged to negotiate the number of formal training sessions based on perceived need. All opted for a least two 1.5-hour sessions in which application of the visual materials to patients, using negotiation and motivational interviewing techniques, was reviewed. Continuing contact with the members of the experimental group was achieved with bimonthly newsletters, personal contacts, and group meetings. In addition, invitations for audio recordings of clinical consultations were accepted by more than 90% of nurses and 56% of doctors. Evaluation via questionnaire halfway through the study (i.e., at 18 months) revealed that 71% of the clinicians frequently used the Agenda Setting Chart and 22% occasionally used the chart; 6% reported rare use and one group failed to return the questionnaire. Many participants also made selective use of the remaining materials designed for “the more selective or ambivalent patient.” For example, the diary was used “sometimes” by 55% of the doctors and nurses in the experimental group and “most of the time” by 19%. The readiness-to-change ruler was used sometimes by 42% and most of the time by 35%. The balance chart was used sometimes by 45% and most of the time by 22%. Additional evaluation and sharing of experiences occurred at group

Appendix B. Reports of the Impact of Disease Management Interventions on Treatment of Diabetes (continued) meetings designed for the experimental group. The first meeting, attended by 83% of participating nurses and 56% of participating doctors, was run as a workshop intended to explore use of the method, identify problems, and share successes. The second meeting (7 months later) was a workshop based on a series of tape-recorded role-plays between a clinician and patient. Evaluations following both meetings confirmed that doctors and nurses found sharing of experiences in a group useful, and 90% of these professionals agreed or strongly agreed that a style of consultation with patients with type 2 diabetes that uses negotiation is helpful. While tape-recorded encounters were used in a number of cases, a detailed analysis of these events was not included in this report. Neither evaluation of the control group nor comparison of parameters between the intervention and control groups was addressed. Patient education in the management of diabetes mellitus. Tan ASL, Yong LS, Wan S, Wong ML. Singapore Med J. 1997;38:156–60. The effects of a diabetes education program on diabetes knowledge, dietary practices, medication compliance, self-monitoring of blood or urine glucose, and glycosylated hemoglobin were assessed in a controlled study of 278 patients with diabetes who attended a primary health care clinic in Singapore. The first 100 patients were assigned to the control group (five patients eventually dropped out), which received usual treatment and advice on managing diabetes from their physician. Subsequent patients were assigned to the intervention group, which participated in a diabetes education program. The program consisted of at least six individual or smallgroup counseling sessions at which patient booklets on the disease, treatment, handling of emergencies, self-monitoring of blood glucose, meal planning, and general health care were reviewed. These sessions were conducted by a nurse trained in counseling techniques. Four larger group sessions also were held, with lectures, videotaped presentations, group discussions, and food displays. Family members were encouraged to attend these sessions.

Diabetes prevention programs are essential in Native American communities that are experiencing high rates of non–insulindependent diabetes. The Zuni Diabetes Prevention program is a community-based primary prevention project designed to reduce the prevalence of diabetes risk factors among high-school-age youths. The size of the population is not specified (it is unclear how many patients participated for the first 2 years of this 4-year study and provided longitudinal data for that period). The program strives to enhance knowledge of diabetes and to support increased physical activity, increased fruit and vegetable intake, and reduced soft drink consumption. The primary mechanisms of intervention are diabetes education, school-based wellness center supportive social networks, and modification of the food supply to reduce fat content and consumption of sugary beverages and increase fiber content. There was collaboration between the public school district and university department of family and community medicine. Mid-project results indicate a significant reduction in soft drink consumption and an increase in glucose/insulin rations, suggesting a decline in the incidence of hyperinsulinemia. Outcomes measured include body mass index, dietary intake of fiber and sugary beverages, heart rate, and fasting and postprandial glucose and insulin levels. Within 2 years after the intervention, the consumption of sugary beverages, body mass index, and heart rate had decreased (a decreased heart rate suggests improved cardiovascular fitness) and dietary fiber intake and glucose-to-insulin ratios had increased (an increase in the glucose-to-insulin ratio suggests a reduced prevalence of hyperinsulinemia and risk for type 2 diabetes). Only the changes in sugary beverage consumption and insulin levels were significant but the other changes suggest the adoption of healthy behaviors that might reduce the prevalence of risk factors for type 2 disease as the study continues. Early lifestyle intervention in patients with non–insulindependent diabetes mellitus and impaired glucose tolerance. Uusitupa MIJ. Ann Med. 1996;28:445–9. A controlled study was conducted in 86 obese Finnish patients recently diagnosed with type 2 diabetes to determine the impact of nonpharmacologic interventions (i.e., diet, exercise) on glycemic control and cardiovascular risk factors (e.g., serum lipids, blood pressure). After a 3-month period during which all patients received basic education about the disease and dietary advice about losing weight and reducing the intake of saturated fat and cholesterol, patients were randomized to an intervention group or a

Disease Management for Diabetes

Compared with baseline, scores on tests of diabetes knowledge (disease, complications, management, and self-care) were significantly higher 1 year after the completion of the education program in both the intervention and control groups. The magnitude of the increase was greater in the intervention group than in the control group. Dietary practices (recommendations to increase fiber intake and decrease intake of calories and fats) improved significantly from baseline in the intervention group. There were no significant changes in the control group. Medication compliance increased significantly in both groups and was higher in the intervention group than in the control group. There was no significant change in the percentage of patients in either group who self-monitored blood or urine glucose. Glycosylated hemoglobin decreased significantly from baseline in the intervention group. However, the change from baseline in the glycosylated hemoglobin was not significant in the control group.

Development of a primary prevention program: insight gained in the Zuni diabetes prevention program. Teufel NI, Ritenbaugh CK. Clin Pediatr. 1998;37:131–42.

[31]

Appendix B. Reports of the Impact of Disease Management Interventions on Treatment of Diabetes (continued) conventional treatment group. The intervention group made six visits to an outpatient clinic at 2-month intervals for intensive therapy with (1) a restricted dietary intake of calories, total fat (<30%), saturated fats (<10%), and cholesterol (<300 mg/day), which was monitored by examining food records and measuring serum lipids and (2) exercise training (30–60 minutes three to four times per week), which was monitored by analyzing daily exercise records. The intervention was not based on specific treatment guidelines. The intervention led to a reduction in intake of saturated fats and calories. Dietary changes in the conventional treatment group were smaller than in the intervention group. The total weight loss from the time of diagnosis until the end of the 12-month intervention period was greater in the intervention group than in the conventional treatment group (6.9 kg versus 3.8 kg, respectively). Weight loss and beneficial changes in metabolic control (fasting blood glucose) and lipid profile (e.g., high-density lipoprotein cholesterol, triglycerides) were observed during the intervention period only in the intervention group. There was a trend toward reduced blood pressure in both groups. Follow-up assessment 1 year after the completion of the intervention period revealed that glycemic control was maintained by a significantly larger percentage of patients in the intervention group than in the conventional treatment group, although deterioration in control was found in both groups. Significantly fewer patients in the intervention group were receiving oral antidiabetic agents than patients in the conventional treatment group (34.8% versus 12.5%, respectively).

Disease Management for Diabetes

A randomized, controlled study to assess the impact of nonpharmacologic interventions (weight reduction, dietary saturated fat restriction, and increased physical activity) on the development of type 2 diabetes in patients with impaired glucose tolerance (a condition that often leads to diabetes) is under way at five centers in Finland. Preliminary results after 1 year of data collection in the 5- or 6-year study suggest a weight loss and improvement in glucose tolerance in only the intervention group.

[32]

Education integrated into structured general practice care for type 2 diabetic patients results in sustained improvement of disease knowledge and self-care. van den Arend IJ, Stolk RP, Rutten GE, Schrijvers GJ. Diabet Med. 2000 Mar;17(3):190–7. The effects of four structured programs for diabetes care with and without integrated education on disease knowledge, self-care behavior (dietary adherence, self-monitoring of blood glucose, physical exercise, and foot inspection), and disease perception were assessed in 243 patients with type 2 diabetes after 6 months (i.e., after the program) and after 12 months (i.e., at follow-up to ascertain whether effects were lasting). The programs included (1) care guided by protocol, with oral and written patient information, (2) care with computerized support and a diabetes nurse avail-

able to answer patient questions, (3) a special educational program focusing on blood glucose monitoring for a subset of patients with poor glycemic control, and (4) an integrated care program, including an educational program with didactic elements and interactive group discussion on basic self-care skills and diabetes pathophysiology and complications. Disease knowledge increased in all four programs and was maintained at follow-up. The increase in disease knowledge was significantly greater in programs with an educational component (programs 3 and 4) than in those without an educational component (programs 1 and 2). Similarly, self-care behavior (dietary adherence, self-monitoring of blood glucose, physical exercise, and foot inspection) was increased in all four programs after 6 months and was maintained after 12 months. The increases in self-care behavior were significantly greater in the programs with an educational component than in those without an educational component. There was no change in disease perception in any of the programs. Disease knowledge and self-care behavior were positively related (i.e., disease knowledge correlated with self-care behavior). A nurse-coordinated intervention for primary care patients with non–insulin-dependent diabetes mellitus: impact on glycemic control and health-related quality of life. Weinberger M, Kirkman MS, Samsa GP, et al. J Gen Intern Med. 1995;10:59–66. The impact of nurse-initiated telephone contacts between office visits to primary care physicians was evaluated in 275 patients with type 2 diabetes at a Veteran Affairs general medical clinic (GMC). A computerized audit identified patients who had ever filled a prescription for insulin or an oral hypoglycemic at the medical center pharmacy and had visited the GMC in the preceding year. Patients who met the inclusion criteria (i.e., diagnosis of type 2 diabetes, current use of insulin or oral hypoglycemic agent, receiving regular care at GMC, telephone access) were invited to participate via letter, phone, or clinic visit. Following baseline data collection, patients were randomly assigned to three intervention groups (followed by three study nurses) or a control group. Patients in the intervention group received monthly (or more frequent) phone calls from the study nurses intended to provide education, facilitate and reinforce compliance, monitor health status, facilitate resolution of identified problems, and facilitate access to primary care; patients were also encouraged to contact study nurses should questions arise. Patients in the control group received usual care (i.e., physician visit every 3–4 months), with no systematic provider-initiated education, monitoring of health status, or telephone reminders between clinic visits. Baseline sociodemographic, clinical, and health-related quality of life (HRQOL) data were similar between groups. During the study, 2,399 nurse-patient telephone contacts were made, with only 162 (6.8%) initiated by the patients. The mean number of contacts per

Appendix B. Reports of the Impact of Disease Management Interventions on Treatment of Diabetes (continued) intervention patient was 13, and advice was provided to the patient in 57% of phone contacts. One-year follow-up data, available for 151 of the 275 randomized patients, revealed significant intergroup differences with respect to both fasting blood sugar (FBS) and glycosylated hemoglobin (GHb)(p<0.05), which favored the intervention group. However, no statistically significant intergroup differences existed for scores on the Medical Outcomes Study 36-item, short-form survey (used to assess HRQOL) or patient-reported diabetes-related signs and symptoms. Thus, between-office nurseinitiated telephone contacts, designed to serve as pragmatic, low-intensity adjuncts to physician-administered care, modestly improved glycemic control, but not HRQOL or diabetes-related signs and symptoms, in patients with type 2 diabetes. Telemedicine improved diabetic management. Whitlock WL, Brown A, Moore K, Pavliscsak H, Dingbaum A, Lacefield D, Buker K, Xenakis S. Mil Med. 2000 Aug;165(8):579–84. The effects of weekly home telemedicine visits by a nurse case manager to review blood glucose levels, body weight, blood pressure, hypoglycemic episodes, exercise and nutrition goals, and well-being were assessed in a 3-month randomized, controlled study of 28 adult patients with type 2 diabetes. A glycosylated hemoglobin A1c level greater than 8% was an eligibility criterion. The nurse case manager maintained frequent contact with the patients’ physician by electronic mail, and the physician made monthly home telemedicine visits to evaluate the patients. A control group received routine diabetes care. Both the telemedicine group and the control group were encouraged to participate in multidisciplinary diabetes education classes held at a medical center.

