The Role of Participation in Disease Management Programs in Improving Glucose Control in Older Adults to Middle-Aged Non-Insulin Dependent Diabetics:

A Review of the Literature

 

 

 

 

 

 

 

 

 

Thomas A. Clobes, C.M.R.

May 13, 2005

 

 

 

 

 

 

 

 

 

University of Medicine & Dentistry of New Jersey

School of Health Related Professions

IDST6400: Critical Literature Review & Scientific Writing

Spring 2005

            Diabetes mellitus is a frustrating condition to treat for all involved: healthcare providers, patients, and even managed care organizations. Providers struggle because they know their diabetic patients are at significantly higher risk for heart disease, nephropathy, retinopathy, neuropathy, and even amputation (Beckman, Creager, & Libby, 2002; Diabetes Research Working Group, 1999). Diabetes is a challenge for patients as well because of the need to monitor their diets, measure blood sugars, maintain an exercise regimen, adhere to medicine schedules, and so forth (Saudek, 2000, p. S664).

The burden begins for managed care organizations with diabetes being one of the most common chronic diseases in the United States. It is estimated that nearly 8% of adults older than 20 years of age and 19% over age 60 suffer from this condition (Harris et al., 1998). Likewise, the National Institutes of Health estimates that 10% of healthcare spending goes directly to managing diabetes (1999).

The most prevalent form of diabetes in the United States is non-insulin dependent diabetes (NIDD), also known as type 2 diabetes. Somewhere between 90 to 95% of diabetes cases are of this type (Wild, Roglic, Sicree, & King, 2004). There are several theories to why this is true. The most common and widely accepted theory is the association between obesity and insulin resistance. Approximately 80% of non-insulin dependent diabetics are obese and the United States has been classified as being in an epidemic of obesity (Khaodhiar, McCowern, & Blackburn, 1999). The connection between obesity and NIDD and the current state of obesity in the United States is bringing another epidemic upon the healthcare community.

The combination of long-term negative patient outcomes from uncontrolled diabetes, the challenges placed on patients, and the financial burdens on managed care have been described as “a matter of national urgency” (Saudek, 2000). This urgency requires efforts beyond typical office visits and routine care. Disease management programs offer patients a level of care beyond what is typically offered to help them with the issues of this chronic condition.

                Unlike the traditional system of medicine that focuses on treating specific problems, disease management programs treat a disease comprehensively. With the example of diabetes, disease management includes patient education, clinician guidelines, regular laboratory tests, referrals to podiatrists and ophthalmologists as needed, provide glucose-testing supplies, and so forth. This is in contrast to traditional medicine that may only occasionally check laboratory values or treat podiatric problems as they arise, independently of any association to the patient’s diagnosis of diabetes.

            Providers benefit from disease management programs because of the assistance from other members of the healthcare delivery system and improved patient outcomes. Kaiser Permanente’s disease management program for diabetic patients offers providers assistance from dietitians, physical trainers, and diabetes support groups; all who can help the patient and physician achieve treatment goals (Dudl & Christensen, 2005). The disease management program at One Health has led to an increase of 20% in their patients meeting minimum laboratory standards and glycosylated hemoglobin A (HbA1c) has decreased by two points for a significant number of their patients (Bootman & Schulman, 2002). HbA1c has become the gold-standard measure of diabetic control and is an indication of one’s average glucose level over the past 2 to 3 months (Goldsetin & Wordell, 2002). A major trial on non-insulin dependent diabetics, the United Kingdom Prospective Diabetes Study concluded that reductions in HbA1c are correlated to significant reductions in risk of myocardial infarction, nephropathy, retinopathy, and neuropathy (UKPDS Group, 1998). Therefore