The effect of diet, exercise, or a combination of diet and exercise on body weight, risk factors for cardiovascular disease (lipid profile, blood pressure), and the development of type 2 diabetes was assessed in a 2-year controlled study of 154 non-diabetic individuals at risk for the disease because of overweight and a parent with diabetes. Subjects were recruited through newspaper advertisements. Subjects randomized to the control group received a selfhelp manual with information about healthy eating, exercise, and behavioral strategies for weight control. Subjects randomized to the diet, exercise, and diet plus exercise groups attended weekly group sessions for the first 6 months of the study, biweekly sessions for the next 6 months, and two 6-week refresher courses during the second year of the study. These sessions were conducted by a multidisciplinary team of health care providers, including registered dietitians, exercise physiologists, and behavior therapists. In the diet group, calorie intake was severely restricted (800–1,000 kcal/day with 20% of calories as fat) for the first 8 weeks of the study and relaxed thereafter (1,200–1,500 kcal/day). Meal plans and shopping lists were provided. The exercise group was encouraged to gradually increase physical activity to 1,500 kcal/week. Supervised exercise was performed at weekly sessions for 10 weeks. Weight loss and improvement in fasting glucose and insulin, lipid parameters, and blood pressure over the first 6 months of the study were significantly greater in the diet group and diet plus exercise group than in the exercise group and control group. However, attendance at group sessions decreased during the second 6 months of the study and some of the weight loss was regained in all groups. After 1 year, the diet group and diet plus exercise group had maintained 60% and 72% of the weight loss, respectively. These two groups had significantly greater weight loss than the exercise group and control group. By the end of the 2-year study, the lipid profile and blood pressure had returned to baseline levels and glycemic control had worsened in all groups. Type 2 diabetes had developed in 17% of subjects. Weight loss over the 2-year study period and impaired glucose tolerance at baseline were predictors of type 2 diabetes. A modest weight loss (4.5 kg) reduced the risk of type 2 diabetes by 30% (compared with no weight loss), even in subjects with impaired glucose tolerance at baseline.

Disease Management for Diabetes

After 3 months, there was a significant reduction in glycosylated hemoglobin A1c in the telemedicine group (from a mean of 9.5% to 8.2%) but not in the control group. Similarly, a significant reduction in total body weight was observed over the course of the study in the telemedicine group (from a mean of 214 to 207 pounds) but not in the control group. Triglyceride levels decreased in the telemedicine group and low-density lipoprotein levels increased in the control group but other lipid values did not change significantly. There was no change in quality of life in either group. The nurse case manager’s and physician’s subjective impression of the usefulness of telemedicine was favorable but the use of the technology was hampered by technical difficulties.

Lifestyle intervention in overweight individuals with a family history of diabetes. Wing RR, Venditti E, Jakicic JM, Polley BA, Lang W. Diabetes Care. 1998;21:350–9.

[33]

Appendix C.

Disease Management for Diabetes

Author(s)

[34]

Size of Population

Method of Identifying Population for Whom Data Are Evaluated

Intervention Strategy

Guideline Based?

Audience for Intervention

Primary Manager of Intervention

American Diabetes Association, American Association of Diabetes Educators, and federal and state regulations governing home health care

Registered nurses (45 at the agency with a diabetes nurse educator and 17 at the agency without an educator)

Diabetes nurse educator

Health care providers at regional and subregional hospitals and facilities and patients with diabetes

Core team

Adams CE, Cook DL, 1994

Two home health care agencies, one with a diabetes nurse educator and the other without a diabetes nurse educator

Home health care agencies were selected at the convenience of the researcher.

Completing needs assessments for both patients and staff; developing nursing standards of care, staff education programs, and specific methods for implementing and evaluating care; designing diabetes self-care content that is user friendly; and providing ongoing support for staff nurses

Amoah AG, Owusu SK, et al., 2000

Patients with diabetes in a developing country with a population of 18 million, 85% of whom are less than 45 years old, served by 2 teaching hospitals, 9 regional hospitals, and 100 subregional hospitals and health care facilities

Not specified

Intensive training No established of two core teams guidelines were comprising a available. physician, a nurse, and a dietitian in a multidisciplinary approach to diabetes care and education, who subsequently developed a patient education booklet and a professional diabetes education curriculum for training health care providers at regional and subregional facilities

RCT=randomized controlled trial

Outcomes Measured

Time Period Studied

Study/Evaluation Design

Patient’s basic knowledge and patient care score on a list of 33 items addressing patient admission, care plan, ongoing care (assessment, patient education, nursing care treatments, and evaluation), and discharge

Not applicable (outcomes provided a snapshot of the status of the two home health care agencies, and the length of time that the diabetes nurse educator was present at the one home health care agency was not specified)

Correlational research design (nonexperimental design)

Training of health care providers and provision of diabetes services

3 years

Not randomized or controlled

Economic Effects Assessed

Setting

Key Results

None

Home health care agencies

Registered nurses at a home health care agency with a diabetes nurse educator scored significantly higher on a test of basic diabetes knowledge than nurses at a home health care agency without a diabetes nurse educator. The standard of patient care provided by the home health care agency with the diabetes nurse educator was significantly higher than that provided by the other agency.

None

Hospital-based clinics

After 3 years, trained diabetes health care teams were in place and diabetes services were provided in all regional and about 63% of subregional hospitals. The number of patients with diabetes registered before the program was 4,719 and after the program, it was 13,643.

Disease Management for Diabetes [35]

Appendix C. (continued)

Disease Management for Diabetes

Author(s)

[36]

Size of Population

Method of Identifying Population for Whom Data Are Evaluated

Intervention Strategy

Guideline Based?

Audience for Intervention

Primary Manager of Intervention

Anderson RM, Fitzgerald JT, et al., 1994

1,863 patients with type 1 or 2 diabetes in Michigan

Recruited from physician offices, pharmacies, clinic waiting rooms, and diabetes patient education programs; patients were asked to return a postagepaid postcard if they wished to receive a free monthly newsletter about diabetes

Newsletter with Not specified readable, concise, action-oriented information about diabetes and encouragement to become assertive consumers of diabetes care who are prepared to interact with physicians, nurses, and dietitians

Patients with type 1 or 2 diabetes

Not specified

Anderson RM, Funnell MM, et al., 1991

23 diabetes educators (20 nurses, 2 dietitians, 1 “other”)

Volunteers responding to mailed flyer

3-day simulated Not specified diabetes care regimen (two daily saline injections, self-monitoring of blood glucose four times daily, 1,200 kcal diet, daily foot care and aerobic exercise, and extensive record keeping) followed by a 3-day intensive skillsbased workshop, using a four-phase learning sequence with information on educational and counseling skills, demonstration of counseling skills, videotaped small-group practice in counseling, and review of videotapes

Diabetes educators

Not specified

Anon, 2003 (Dis Manag Advis. 2003;9:120–122)

105 patients 8–17 years of age

Not specified

Family-focused teamwork

Teenagers

College graduates without health care background

RCT=randomized controlled trial

Not specified

Time Period Studied

Study/Evaluation Design

Economic Effects Assessed

Overall response to newsletters (patient ratings of whether the newsletters were helpful) and specific reactions to individual issues (i.e., topics) of newsletter (whether and with whom issue was discussed and whether change was made)

2 years (21 newsletters)

Nonrandomized

Diabetes educators’ counseling skills, attitudes toward patient autonomy, and need for special training for diabetes educators

6 days

Glycemic control

1 year

Setting

Key Results

None

Community (patients were at home)

The newsletters were most helpful to patients with type 2 diabetes who were elderly or in poor overall health or who had complications, low incomes, or a poor understanding of diabetes. The likelihood of change in diabetes care in response to newsletters was low in all patients.

Not randomized or controlled

None

Not applicable

Counseling skills and attitudes toward patient autonomy improved significantly as a result of the intervention.

RCT

None

Home

Glycemic control was significantly better in the intervention group than the usual care group.

Disease Management for Diabetes

Outcomes Measured

[37]

Appendix C. (continued)

Intervention Strategy

Guideline Based?

Audience for Intervention

Primary Manager of Intervention

Anon, 2003 (Hosp 58,000 patients Not specified Case Manag. with diabetes 2003;11(7):103–104)

Care standardization, patient and staff education

Not specified

Patients, staff

Lead diabetic educator or coordinator

Arsenau DL, Mason AC, et al., 1994

40 patients with diabetes

Patients attending a diabetes education program were asked to participate.

An individualized method of learning instead of classroom instruction

American Diabetes Patients with Association and diabetes American Dietetic Association

Aubert RE, Herman WH, et al., 1998

17 patients with type 1 diabetes and 121 patients with type 2 disease

Group-model HMO pharmacy records and a database of patients who had visited a physician for diabetes care, been hospitalized for diabetes, been seen by a utilization management nurse, or been referred to an ophthalmologist for a diabetic retinal examination

Nurse case Not specified management using written algorithms under the direction of a family physician and endocrinologist (close patient monitoring, continuous reinforcement of dietary and exercise recommendations, and systematic drug therapy adjustments by a nurse case manager)

Disease Management for Diabetes

Author(s)

[38]

Size of Population

Method of Identifying Population for Whom Data Are Evaluated

RCT=randomized controlled trial

Patients with diabetes

Registered dietitians, nutrition specialist, and education specialist reviewed content and design of individualized learning programs from existing sources.

Registered nurse who was a certified diabetes educator

Economic Effects Assessed

Outcomes Measured

Time Period Studied

Study/Evaluation Design

Glycosylated hemoglobin

6 months

Non-randomized

Hourly clinic operational costs decreased by 20%.

Fasting plasma 5 months glucose, glycosylated hemoglobin, percentage of ideal body weight, knowledge, and behavior scores

RCT

Individualized Hospital-based learning was judged diabetes center cost-effective. (presumably clinic)

The individualized method of education was as effective as traditional classroom instruction in educating patients with diabetes and may be more cost-effective but the individualized method did not improve glycemic control.

Glycosylated 12 months hemoglobin, fasting blood glucose levels, medication requirements, body weight, blood pressure, serum cholesterol and triglyceride levels, self-rated health status, episodes of severe hypoglycemia, emergency department visits, and hospital admissions

RCT

None

Glycemic control improved to a significantly greater extent with nurse case management than with usual care.

Setting

Key Results

Health system (clinic)

The average glycosylated hemoglobin decreased from 9.2% to 7.0%.

Primary care clinics in a group-model HMO

Disease Management for Diabetes [39]

Appendix C. (continued)

Disease Management for Diabetes

Author(s)

[40]

Size of Population

Method of Identifying Population for Whom Data Are Evaluated

Intervention Strategy

Guideline Based?

Audience for Intervention

Primary Manager of Intervention

Baker SB, Vallbona C, et al., 1993

4,300 patients with diabetes

Not specified

Development and American Diabetes implementation Association, of protocols for National prevention and Cholesterol care of diabetes- Education related Program complications, general management of diabetes, education of patients and professionals

Patients with diabetes and health care professionals

Physicians, physician assistants, nurses, diabetes nurse educator, nutritionists, social workers, podiatrists

Barth R, Campbell LV, et al., 1990

62 participants

Patients were recruited through an extensive radio and newspaper campaign, from referrals by general practitioners and from people attending the Diabetes Center, St. Vincent’s Hospital, and other diabetes centers in Sydney, Australia

Conventional program: Diabetes education, lectures on foot care, footwear, and dealing with extreme temperatures.

Patients with type 2 diabetic patients, (age >30 years, overweight)

Not specified

Patients with type 1 or 2 diabetes

Diabetes educators

Bernbaum M, Wittry S, et al., 2000

Programs were designed by the current standards of practice in Australia (unclear if there were actual guidelines).

Intensive program: In addition to diabetes education, foot care information and demonstrations were given.

163 visually Participants were Adaptive diabetes Not specified impaired referred but method education program patients with not specified. with adaptive type 1 or equipment (e.g., type 2 diabetes syringe magnifiers, syringe loading devices, glucose monitoring systems with speech capability, and tactile aids for proper blood sample placement)

RCT=randomized controlled trial

Outcomes Measured

Time Period Studied

Study/Evaluation Design

Economic Effects Assessed

Setting

Key Results

Retrospective chart reviews (not randomized or controlled)

Costs of 12 months of treatment for diabetic eye disease (benefits and benefit-to-cost ratio were projected)

Community health centers in urban low-income neighborhoods

Preliminary results suggest that screening and preventive treatment for diabetic eye disease is cost-effective. The percentage of patients with eye examinations increased and the incidence of legal blindness decreased. Foot examinations and blood pressure control both improved, although body weight and blood glucose control have not improved. Patient knowledge and physician accuracy in eye examinations both increased.

Routine follow-up 6 months visits to the podiatrist (other than project podiatrist), foot care knowledge, and routine foot care compliance

Experimental design with 2 samples

Implementation of the intensive program is more costly than the conventional program because the intensive group requires more manpower per person.

Diabetes Center and Garvan Institute of Medical Research, St. Vincent’s Hospital, Sydney, Australia

The intensive group showed greater improvements than the conventional group in foot care knowledge, compliance with the recommended foot care routine, and compliance with the initial advice to consult a podiatrist for further treatment. The intensive group also showed a significant reduction in the number of foot problems that required treatment compared to the conventional group.

Demonstration of proficient self-care technique

Not randomized or controlled

None

University-based hospital clinic

The success rate in restoring diabetes self-care practices was 72% overall, including 81% in patients with early visual impairment (legal blindness with some residual functional vision) and 57% in patients with late visual impairment (legal blindness with some residual functional vision).

10 years

Disease Management for Diabetes

Blood glucose level, 5+ years (study body weight, is ongoing) prevalence of uncontrolled hypertension, number of patients receiving eye examinations, incidence of legal blindness, number of patients receiving foot examinations, patient knowledge, appropriateness of prescribing, physician proficiency in assessing visual impairment

[41]

Appendix C. (continued)

Disease Management for Diabetes

Author(s)

[42]

Size of Population

Method of Identifying Population for Whom Data Are Evaluated

Intervention Strategy

Guideline Based?

Audience for Intervention

Primary Manager of Intervention

Birke JA, Horswell R, et al., 2003

>14,000 patients

Not specified

Disease American Diabetes Patients management Association initiative (DMI) with or without staged management diabetes foot program (DFP)

Not specified (multidisciplinary)

Bloomgarden ZT, Karmally W, et al., 1987

266 diabetic patients

Clinic roster was used to identify all insulin-treated clinic patients who were contacted by telephone to establish an appointment for interview

Nine monthly Not specified education sessions (audiovisual materials, card games) reviewing diabetes physiology; foot, skin, and dental hygiene; insulin administration and emergencies; complications and risk factors; and nutrition. Separate sessions and audiovisual materials in Spanish were available for Hispanic participants.

Patients

Not specified

Bott U, Bott S, et al., 2000

83 patients with Patients were type 1 diabetes referred because of inability to achieve therapeutic goals for glycemic control or quality of life on an outpatient basis (e.g., frequent hypoglycemic events, motivational problems, need for diabetes knowledge refresher, need for greater flexibility in insulin treatment regimen)

5-day inpatient Not specified small-group educational sessions (20 hours) on insulin therapy, hypoglycemia, diet, psychosocial problems, and other patientspecific concerns

Patients with type 1 diabetes

Educators and others not specified

RCT=randomized controlled trial

Setting

Key Results

Retrospective, nonrandomized

None

Hospital

The hospitalization rate was reduced by both the DMI alone and DMI + DFP but the reduction was greater with the combination. DMI alone and DMI + DFP both reduced amputation rates; there was no difference in the reduction between the two groups.

1.6 years in education group; 1.5 years in control group

RCT

None

Large diabetes clinic in a metropolitan teaching hospital

A traditional diabetes patient education program failed to significantly change glycosolated hemoglobin levels, fasting blood glucose levels, lipid levels, body weight, foot lesion scores, and use of medical care in a clinic population of insulin-treated diabetics. While knowledge of diabetes and behavior was improved in program graduates, the authors concluded that patient education may not be an efficacious therapeutic intervention in most adults with type 2 diabetes.

Follow-up after a mean of 17.5 months (range 9–31 months)

Not randomized or controlled

Days of hospitalization and sick leave

Hospital

There was no change from baseline in the mean glycosylated hemoglobin at follow-up but there was a significant decrease in the incidence of severe hypoglycemia. The number of sick leave days per patient/year decreased significantly from 17.0 at baseline to 7.7 at follow-up.

Time Period Studied

Study/Evaluation Design

Diabetes foot-related hospitalizations and lower extremity amputations

2 years

Glycosolated hemoglobin levels, fasting blood glucose, weight, lipid levels, development of foot lesions, control of hypertension, use of medical care, knowledge of diabetes and behavior

Glycosylated hemoglobin levels, incidence of severe hypoglycemia (need for glucose i.v. or glucagon injection), days of hospitalization and sick leave

Disease Management for Diabetes

Economic Effects Assessed

Outcomes Measured

[43]

Appendix C. (continued)

Disease Management for Diabetes

Author(s)

[44]

Size of Population

Method of Identifying Population for Whom Data Are Evaluated

Guideline Based?

Audience for Intervention

Primary Manager of Intervention

Eight weekly 2-hour culturally sensitive diabetes education sessions (videotape presentations, discussions, demonstrations, visit to a grocery store) and one 2-hour support group session conducted in Spanish language

Curriculum guide from American Association of Diabetes Educators

Mexican American patients with diabetes and a family member or friend for support

Clinical nurse specialist, registered dietitian, and community lay worker, all of whom were Mexican American

Brown SA, Hanis CL, 1995

353 Mexican American patients with type 2 diabetes

Cabrera-Pivaral CE, Gonzalez-Perez G, et al., 2000

49 patients with Not specified type 2 diabetes not using insulin

Behavior-modifying Not specified intervention (basic and applied nutrition concepts with workshops, teamwork, and discussions)

Patients with type 2 diabetes not using insulin

Not specified

Canga N, De Irala J, et al., 2000

280 patients with type 1 or 2 diabetes registered at 2 urban hospitals and 15 urban primary care centers

A nurse-led, faceto-face, individually structured intervention to help smokers quit smoking, optional transdermal nicotine replacement therapy, and follow-up support program

Patients with type 1 or 2 diabetes

A nurse trained in smoking cessation counseling

RCT=randomized controlled trial

Random selection of seven Mexican American subjects with diabetes from a cohort of 353 patients who had been followed in another study for 12 years

Intervention Strategy

Review of clinical records and contact by telephone or written correspondence

Protocols established in How to Help Your Patients Stop Smoking: A National Cancer Institute Manual for Physicians

Outcomes Measured

Time Period Studied

Study/Evaluation Design

Knowledge of diabetes, health behaviors, body weight, metabolic control

9 weeks in 1992 or 1993 with follow-up 3 months later

Serum glucose, total cholesterol, and triglyceride levels

Economic Effects Assessed

Setting

Key Results

Pilot study of randomly None selected subjects (no control group)

County agricultural extension office in a rural community

Glycosylated hemoglobin and fasting blood glucose levels and knowledge of diabetes self-management were significantly improved 3 months after the intervention.

9 months

RCT

None

Not specified

Significant reductions in serum glucose, total cholesterol, and triglyceride levels occurred in the experimental group over the course of the study. No significant changes occurred in the control group.

Success in smoking 6 months cessation (with biochemical validation of urine cotinine concentrations) or reduction in number of cigarettes smoked daily

RCT

None

Hospitals and primary care centers

The incidence of smoking cessation was 17.0% in the intervention group and 2.3% in the control group. In subjects who failed to quit, the mean number of cigarettes smoked daily decreased significantly in both groups but the reduction was greater in the intervention group than in the control group.

Disease Management for Diabetes [45]

Appendix C. (continued)

Disease Management for Diabetes

Author(s)

[46]

Size of Population

Method of Identifying Population for Whom Data Are Evaluated

Intervention Strategy

Guideline Based?

Audience for Intervention

Primary Manager of Intervention

Health care

Health care providers and

Carlson A, Rosenqvist U, 1991

A 20% sample Diabetic patient (N=806 patients) selected center randomly from 4,492 patients with diabetes who visited 34 primary health care centers in the past year in Sweden

Continuing Not specified registry at each medical education courses for health care providers and organizational change

Clancy DE, Brown SB, et al., 2003

120 patients with poorly controlled type 2 diabetes

Electronic medical records search and telephone or on-site solicitation

Group visits with presentations, questions and answers, and individual care

American Diabetes Patients Association (ADA)

Primary care internal medicine physician and diabetes nurse educator

Coast-Senior EA, Kroner BA, et al., 1998

23 patients with type 2 diabetes who require insulin therapy

Referral by primary care providers because of inadequate glycemic control

Patient education, medication counseling, monitoring, insulin initiation or adjustment

Goals for glycemic control were in accordance with American Diabetes Association standards

Pharmacists

RCT=randomized controlled trial

patients with diabetes

Patients with type 2 diabetes referred to a pharmacist by primary care providers to improve glycemic control

Outcomes Measured

Time Period Studied

Organization 18 months providers variables (nurse involvement, resources, competence), performance of routine measurements (glycosylated hemoglobin, eye examination), patient access to and continuity of care, patient education, dietary knowledge, self-monitoring of blood glucose, metabolic control, patient-rated quality of care

Study/Evaluation Design RCT (centers, not

Economic Effects Assessed

Setting

Key Results

None individual patients,

Not specified

The intervention improved quality of care and self-monitoring of blood glucose but it did not improve metabolic control.

were randomized) with multivariate analysis by multiple regression to identify factors contributing to variance in selected organization, process, and outcome variables

Glycemic control, 6 months lipid profiles, trust in physician, patient satisfaction with care, compliance with ADA process of care indicators

RCT

None

Clinic

Glycemic control and lipids improved in both groups. Trust in physician, patient satisfaction, and compliance with process indicators were greater in the intervention group than in the control group.

Glycemic control Up to 45 weeks (glycosylated hemoglobin, fasting blood glucose and random blood glucose levels), symptomatic hypoglycemic episodes, emergency department visits, hospitalizations

Not randomized or controlled

None

Two primary care clinics at a university-affiliated Veterans Affairs medical center

Pharmacists’ efforts as part of a multidisciplinary team improved glycemic control significantly in patients with type 2 diabetes who require insulin.

Disease Management for Diabetes [47]

Appendix C. (continued)

Author(s)

Size of Population

Disease Management for Diabetes

Diabetes Control 1,441 patients and Complications with type 1 Trial Research Group. diabetes The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus, 1993

[48]

Method of Identifying Population for Whom Data Are Evaluated Not specified

Domurat ES, 1999

30% of HMO HMO members members with diabetes (2,617 patients)

Edlin M, 2003

3,118 of 6,799 health plan enrollees with diabetes

RCT=randomized controlled trial

Intervention Strategy

Guideline Based?

Intensive therapy Not specified with an external insulin pump or three or more daily insulin injections and frequent blood glucose monitoring (versus conventional therapy with one or two daily insulin injections)

Use of Diabetes Care System (DCS), a computersupported diabetes care management program

Patient self-referral Promotion of or physician referral guideline use by physicians, 1–4 annual patient visits, and patient education on self-management techniques and preventive care

Audience for Intervention

Primary Manager of Intervention

Patients with type 1 diabetes

Physicians

American Diabetes Patients with Association diabetes

Team of health care professionals

Health Plan Physicians and Employer Data and patients Information Set measures

Nurse educators

Outcomes Measured

Time Period Studied

Study/Evaluation Design

Economic Effects Assessed

Setting

Key Results

Glycosylated 6.5 years (average) Multicenter, RCT hemoglobin and blood glucose levels, eye examinations for retinopathy, urinary albumin excretion for nephropathy, clinical neurologic examination for neuropathy; lowdensity lipoprotein cholesterol, cardiovascular and peripheral vascular events

None

Not specified (presumably at home with intermittent clinic visits)

Intensive therapy delays the onset and slows the progression of retinopathy, nephropathy, and neuropathy in patients with type 1 diabetes.

Glycosylated hemoglobin, blood pressure, urinary protein, and serum lipids

3 years

Pre-post comparison

None

California HMO

Screening rates for glycosylated hemoglobin, urinary protein, and serum lipids were significantly higher in the 2,617 patients managed by DCS compared with the 5,993 patients receiving usual care. The rate of follow-up testing in patients with an elevated glycosylated hemoglobin was significantly higher in the DCS patients than in the usual care patients.

Glycosylated hemoglobin testing, lipid screening, eye exams, kidney screening; glycemic control, health care utilization

24 months

Chart review

$1,300 per member Home, clinic, and or total of $4 million hospital saved annually, which more than offsets program cost of $1.8 million

Participants had higher rates of testing, screening, and eye exams; better glycemic control; and fewer emergency department visits and inpatient days than nonparticipants.

Disease Management for Diabetes [49]

Appendix C. (continued)

Disease Management for Diabetes

Author(s)

[50]

Size of Population

Method of Identifying Population for Whom Data Are Evaluated

Intervention Strategy

Guideline Based?

Audience for Intervention

Primary Manager of Intervention

Engelgau MM, Narayan KM, et al., 1998

4,400 households in each of two communities (one with large population of African Americans at high risk and the other as a control)

Not specified

Health promotion Not specified initiatives (e.g., walking programs, publication of articles on low-fat diets); outreach efforts to increase awareness of diabetes risk factors, screening for undiagnosed diabetes, and tracking of previously diagnosed patients; selfmanagement workshops and medical record reviews and other efforts involving primary care providers to improve quality of diabetes care

Community in southeast section of Raleigh, N.C., with large African American population

Not specified (collaborative effort involving community leaders, health care providers, Centers for Disease Control and Prevention)

Gilden JL, Hendryx MS, et al., 1992

32 elderly (average age 68 yrs) male patients with diabetes at a Veterans Affairs medical center

Not specified

18 monthly Not specified support group sessions with continuing education, coping skills training, group discussions, and structured social activities

Patients

Social workers

Ginn M, Frate DA, et al., 1999

754 patients with diabetes, hypertension, or both

Not specified

Community-based Not specified case management (client assessment, care plan development, implementation, client service monitoring, and periodic assessment)

Patients with diabetes, hypertension, or both

Case managers hired from the local community and trained in methods for educating and managing patients

RCT=randomized controlled trial

Economic Effects Assessed

Outcomes Measured

Time Period Studied

Study/Evaluation Design

Patient participation in programs, screening rates, changes in diabetes practice, glycosylated hemoglobin concentration (i.e., long-term blood glucose control), patient knowledge and skills

Pilot study was conducted in 1993 (duration unspecified); Project DIRECT will take 5 years.

Process and outcome measures assessed primarily by questionnaire

None

Community Provided only for (presumably clinic) pilot study in which 902 individuals participated (45% African American); half of African Americans were physically inactive and half were overweight; African Americans had a higher prevalence of diagnosed and undiagnosed diabetes and were more likely to smoke and have uncontrolled hypertension and less likely to have single health care providers than members of other races; all diabetic patients had a low level of preventive care (eye and foot exams).

Diabetes knowledge, psychosocial factors (self-care-related quality of life, stress, family involvement in care, and social involvement), depression, and glycemic control

2 years

Partially randomized and controlled

None

Clinic

Diabetes knowledge and quality of life were better in patients who attended support group sessions than in patients who did not.

Knowledge of hypertension and diabetes, selfreported health status, blood pressure control status, diabetes control status

24 months

Uncontrolled longitudinal study

None

12 ambulatory clinics and one hospital in rural Mississippi

Knowledge about diabetes and hypertension, self-reported health status, and blood pressure control increased significantly. Blood glucose control also increased, although the difference from baseline was not significant.

Setting

Key Results

Disease Management for Diabetes [51]

Appendix C. (continued)

Disease Management for Diabetes

Author(s)

[52]

Size of Population

Method of Identifying Population for Whom Data Are Evaluated

Intervention Strategy

Guideline Based?

Audience for Intervention

Primary Manager of Intervention

Glasgow RE, Toobert DJ, et al., 1995

Approximately 200 adult patients with type 1 or 2 diabetes who are at least 40 years of age (data for the initial 95 are presented in this article)

Letter about study with informed consent statement sent to patients who have diabetes and a scheduled appointment at an office-based practice, followed by a phone call from investigator

Computer-assisted Not specified and interactive video procedures that assess patient views and behaviors related to diabetes care and selfmanagement; assistance for patients in setting dietary goals and problemsolving strategies; follow-up phone contact and office visits for support

Patients with diabetes

An endocrinologist and internist who specializes in diabetes

Glasgow RE, Toobert DJ, et al., 1996

206 adult diabetic patients

Letter about study with informed consent statement sent to patients who have diabetes and a scheduled appointment at an office-based practice, followed by a phone call from investigator

Computer-assisted and interactive video procedures that assess patient views and behaviors related to diabetes care and selfmanagement; assistance for patients in setting dietary goals and problem-solving strategies; follow-up phone contact and office visits for support

Patients with diabetes

An endocrinologist and internist who specializes in diabetes

RCT=randomized controlled trial

Diet chosen as focus because is an objective of Healthy People 2010

Study/Evaluation Design

Economic Effects Assessed

Weight, eating Not specified patterns (fat content), serum cholesterol, and glycemic control (glycosylated hemoglobin)

RCT

Weight, eating 12 months patterns (fat content), serum cholesterol, and glycemic control (glycosylated hemoglobin)

RCT

Outcomes Measured

Time Period Studied

Key Results

None

Physicians’ offices

Baseline data from only initial visit and 1- and 3-week follow-up phone calls for 51 of initial 95 patients reveal that 90% of those randomized to intervention achieved dietary goals at 1-week follow-up and 96% did so at 3-week follow-up.

None

Physicians’ offices

The intervention produced greater improvement than usual care in a number of measures of dietary behavior (e.g., fewer calories from saturated fat, fewer high-fat eating habits and behaviors) at the 3-month follow-up. There were also significant differences favoring intervention in changes in serum cholesterol levels and patient satisfaction but not in glycosolated hemoglobin. The intervention effects were relatively robust across a variety of patient characteristics, the two participating physicians, and intervention staff members.

Disease Management for Diabetes

Setting

[53]

Appendix C. (continued)

Disease Management for Diabetes

Author(s)

[54]

Size of Population

Method of Identifying Population for Whom Data Are Evaluated

Intervention Strategy

Guideline Based?

Audience for Intervention

Primary Manager of Intervention

Glasgow RE, Toobert DJ, 2000

320 adult Letter about patients with project sent by type 2 diabetes primary care providers to patients, followed by a phone call from investigator

Behavioral dietary Not specified intervention (dietary goal setting) with or without diabetes self-management support: (1) follow-up telephone calls and (2) community resources enhancement (e.g., newsletters, information on dining out and shopping)

Patients with type 2 diabetes

A nurse/certified diabetes educator, registered dietitian, doctoral-level psychologist, and an education major

Grey M, Boland EA, et al., 1999

77 adolescents with type 1 diabetes who were beginning intensive insulin therapy

Coping (problem Not solving) and skills guideline-based training (four to eight 1- to 1.5-hour weekly small-group sessions followed by monthly sessions using role-playing in various social situations)

Patients with diabetes

A master’s prepared nurse practitioner with experience in pediatric psychiatry and diabetes

Halbert RJ, Nichol JM, et al.,1999

19,523 diabetic Reviewed claims members of and pharmacy an HMO databases to identify all diabetic members of HMO ≥18 years as well as their diabetic retinal examination (DRE) status

Mailed patient reminders for screening examinations; also provided reports of patient DRE status to patients and treating physicians, educational materials to patients, and current American Diabetes Association retinal examination guidelines to treating physicians

Physicians were sent a letter explaining the program that included current guidelines

Physicians

RCT=randomized controlled trial

Not specified (selected from a university-based pediatric diabetes service)

Current ADA guidelines sent to physicians

Outcomes Measured

Time Period Studied

Study/Evaluation Design

Economic Effects Assessed

Setting

Key Results

Reach (percentage of eligible patients participating), adoption (willingness of primary care physicians to participate), patient behavior (e.g., fat intake), physiologic measures (glycosylated hemoglobin, lipids), illness-related quality of life

6 months

RCT

None

Clinic

Seventy-six percent of eligible patients and 40 of 42 primary care physicians agreed to participate. On average, there was a 50% reduction in dietary fat intake. However, there were small reductions in glycosylated hemoglobin and the ratio of total cholesterol to high-density lipoprotein cholesterol that were not significant. There was relatively little change in illness-related quality of life. Follow-up phone calls and community resources enhancement did not add to efficacy of the behavioral intervention.

Metabolic control, self-efficacy (personal competence, power, resourcefulness), impact of diabetes on quality of life, worries about diabetes, rate of severe hypoglycemic episodes

6 months

RCT (health care providers blinded to study group assignment)

None

Admitted for 1 day to university-based research center to obtain baseline data and review treatment goals and methods

After 6 months, coping and skills training (CST) improved metabolic control and quality of life in adolescents with type 1 diabetes who were initiating intensive insulin therapy. CST did not affect the rate of severe hypoglycemic episodes.

Diabetic retinal examination (DRE)

7 months pre- and RCT (no 1 year postnonintervention control intervention group, but single intervention treated as control group)

Rates of DRE were higher in the multiple intervention group than the single intervention group after a second patient reminder, but significant intergroup differences did not exist following subsequent reminders. Financial resources for multiple patient reminders for DRE may be better spent on alternative approaches for reducing complications of diabetes.

Disease Management for Diabetes

Analysis of direct HMO costs of reminders (only printing and postage) in multiple reminders group showed incremental cost of $80 per additional eye examination.

[55]

Appendix C. (continued)

Disease Management for Diabetes

Author(s)

[56]

Size of Population

Method of Identifying Population for Whom Data Are Evaluated

Intervention Strategy

Guideline Based?

Audience for Intervention

Primary Manager of Intervention

Not guideline based

Patients

Clinical pharmacist (PharmD with 2 years of clinical training in general medicine)

Patients with diabetes

Specialists (faculty diabetologists, endocrine fellows, medical residents, diabetic nurse educator, podiatrist, and optometrist) at diabetes mellitus clinic

Hawkins DW, Fiedler FP, et al., 1979

1,722 patients with diabetes, hypertension, or both

Not specified, but included all patients enrolled in a hospital-based medical follow-up clinic

Management of patients by pharmacists (with physician supervision) compared with physician management

Ho M, Marger J, et al., 1997

112 medical records for diabetes patients is the sample size in a population of 35,000 patients, not necessarily with diabetes

Randomly selected from computerized medication profiles and clinic enrollment list

Comparison is American between a Diabetes general medicine Association clinic staffed by guidelines generalists (faculty internists, medical fellows, medical residents, and nurse practitioners) and a diabetes mellitus clinic staffed by specialists (faculty diabetologists, endocrine fellows, medical residents, diabetic nurse educator, podiatrist, and optometrist)

RCT=randomized controlled trial

Study/Evaluation Design

Economic Effects Assessed

Setting

Key Results

Kept-clinic29 months appointment rate, medication compliance rate, emergency department visits, hospital admissions, blood pressure and fasting blood glucose measurements

RCT

None

Clinic

Care provided by the clinical pharmacist and care provided by the physician were equivalent in controlling blood glucose and diastolic blood pressure. The kept-clinicappointment rate was higher and the clinic dropout rate was lower in the experimental group (pharmacist-managed care) compared with the control group (physician-managed care), suggesting greater patient satisfaction with care provided by the clinical pharmacist.

Compliance with 2 years process-of-care criteria and a subset of minimally acceptable criteria chosen because of a strong link with good patient outcomes in clinical trials or importance for continuity of care

Retrospective chart review

None

Universityaffiliated Veterans Affairs medical center, general medical clinic, and diabetes mellitus clinic

Quality of care at a diabetes specialty clinic was better than at a general medicine clinic.

Outcomes Measured

Time Period Studied

Disease Management for Diabetes [57]

Appendix C. (continued)

Disease Management for Diabetes

Author(s)

[58]

Size of Population

Method of Identifying Population for Whom Data Are Evaluated

Intervention Strategy

Guideline Based?

Howorka K, Pumprla J, et al., 2000

32 and 68 adult patients with type 1 diabetes for study 1 and study 2, respectively

Patients were recruited from a population of about 400 according to eligibility criteria (e.g., duration of diabetes at least 1 year)

Huff PS, Ives TJ, et al., 1983

Not specified

Referral by primary Diabetes American care provider questionnaire Diabetes assessed Association patient knowledge of diabetes and was used to establish a teaching plan and treatment goals. An initial 1-hour counseling session was conducted to address disease pathophysiology, complications, importance of patient compliance, role of diet and drug therapy, administration technique, adverse effects, and self-monitoring. Patient progress in the educational process was documented. Additional 15-minute sessions at followup visits were provided as needed.

RCT=randomized controlled trial

Structured Not specified education about functional insulin treatment (insulin dosing according to blood glucose levels and food intake, thereby avoiding the need for adherence to a prescribed schedule of doses and meals)

Audience for Intervention

Primary Manager of Intervention

Patients with type 1 diabetes

Not specified

Patients

Pharmacists and physicians

Economic Effects Assessed

Outcomes Measured

Time Period Studied

Study/Evaluation Design

Perceived control over diabetes and diabetes-related health beliefs

4 weeks for study 1 and 3 years for study 2

RCT for study 1, uncontrolled pilot study for study 2

Perceived treatment Not specified cost-effectiveness as the difference between measures of benefits from treatment and barriers to treatment

On a short-term basis, functional-insulin-treatment (FIT) training induced feelings of independence because of situational control. After 3 years, FIT training improved perceived self-efficacy, treatment satisfaction, and glycemic control, resulting in a feeling of empowerment. Perceived treatment costeffectiveness improved significantly due to decreases in barriers to treatment.

None specified

Not specified

Survey

None

Pharmacists were able to provide more instructional time than typically is provided by physicians, thereby improving patient understanding. Patients were grateful to have ready access to pharmacists for information or help solving problems. Physicians had more time available to spend with other patients once pharmacists assumed the patient education responsibility. Communication between pharmacists and physicians improved.

Setting

Ambulatory care clinic

Key Results

Disease Management for Diabetes [59]

Appendix C. (continued)

Author(s)

Disease Management for Diabetes

Irvine AA, Mitchell CM, et al., 1992

[60]

Size of Population 61 adults with diabetes

RCT=randomized controlled trial

Method of Identifying Population for Whom Data Are Evaluated 250 patients referred by physicians were sent a letter inviting them to participate and attend initial meetings in a local church three evenings during 1 week.

Intervention Strategy

Guideline Based?

Diabetes Not specified education program to increase knowledge, selfcare, and metabolic control

Audience for Intervention

Primary Manager of Intervention

Patients with diabetes

Nurse educator, nutritionist, psychologist, and physical therapist

Outcomes Measured

Time Period Studied

Self-reported 3 months adherence to treatment regimen (diet and weight, exercise, selfmonitoring of glucose levels, medication use, and foot care), metabolic control (glycosylated hemoglobin), attitudes toward living with diabetes (self-esteem, denial, locus of control, degree to which regimen intrudes on lifestyle, attitude toward physician), knowledge of diabetes, barriers to self-care

Study/Evaluation Design Controlled, but not randomized

Economic Effects Assessed None

Setting

Key Results

Community (three Appalachian communities that were comparable in size and character)

Subjects who did not drop out of the education program and subjects in the control group who attended all three testing sessions (Attenders) had significant increases in diabetes knowledge and foot care over the course of the study (however, these Attenders may have characteristics that allow for improvement regardless of whether education is provided). Subjects who dropped out of the education program and subjects in the control group who did not attend all three testing sessions (Nonattenders) had no improvement in diabetes knowledge or foot care. These Nonattenders tended to be less educated, have a lower income, be younger, have had diabetes for twice as long, report more barriers to self-care, and have poorer health than Attenders. Special efforts may be required to promote program attendance in patients with these characteristics.

Disease Management for Diabetes [61]

Appendix C. (continued)

Disease Management for Diabetes

Author(s)

[62]

Size of Population

Method of Identifying Population for Whom Data Are Evaluated

Intervention Strategy

Guideline Based?

Audience for Intervention

Primary Manager of Intervention

Jaber LA, Halapy H, et al., 1996

39 adults

Diabetic patients currently attending a clinic who met the inclusion criteria and responded to contact via mail and telephone

Comprehensive Not specified pharmaceutical care model, including education, medication counseling, instruction about diet, exercise, and home blood glucose monitoring, and evaluation/ adjustment of hypoglycemic regimen

Urban African American clinic patients

Pharmacists

Kaplan RM, Hartwell SL, et al., 1987

70 adults

Invitations to participate via public radio announcements, newspaper notices, and physicians

Four types of intervention strategies—diet alone, exercise alone, diet plus exercise, or education alone (control)

Patients

Not specified

Karlander SG, Kindstedt K, et al., 1983

83 patients with diabetes (and another 32 patients with diabetes as controls)

All patients attending a diabetes clinic were invited to participate (32 patients with diabetes matched for age were not asked to participate and served as controls; how these patients were identified was not explained in the article).

5-day formal Not specified diabetes education program with two daily lessons about the disease and nutrition, respectively, using lecture, slides, questions-andanswers, and printed text

Patients with diabetes

Not specified

RCT=randomized controlled trial

Not specified

Time Period Studied

Study/Evaluation Design

Economic Effects Assessed

Primary outcome measures: fasting plasma glucose and glycosylated hemoglobin; secondary outcome endpoints: blood pressure, lipid levels, renal function, and quality of life

4 months

RCT

Glycosylated hemoglobin, body weight, quality of life

18 months

Knowledge about diabetes and nutrition, metabolic control (blood glucose, urinary glucose excretion, relative body weight, serum cholesterol and triglycerides, and, in some patients, glycosylated hemoglobin)

1 year

Setting

Key Results

None

A universityaffiliated internal medicine outpatient clinic

A comprehensive model of pharmaceutical care effectively improved glycemic control, but not blood pressure, lipid levels, body weight, or measured quality-of-life parameters, in a clinic-based population of urban African American patients with type 2 diabetes. Changes in glycemic control were attributed to improved patient understanding of diabetes and optimization of oral hypoglycemic regimens.

Randomized trial (no non-intervention control group, but education group as control group)

Estimated cost of diet and exercise program was $1,000 per patient; estimated cost of a single “well year” produced by this program was $10,870.

Not specified

A combination diet-plus-exercise program produced the greatest reduction in glycosylated hemoglobin, as well as significant improvement in quality of life, compared with interventions consisting of diet, exercise, or education alone. These improvements were largely uncorrelated with body weight changes.

Nonrandomized, but controlled

None

Ambulatory clinic

Knowledge of diabetes and nutrition were significantly improved immediately and 1 year after the intervention. Knowledge did not change between the initial visit and second visit 2 months later in the control group. Metabolic control was not affected by the intervention.

Disease Management for Diabetes

Outcomes Measured

[63]

Appendix C. (continued)

Disease Management for Diabetes

Author(s)

[64]

Size of Population

Method of Identifying Population for Whom Data Are Evaluated

Intervention Strategy

Guideline Based?

Audience for Intervention

Primary Manager of Intervention

Legorreta AP, Peters AL, et al., 1996

390 patients Use of ICD-9 codes with type 1 or to generate list of type 2 diabetes diabetic patients at study sites

Provider teams Not specified (nurses, physician assistants, endocrinologists) were trained in application of diabetes management program that uses protocols linked to a computer system to improve disease management and compliance with referrals and laboratory tests.

Health care Not specified who providers (nurses, assessed physician assistants, adherence to endocrinologists) protocols by received training health professional in program teams; nurses administration; managed patients were application of recipients of program to program. patients

Levetan CS, Salas JR, et al., 1995

104 patients with diabetes

Consultation provided by a diabetes team

Not specified

Patients

Diabetes team (nurse educator, registered dietitian, and endocrinologist)

Lieberman DA, 2001

59 children and Not specified adolescents with diabetes

Diabetes selfmanagement video game

Not specified

Children and adolescents with diabetes

Not specified

RCT=randomized controlled trial

Consecutive patients

Outcomes Measured

Time Period Studied

Study/Evaluation Design

Economic Effects Assessed

Setting

Key Results

Prospective, nonrandomized, controlled trial

None

Large medical groups contracted to provided health care to HMO members, including a typical participating medical group and IPA

A nurse-administered diabetes management program, employing special protocols linked to a computer system, effectively improved glycemic control, lipid levels, and referral rates in diabetic patients treated in an outpatient setting.

Length of stay

Not specified

None

Hospital

Length of stay in the diabetes team consultation group (3.6 days) was significantly shorter than in the endocrinologist consultation group (5.5 days) and the noconsultation group (8.2 days).

RCT

None

Home

There was a 77% decrease from baseline in urgent care and emergency medical visits in the diabetes video game group but no change in the control group. There were improvements in diabetes-related self-efficacy, communication with parents about diabetes, and daily diabetes self-care in the diabetes video game group but not in the control group.

Not specified

Amount of video 6 months game playing time, diabetes-related urgent care and emergency medical visits, diabetes-related self-efficacy (ability to affect outcomes), communication with parents about diabetes, daily diabetes self-care

Disease Management for Diabetes

Provider 18 months adherence to protocols by measurement of patient laboratory values (glycosolated hemoglobin, fasting plasma glucose, full lipid panel, serum creatinine, urinalysis)

[65]

Appendix C. (continued)

Disease Management for Diabetes

Author(s)

[66]

Size of Population

Method of Identifying Population for Whom Data Are Evaluated

Intervention Strategy

Guideline Based?

Audience for Intervention

Primary Manager of Intervention

American Diabetes Association guidelines for metabolic control (metabolic control was not a primary outcome measure)

Patients with diabetes receiving oral antidiabetic drug therapy

Pharmacists

Adults with diabetes who were receiving insulin therapy and had poor glycemic control

Community health nursing students in their senior year who were under the supervision of a nursing faculty investigator

Matsuyama JR, Mason BJ, et al., 1993

47 patients with Screening of poor to fair computer metabolic medication control of profiles and diabetes laboratory values mellitus of all patients receiving a sulfonylurea for 3 months or longer with a consistent dosage and poor to fair metabolic control

Use of an electronic medicationevent monitoring system (MEMS), a medication vial cap with a microprocessor that records each date and time that the vial is opened

Mazzuca KB, Farris NA, et al., 1997

22 adults with diabetes who were receiving insulin therapy and had poor glycemic control

Community Not specified health nursing intervention (weekly or biweekly home visits to provide health teaching and guidance, health referrals, coordination of care, and client advocacy; teaching addressed nutrition, exercise, foot care, and blood glucose monitoring)

RCT=randomized controlled trial

Patients who had received care at a university-based internal medicine clinic were contacted by telephone by investigators.

Outcomes Measured

Time Period Studied

Study/Evaluation Design

Economic Effects Assessed

Dates and times medication vial was opened and pill counts after 30 and 60 days

2 months

Double-blind RCT

Self-reported self-care behaviors (managing complications, blood glucose monitoring, diet, reporting foot changes, exercise), dietary adherence (3-day dietary recall), foot care (ulcer formation), blood glucose levels, diabetes knowledge, functional health status and well-being

32 weeks

RCT

Setting

Key Results

None

Veterans Affairs medical center ambulatory clinics

The MEMS data allowed pharmacists to individualize recommendations to a greater extent than the pill count method

None

Community (patient homes)

Community health nursing produced a significant improvement in self-care competency in adults with insulin-treated diabetes, although it did not improve health status.

Disease Management for Diabetes [67]

Appendix C. (continued)

Disease Management for Diabetes

Author(s)

[68]

Size of Population

Method of Identifying Population for Whom Data Are Evaluated

Intervention Strategy

Guideline Based?

Audience for Intervention

Primary Manager of Intervention

McCulloch DK, Price MJ, et al., 1998

30 practices (unclear as to the number of patients)

Not specified

Support program American comprised an Diabetes online patient Association registry and use of guidelines for eye and foot examinations, screening for microalbuminuria, and glycemic control; registry served as a reminder of the recommended elements of care; delivery of care redesigned to provide for patient group visits as well as individual visits and to establish a decentralized team of diabetes experts that sees patients jointly with primary care providers; team traveled to each clinic several times a year and, along with the primary care team, saw each patient for approximately 30–40 minutes

Patients

Physicians

Meigs JB, Cagliero E, et al., 2003

598 patients with diabetes

ICD-9 codes, with randomization by coin toss

Web-based patient-specific information management/ decision support tool

Physicians and patients

Data collection by nurses

RCT=randomized controlled trial

American Diabetes Association, National Cholesterol Education Program, and Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure

Time Period Studied

Study/Evaluation Design

Economic Effects Assessed

Glycosylated hemoglobin and eye examinations

3 years

RCT

Testing rates and values for glycosylated hemoglobin, lowdensity lipoprotein (LDL) cholesterol, and blood pressure; eye and foot examination rates

12 months

RCT (1 yr before and after intervention)

Setting

Key Results

None

Not-for-profit staff-model health maintenance organization

The prevalence of testing for glycosylated hemoglobin and eye examinations both increased over the 3-year period after implementation of the program. Nearly two-thirds of patients with diabetes received annual eye examinations by the end of the 3-year period. In the first year of the program, half of all patients with diabetes had a foot examination compared with fewer than 20% before program implementation. Microalbuminuria screening also increased markedly after program implementation. The prevalence of smoking decreased from 14% in 1994 to 10% in 1996.

None

Hospital-based clinic

Testing of glycosylated hemoglobin and LDL cholesterol and foot examinations increased and glycemic control improved in the intervention group.

Disease Management for Diabetes

Outcomes Measured

[69]

Appendix C. (continued)

Disease Management for Diabetes

Author(s)

[70]

Size of Population

Method of Identifying Population for Whom Data Are Evaluated

Nilasena DS, Lincoln MJ, et al., 1995

31 Physicians

Nine SL, Lakies CL, et al., 2003

75 African Patients Americans with volunteered to diabetes, participate hypertension, or both

RCT=randomized controlled trial

Internal medicine resident physicians in their 3rd or 4th year

Intervention Strategy

Guideline Based?

Audience for Intervention

Primary Manager of Intervention

Impact of a American computerDiabetes generated Association reminder system on physician compliance with guidelines for diabetes preventive care; residents in intervention group were provided with a patient-specific report listing patient health data and upcoming or overdue preventive health activities (e.g., physical examinations, laboratory tests, referrals, patient education)

Internal medicine resident physicians in their 3rd or 4th year

Not specified

Exercise, medical nutrition therapy, support groups, cooking schools, and service coordination

Patients

Registered nurse or dietitian

Not specified

Outcomes Measured

Time Period Studied

Study/Evaluation Design

Economic Effects Assessed

Setting

Key Results

Compliance with guidelines for preventive care

6 months

RCT

None

Hospital

The average compliance score 1 month after implementation of the reminder system was significantly higher compared with baseline in the intervention group (54.9% after 1 month versus 38.0% at baseline). The average compliance score in the control group also increased significantly over the same period (51.0% after 1 month versus 34.6% at baseline); there was no significant difference between the two groups in the change in compliance. The highest compliance scores were for laboratory tests and referrals, and the lowest scores were for patient education. The use of encounter forms facilitated documentation of clinical data and compliance with guidelines for diabetes preventive care.

Blood pressure, glycosylated hemoglobin, and quality of life

1 year

Quasi-experimental

None

Community

Glycosylated hemoglobin decreased significantly in patients with elevated initial values. Quality of life improved but the change was not significant.

Disease Management for Diabetes [71]

Appendix C. (continued)

Disease Management for Diabetes

Author(s)

[72]

Size of Population

Method of Identifying Population for Whom Data Are Evaluated

Intervention Strategy

Guideline Based?

Audience for Intervention

Primary Manager of Intervention

Registered nurses and licensed practical nurses at longterm care facilities

Diabetes educators

Parker MT, Keggett-Frazier N, et al., 1995

Three longterm-care facilities (LTCFs); two received intervention and one served as a control group

A letter was sent to administrators of 40 long-termcare facilities (the method for identifying these facilities was not specified) in eastern North Carolina that met certain criteria for number of beds and number of patients with diabetes.

Education program (7 biweekly 20-minute sessions using a lecture format with slides and a question-andanswer period on medications, patient monitoring, managing hyperglycemia and hypoglycemia, diet, foot care, illness care, and exercise and patient-family education)

American Association of Diabetes Educators and American Diabetes Association guidelines were used to develop the diabetes knowledge test

Piette JD, Weinberger M, et al., 2000

248 Englishor Spanishspeaking adult patients with diabetes managed by medication

Culled from medical records of patients with scheduled appointments at one of two general medicine clinics of a county health care system

Biweekly automated telephone assessment (patient use of the touch-tonetelephone keypad to enter blood glucose readings and other data for review by a nurse) and self-care education calls with nurse follow-up

Self-care Patients education recorded messages were based on literature published by the Centers for Disease Control and Prevention and American Diabetes Association

RCT=randomized controlled trial

Diabetes nurse educator

Outcomes Measured

Time Period Studied

Study/Evaluation Design

Diabetes knowledge and diabetes care practices (providing carbohydrates for patients with hypoglycemia, testing for ketones and contacting a physician for patients with hyperglycemia, administering insulin in the abdomen, and having a podiatrist cut toenails)

3 months

LTCFs were randomized to the intervention or control group; pretests and posttests were used to test diabetes knowledge and a retrospective chart review (randomly selected charts for LTCFs with more than 20 residents with diabetes and the entire population of patients with diabetes for LTCFs with fewer residents with diabetes) was used to evaluate diabetes care.

Self-monitoring of blood glucose, foot inspection, and weight monitoring, medication adherence, glycemic control (glycosylated hemoglobin, serum glucose)

12 months

RCT

Economic Effects Assessed

Setting

Key Results

None

Long-term-care facilities

The intervention significantly improved diabetes knowledge but did not affect diabetes care practices.

None

Home

Self-monitoring of blood glucose, foot inspection, and weight monitoring were more frequent; problems with medication adherence were less common; and glycemic control was better in the intervention group than in the control group, which received usual care.

Disease Management for Diabetes [73]

Appendix C. (continued)

Disease Management for Diabetes

Author(s)

[74]

Size of Population

Method of Identifying Population for Whom Data Are Evaluated

Guideline Based?

Biweekly automated telephone assessment (patient use of the touch-tonetelephone keypad to enter blood glucose readings and other data for review by a nurse) and self-care education calls with nurse follow-up

Self-care Patients education recorded messages were based on literature published by the Centers for Disease Control and Prevention and American Diabetes Association.

Piette JD, Weinberger M, et al., 2000

248 English- or Spanishspeaking adult patients with diabetes managed by medication

Piette JD, Weinberger M, 2001

272 patients Recruited from with diabetes at three general a Veterans medicine clinics Affairs and one diabetes clinic specialty clinic within a university-affiliated Veterans Administration health care system, using patient medical records

Biweekly Not specified automated telephone assessment (patient use of the touch-tonetelephone keypad to enter blood glucose readings and other data for review by a nurse) and self-care education calls with nurse follow-up

Pijls LT, de Vries H, et al., 2000

125 patients with type 2 diabetes and microalbuminuria, relatively high albuminuria, or diabetes for at least 5 years and high dietary protein intake

Dietary counseling

RCT=randomized controlled trial

Culled from medical records of patients with scheduled appointments at one of two general medicine clinics of a county health care system

Intervention Strategy

Not specified (described elsewhere in published literature)

Audience for Intervention

Patients

European Patients Association for the Study of Diabetes and American Diabetes Association

Primary Manager of Intervention Diabetes nurse educator

Nurse

Dietitians

Time Period Studied

Study/Evaluation Design

Economic Effects Assessed

Setting

Key Results

Depression, anxiety, self-efficacy (patients’ confidence in participating in their own care), days in bed because of illness, diabetes-specific health-related quality of life, general quality of life

12 months

RCT

None

Home

Significantly fewer symptoms of depression and days in bed because of illness and significantly greater self-efficacy to perform self-care activities were associated with the intervention (automated telephone assessment and nurse follow-up) than with usual care (the control group). Anxiety and quality of life were similar in the two groups.

Self-care (selfmonitoring of blood glucose, foot inspections), symptoms, satisfaction with care, and glycemic control

12 months

RCT

None

Clinic

Patients in the intervention group reported significantly more frequent selfmonitoring of blood glucose and foot inspections, fewer symptoms of poor glycemic control, and greater satisfaction with care. Among patients with an elevated baseline glycosylated hemoglobin, the mean value after 12 months was significantly lower in the intervention group than in the control group.

Protein intake estimated from urinary urea excretion

12 months

RCT

None

Not specified

Dietary protein intake was significantly lower in the experimental group than in the control group after 6 months but the difference was smaller and not significant after 12 months. Dietary counseling resulted in only modest protein restriction.

Disease Management for Diabetes

Outcomes Measured

[75]

Appendix C. (continued)

Disease Management for Diabetes

Author(s)

[76]

Size of Population

Method of Identifying Population for Whom Data Are Evaluated

Intervention Strategy

Guideline Based?

Audience for Intervention

Primary Manager of Intervention

Pill R, Stott NCH, et al., 1998

33 general practices with 252 patients with type 2 diabetes

Not specified

Training (at least Not specified 3 hours divided into two sessions using discussions, demonstrations, and role-play) of general practitioners and practice nurses about an unspecified patient-centered intervention (“described elsewhere”) designed to encourage clinicians to negotiate individual care plans based on patients’ perceptions of their disease and readiness to change their lifestyles and to work toward realistic targets for behavior change

General practitioners and practice nurses

A general practitioner, research nurse, and clinical psychologist

Ridgeway NA, Harvill DR, et al., 1999

28 adults

Computerized audit identified diabetic patients who visited clinic in preceding year and physicians recommended participation to those patients who met the inclusion criteria (i.e., diagnosis of type 2 diabetes, ≥20% over ideal weight, able to participate in monthly visits and comprehend presented material, inadequate glycemic control).

Six 90-minute Not specified education/behavior modification sessions offered monthly, with a follow-up session at 12 months

Patients

Registered dietitians and registered nurses who were certified diabetes educators

RCT=randomized controlled trial

Economic Effects Assessed

Setting

Key Results

Change in 3 years glycosylated hemoglobin, complications, patient satisfaction with care, functional health status, ability of health care professionals to apply the intervention

RCT

None

General practice

Although most clinicians had a moderate to good understanding of the underlying principles of the intervention, few routinely applied the principles in practice. An impact on patient outcomes and health status was not demonstrated.

Glycemic control (fasting blood glucose and glycosylated hemoglobin), lipid profiles, body weight, knowledge about diabetes, health-related quality of life (HRQOL), medication requirements

RCT

Cost of educational materials and salaries for the program

Primary care clinic

Significant but transient improvements in metabolic parameters as well as body weight occurred in patients with type 2 diabetes who participated in education/behavior modification classes. Patients who took the classes also improved their knowledge of diabetes and demonstrated significantly more knowledge than controls post-intervention.

Time Period Studied

During and 12 months after 6-month intervention

Disease Management for Diabetes

Study/Evaluation Design

Outcomes Measured

[77]

Appendix C. (continued)

Disease Management for Diabetes

Author(s)

[78]

Size of Population

Method of Identifying Population for Whom Data Are Evaluated

Intervention Strategy

Guideline Based?

Audience for Intervention

Primary Manager of Intervention

Patients

Pharmacists

Rothman R, Malone R, et al., 2003

138 adults General internal with poorly medicine practice controlled type 2 diabetes

Diabetes American education, Diabetes limited physical Association examination, initial treatment recommendations, the use of medication algorithms, and frequent patient follow-up

Schmidt SO, Burns C, et al., 2003

231 patients with diabetes at six clinics

Use of a diabetes American Diabetes Physicians management Association (ADA) flow sheet, with or without provider feedback on performance

Not specified

Smith DE, Heckemeyer CM, et al., 1997

22 older obese Advertisements women with and a patient type 2 diabetes letter

Three individualized motivational interviewing sessions (to assess attitude toward change, solve problems, assess objective data on health and behavior, delineate discrepancies between current status and goals, and develop realistic and objective goals)

Psychologists experienced in the intervention technique

RCT=randomized controlled trial

Random selection of charts

Not specified

Patients

Outcomes Measured

Time Period Studied

Study/Evaluation Design

Glycosylated hemoglobin

6 months

Retrospective, not controlled

Compliance with ADA recommended frequencies for various tests and exams and patient outcomes

2 months

Body weight, glycosylated hemoglobin, treatment adherence (group meeting attendance, maintaining and submitting diaries of eating, exercise, and home blood glucose monitoring)

16 weeks

Economic Effects Assessed

Setting

Key Results

None

University-based clinic

Glycosylated hemoglobin decreased significantly, especially in patients with high baseline levels or a recent diagnosis.

Quasi-experimental

None

Clinic

Both interventions significantly improved the performance of foot exams but had little impact on other tests and patient outcomes.

RCT

None

Home with intermittent meetings in unspecified setting

Treatment adherence and blood glucose control were significantly better in the motivational (intervention) group than in the standard (control) group. Both groups lost a significant amount of weight but the two groups did not differ in the amount of weight loss.

Disease Management for Diabetes [79]

Appendix C. (continued)

Disease Management for Diabetes

Author(s)

[80]

Size of Population

Method of Identifying Population for Whom Data Are Evaluated

Intervention Strategy

Guideline Based?

Audience for Intervention

Primary Manager of Intervention

Smith L, Weinert C, 2000

30 women with type 1 or 2 diabetes living in rural areas

Letters were sent to certified diabetes educators, and flyers about the program for distribution to patients were enclosed; the mailing list for the state chapter of the American Diabetes Association was also used to identify potential participants who then received flyers.

Program using None computer-based telecommunications technology to provide support, information, and education

Female patients with diabetes

Certified diabetes educator

Stott NC, Rees M, et al., 1996

30 primary care doctors, 33 nurses, and 200 patients

Identified family practices with doctors and nurses who are “interested and active in diabetes care” (e.g., attend continuing education sessions)

Introduction of Not specified visual agendasetting technology (charts, diaries, readiness-tochange ruler), as well as techniques in negotiation and motivational interviewing that health professionals can use to facilitate clinician-patient interactions, patient lifestyle changes, and compliance

Doctors and nurses; patients secondarily involved as recipients of technique

Not specified who managed intervention for doctors and nurses; nurses and doctors applied techniques to patients

RCT=randomized controlled trial

Outcomes Measured

Time Period Studied

Study/Evaluation Design

Economic Effects Assessed

Measures of social support, quality of life, psychosocial adaptation to illness, attitudes about the impact of the computer intervention

10 months

RCT

Uptake of training, use of the method, attendance/ participation in group discussions, willingness to have consultations tape-recorded

3 years

RCT

Setting

Key Results

None

Patient homes, many of which were far from the nearest source of health care

Quality of life scores were higher for women in both groups who had more education and improving health. Attitudes toward the use of telecommunications technology were positive, and 77% of subjects recommended its use “very highly.” The program provided a great deal of support to 77% of women in the computer group.

None

29 family practices in Wales

A high percentage of family practice-based clinicians reported frequent (71%) or occasional (22%) use of agenda-setting technology in their interactions with type 2 diabetics following an intervention directed at introducing and promoting this technique. While levels of engagement were higher among nurses, doctors also reported benefits of using visual charts and techniques, negotiation, and motivational interviewing to encourage patient compliance and lifestyle changes.

Disease Management for Diabetes [81]

Appendix C. (continued)

Disease Management for Diabetes

Author(s)

[82]

Size of Population

Method of Identifying Population for Whom Data Are Evaluated

Intervention Strategy

Guideline Based?

Audience for Intervention

Primary Manager of Intervention

Tan ASL, Yong LS, et al., 1997

278 patients with diabetes (183 for the intervention) out of a total population of about 2,500 with diabetes served by the clinic

Recruited as they came to the clinic for treatment (first 100 assigned to control group, subsequent patients to the intervention group)

Diabetes Not specified education program (at least six individual or small-group counseling sessions using patient booklets and four group sessions, with lectures, videotaped presentations, group discussions, and food displays)

Patients

Nurse trained in counseling techniques

Teufel NI, Ritenbaugh CK, 1998

Not specified (it is unclear how many patients participated for the first 2 years of this 4-year study and provided longitudinal data for that period)

Not specified

Establishment of Not specified supportive social networks, development of a wellness (exercise) facility, incorporation of diabetes education into the high school curriculum, and modification of the food supply to reduce fat content and consumption of sugary beverages and increase fiber content

Zuni high school students in New Mexico, a population with a high prevalence of type 2 diabetes and an early age at diagnosis

Collaboration between public school district and university department of family and community medicine

RCT=randomized controlled trial

Outcomes Measured

Time Period Studied

Study/Evaluation Design

Economic Effects Assessed

Setting

Key Results

Controlled, but not randomized

None

Government primary health care clinic

The intervention (diabetes education program) significantly improved diabetes knowledge, dietary practices, medication compliance, and glycemic control in patients with diabetes.

Body mass index, dietary intake of fiber and sugary beverages, heart rate, fasting and postprandial glucose and insulin levels

Not controlled

None

Two high schools (community)

Within 2 years after the intervention, the consumption of sugary beverages, body mass index, and heart rate had decreased (a decreased heart rate suggests improved cardiovascular fitness), and dietary fiber intake and glucose-toinsulin ratios had increased (an increase in the glucose-to-insulin ratio suggests a reduced prevalence of hyperinsulinemia and risk for type 2 diabetes). Only the changes in sugary beverage consumption and insulin levels were significant, but the other changes suggest the adoption of healthy behaviors that might reduce the prevalence of risk factors for type 2 disease as the study continues.

4 years (results available only for first 2 years)

Disease Management for Diabetes

Diabetes 1.5 years knowledge (disease, complications, management, and self-care), dietary practices, medication compliance, selfmonitoring of blood or urine glucose, glycosylated hemoglobin

[83]

Appendix C. (continued)

Author(s)

Disease Management for Diabetes

Uusitupa MIJ, 1996

[84]

Size of Population

Method of Identifying Population for Whom Data Are Evaluated

86 obese Not specified Finnish patients

RCT=randomized controlled trial

Intervention Strategy

Guideline Based?

Intervention group Not specified made six visits to an outpatient clinic at 2-month intervals for intensive therapy with (1) a restricted dietary intake of calories, total fat (<30%), saturated fats (<10%), and cholesterol (<300 mg/day), which was monitored by examining food records and measuring serum lipids, and (2) exercise training (30–60 minutes three to four times per week), which was monitored by analyzing daily exercise records.

Audience for Intervention

Primary Manager of Intervention

Patients

Not specified

Outcomes Measured

Time Period Studied

Study/Evaluation Design

Economic Effects Assessed

Setting

Key Results

Dietary intake of calories via food records, serum lipids, and exercise training

1 year and 3 months

RCT

None

Outpatient clinic

The intervention led to a reduction in intake of saturated fats and calories. Weight loss and beneficial changes in metabolic control (fasting blood glucose) and lipid profile (e.g., high-density lipoprotein cholesterol, triglycerides) were observed during the intervention period only in the intervention group. Follow-up assessment 1 year after the completion of the intervention period revealed that glycemic control was maintained by a significantly larger percentage of patients in the intervention group than in the conventional treatment group.

Disease Management for Diabetes [85]

Appendix C.

Disease Management for Diabetes

(continued)

[86]

Author(s)

Size of Population

van den Arend IJ, Stolk RP, et al., 2000

243 patients with type 2 diabetes

RCT=randomized controlled trial

Method of Identifying Population for Whom Data Are Evaluated All eligible patients were included or some were selected randomly or based on poor glycemic control (depending on the program)

Intervention Strategy

Guideline Based?

Audience for Intervention

Primary Manager of Intervention

Four structured programs for patient care, including (1) care guided by protocol, with oral and written patient information, (2) care with computerized support and a diabetes nurse available to answer patient questions, (3) a special educational program focusing on blood glucose monitoring for a subset of patients with poor glycemic control, and (4) an integrated care program with an educational program with didactic elements and interactive group discussion on basic self-care skills and diabetes pathophysiology and complications

Guidelines on type 2 diabetes of the Dutch College of General Practitioners

Patients

General practitioners, dietitians, diabetes nurses, ophthalmologists, podiatrists

Outcomes Measured

Time Period Studied

Study/Evaluation Design

Disease knowledge, self-care behavior (dietary adherence, self-monitoring of blood glucose, physical exercise, and foot inspection), and disease perception

12 months

Randomized, but not controlled

Economic Effects Assessed None

Setting

Key Results

Primary care setting

Disease knowledge and self-care behavior increased in all four programs and was maintained at follow-up. The increases were significantly greater in programs with an educational component (programs 3 and 4) than in those without an educational component (programs 1 and 2).

Disease Management for Diabetes [87]

Appendix C. (continued)

Disease Management for Diabetes

Author(s)

Size of Population

Intervention Strategy

Guideline Based?

Audience for Intervention

Primary Manager of Intervention

Weinberger M, Kirkman S, et al., 1995

275 adults receiving care at Veterans Affairs medical clinic

Computerized Nurse-initiated Not specified audit identified telephone contacts patients who between clinic had ever filled visits to primary a prescription for care physicians to insulin or an oral provide education, hypoglycemic reinforce agent at the compliance, pharmacy and had monitor health visited the clinic in status, solve the preceding year. problems, and Patients who met facilitate access the inclusion criteria to primary care (i.e., diagnosis of type 2 diabetes, currently using insulin or oral hypoglycemic agent, receiving regular care at clinic, access to phone) were invited to participate by letter, phone, or clinic visit.

Patients

Nurses

Whitlock WL, Brown A, et al., 2000

28 adult patients with type 2 diabetes and glycosylated hemoglobin >8%

Recruited based on glycosylated hemoglobin value from internal medicine, family practice, and primary care clinics using hospital information systems at an Army medical center

Weekly home American telemedicine Diabetes (voice and video Association interaction) visits with patient by nurse case manager to review blood glucose levels, body weight, blood pressure, hypoglycemic episodes, exercise and nutrition goals, and well-being, and monthly physician telemedicine visits

Patients

Nurse case manager under direction of primary care physician

Wing RR, Venditti E, et al., 1998

154 non-diabetic individuals at risk for diabetes because of overweight and a parental history of the disease

Newspaper advertisement for overweight individuals with a parent who has diabetes

Group meetings Not specified about diet, exercise, or both; restricted diet with structured meal plans; group exercise sessions plus individual exercise

Non-diabetic individuals

Behavior therapist, registered dietitian, exercise physiologist

RCT=randomized controlled trial

[88]

Method of Identifying Population for Whom Data Are Evaluated

Time Period Studied

Study/Evaluation Design

Economic Effects Assessed

Glycosolated hemoglobin, fasting blood sugar health-related quality of life (HRQOL), and symptoms

1 year

RCT

Laboratory tests (glycosylated hemoglobin, lipid values), total body weight, quality of life

3 months

Body weight, oral glucose tolerance test, fasting glucose and insulin, glycosylated hemoglobin, lipid profile, blood pressure

2 years

Setting

Key Results

None

Veterans Affairs general medical clinic

The intervention modestly improved glycemic control but not HRQOL or diabetes-related symptoms.

RCT

None

Home

There were significant reductions in glycosylated hemoglobin and total body weight in the telemedicine group but not in the control group.

RCT

None

Not specified

Diet and diet plus exercise produced significant weight loss and improvement in cardiovascular risk factors (lipid profile, blood pressure) initially but these improvements were not maintained on a long-term basis. The risk for type 2 diabetes was significantly reduced by even a modest weight loss.

Disease Management for Diabetes

Outcomes Measured

[89]

Appendix D. Diabetes Disease Management Programs Group Health Cooperative of Puget Sound Seattle, Washington

U.S. Public Health Service Carville, Louisiana

Group Health Cooperative of Puget Sound, a Seattle-based HMO, is developing a “clinical road map” for members with diabetes that includes a sophisticated, individualized method of patient education and empowerment and a restructuring of clinic operations. A randomized, controlled trial is under way to compare outcomes in three groups: one using a new educational model in conjunction with a diabetes clinic, another incorporating the model into primary care practices, and a control group receiving usual care with no educational intervention. The educational model is based on the transtheoretical model of behavior change. Intervention guidelines have been developed for each of five stages of behavior change (precontemplation, contemplation, preparation, action, maintenance).

Reducing lower-extremity amputations in patients with diabetes by 40% is an objective of a project called the Lower Extremity Amputation Prevention (LEAP) Program, which is headquartered at the Gill W. Long Hansen’s Disease Center in Carville, Louisiana. Annual foot screenings, patient education, proper footwear selection, daily self-inspection of the feet, and management of simple foot problems are components of the LEAP program. A 25% reduction in lower-extremity amputations has already been reported, although additional progress is sought. A study of a patient empowerment program in which patients self-screen for loss of sensation in the feet is under way at 40 sites in 18 states.

Texas A&M School of Rural Public Health Bryan, Texas Baylor College of Medicine Houston, Texas Improving the health of Mexican Americans, who have high rates of obesity and are at high risk for type 2 diabetes, was the objective of Unidos en Salud (United in Health), a 5-year research project conducted by Baylor College of Medicine. The researchers focused on a population of 40,000 people who were primarily (97%) Hispanic and poor in Starr County, which borders northern Mexico. The project was built on the results of Cuidando el Corazon (Caring for Your Heart), an earlier research project that involved culturally sensitive approaches to reducing dietary fat intake and other dietary modifications.

Disease Management for Diabetes

A randomized, controlled study was undertaken to determine whether 6 months of professional nutrition, lifestyle management, and exercise interventions followed by 6 months of social support and peer support as maintenance would result in an individual weight loss of at least 10 pounds in 300 women. Other outcome measures include body mass index, serum glucose, insulin, glycosylated hemoglobin, lipids, and blood pressure. Women were randomized to the intervention or a wait-listed group (the latter served as a control group).

[90]

Many women did not reach the goal of losing at least 10 pounds during the intervention period, although some women achieved success during the maintenance period. Peer support provided significant benefits. Women with the least education and income appeared to benefit most from the social support.

A 3-year study by researchers at Texas A&M’s School of Rural Public Health is under way to explore cost-effective strategies for bringing disease management to rural populations. A model developed at Mahomet, Illinois–based Carle Health Systems Research Center was used. This model involves branch clinics to improve patient access to health care in rural areas, a care team made up of a registered nurse partnered with 5–10 primary care providers, and the use of authoritative clinical guidelines. Another model of care under study by the Texas A&M researchers is the St. Elizabeth of Hungary Clinic in Tucson, Arizona, which takes care of many uninsured patients with diabetes. A diabetes clinical flow sheet is used to track clinical assessments, laboratory test results, interventions, physician exams, and self-care activities for each patient. Diabetes day group visits are used to make care more accessible or convenient for patients and to improve compliance (patients can obtain all needed services in a single visit instead of multiple visits). The average glycosylated hemoglobin measurement decreased from 8.9% to 8.2% over a 2-year period as a result of the intervention. Glycemic control improved over the course of 1 year in nearly half of patients with diabetes who were considered at high risk. Plans are under way to use telemedicine for patients with diabetes in rural areas and to implement preventive initiatives (e.g., nutrition counseling) for patients at risk for developing diabetes (i.e., patients with prediabetes). See: The new frontier: delivering cost-effective DM to rural populations. Dis Manag Advis. 2003;9:122–26.

Appendix D. Diabetes Disease Management Programs

(continued)

Gottleib Memorial Hospital Melrose Park, Illinois

UnitedHealthcare Diabetes Management Program Minneapolis, Minnesota

Gottlieb Memorial Hospital offers a subcapitated disease management program for diabetes to the 200 physician members of its physician/hospital organization. The program uses a multidisciplinary approach to manage diabetes and reduce the incidence of complications. About 100 patients have been enrolled. Dramatic improvements in glycemic control have been observed with the program. There were no hospital admissions because of uncontrolled diabetes in the first 6 months of the program. Annual cost savings of $750,000 are projected, based on 50% reductions in both hospital admissions and hospital days for diabetes.

Visions for Diabetes was a UnitedHealthcare program designed to help members with diabetes take better care of themselves and prevent serious complications. It sought to promote physician compliance with guidelines established by the American Diabetes Association, improve quality of care and clinical outcomes, and modify member and physician behaviors. The company delivered a comprehensive diabetes health management program in collaboration with Merck-Medco Managed Care, LLC. The International Diabetes Center, a world leader in the development of education and treatment models to improve the health and lives of people with diabetes, provided continuing medical education credits for provider “rollout” and case management training. UnitedHealthcare also has a disease management program for asthma and a wellness program for pregnancy. (For additional information, go to http://www.unitedhealthcare.com.)

Disease Management for Diabetes [91]

Disease Management for Diabetes

References

[92]

1. Coons SJ. Disease management: definitions and exploration of issues. Clin Ther. 1996;18(6):1321–26. 2. Rossiter LF, Whitehurst-Cook MY, Small RE, et al. The impact of disease management on outcomes and cost of care: a study of low-income asthma patients. Inquiry. 2000;37(2):188–202. 3. Bernard S, Bernard M, Donnica L. Live long and prosper with disease management. Managed Healthcare. 1997;7(1):42–4. 4. Terry K. The disease management boom. Med Econ. 1997;74(14):62–4. 5. Lewis A. The myths and realities of disease management. Managing Employee Health Benefits. 1997;5(2):10–19. 6. Novartis Pharmacy Benefit Report: Facts and Figures. 1999 edition. 7. Gore M. Industry partnerships: disease management programs flourish. J Manag Care Pharm. 1995;1:154–72. 8. The Genesis Report. Disease management: compelling yet repelling. 1997;4(7):17–25. 9. Disease Management Association of America. Definition of disease management. Available at: http://www.dmaa.org/definition.html. Accessed November 30, 2003. 10. Centers for Medicare and Medicaid Services. Discussion of disease management. Available at: http://cms.hhs.gov/media/press/release.asp?Counter=418. Accessed November 30, 2003. 11. National Pharmaceutical Council. Medicaid disease management & health outcomes: what is disease management? Available at: http://www.dmnow.org/. Accessed November 29, 2003. 12. National Pharmaceutical Council. Medicaid disease management & health outcomes: four steps to implement a Medicaid disease management program. Available at: http://www.dmnow.org. Accessed November 29, 2003. 13. Joint Commission on Accreditation of Healthcare Organizations. Disease-specific care certification. Available at: http://www.jcaho.org/dscc/index.htm. Accessed November 30, 2003. 14. National Committee for Quality Assurance. NCQA disease management accreditation/certification information. Available at: http://www.ncqa.org/Programs/Accreditation/DM/dmmain.htm. Accessed November 30, 2003. 15. Final NCQA DM accreditation standards hit the street. DM News. December 25, 2001;7(5):1,4,5. 16. American Accreditation HealthCare Commission. URAC accreditation programs. Available at: http://www.urac.org. Accessed May 22, 2003. 17. Novartis Pharmacy Benefit Report: Trends and Forecasts. 1998 edition. 18. American Association of Health Plans. 2002 AAHP Annual Survey of Health Plans: Chart Book of Findings. Available at: http://www.aahp.org/Content/NavigationMenu/Inside_AAHP/AA HP_Surveys/2002Survey.pdf. Accessed October 29, 2003. 19. National Pharmaceutical Council. Medicaid disease management & health outcomes: four steps to implement a Medicaid disease management program. Available at: http://www.dmnow.org. Accessed November 30, 2003. 20. Gillespie JL, Rossiter LF. Medicaid disease management programs: findings from three leading US state programs. Dis Manage Health Outcomes. 2003;11:345–61.

21. The Health Strategies Consultancy, LLC, and Duke University, The Fuqua School of Business. Medicaid cost containment and potential effects on diabetic beneficiaries [white paper]. October 2003. Available at: http://www.healthstrategies.net/about/archive/110403.html. Accessed November 18, 2003. 22. American Diabetes Association. Diabetes statistics. Available at: http://www.diabetes.org. Accessed October 14, 2003. 23. U.S. Department of Health and Human Services. Healthy People 2010 Leading Health Indicators. Available at: http://www.healthypeople.gov/LHI/Touch_fact.htm. Accessed October 20, 2003. 24. American Diabetes Association.. Economic costs of diabetes in the U.S. in 2002. Diabetes Care. 2003;26:917–32. 25. Oki JC, Isley WL. Diabetes mellitus. In: DiPiro JT, Talbert RL, Yee GC, et al., eds. Pharmacotherapy: A Pathophysiologic Approach. 5th ed. New York, NY: McGraw-Hill; 2002:1335–58. 26. American Diabetes Association. Report of the expert committee on the diagnosis and classification of diabetes mellitus. Diabetes Care. 2003;26(suppl 1):S5–20. 27. The Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med. 1993;329:977–86. 28. American Diabetes Association. Standards of medical care for patients with diabetes mellitus. Position statement. Diabetes Care. 2003;26(suppl 1):S33–50. 29. American Diabetes Association. Implications of the Diabetes Control and Complications Trial. Position statement. Diabetes Care. 2003;26(suppl 1):S25–7. 30. American Diabetes Association. Implications of the United Kingdom Prospective Diabetes Study. Position statement. Diabetes Care. 2003;26(suppl 1):S28–32. 31. UK Prospective Diabetes Study Group. Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet. 1998;352:837–53. 32. UK Prospective Diabetes Study Group. Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes (UKPDS 34). Lancet. 1998;352:854–65. 33. American Diabetes Association. Evidence-based nutrition principles and recommendations for the treatment and prevention of diabetes and related complications. Position statement. Diabetes Care. 2003;26(suppl 1):S51–61. 34. American Diabetes Association. Physical activity/exercise and diabetes mellitus. Position statement. Diabetes Care. 2003;26(suppl 1):S73–7. 35. American Diabetes Association. The pharmacological treatment of hyperglycemia in NIDDM. Diabetes Care. 1996;19 (suppl 1):S54–61. 36. Physicians’ Desk Reference. 57th ed. Montvale, NJ: Thomson PDR; 2003. 37. Murphy EJ, Davern TJ, Shakil AO, et al., Troglitazone-induced fulminant hepatic failure. Acute Liver Failure Study Group. Dig Dis Sci. 2000;45:549–53.

The National Pharmaceutical Council 1894 Preston White Drive Reston, VA 20191-5433 Phone: 703-620-6390 Fax: 703-476-0904 www.npcnow.org www.dmnow.org 1DXM0140204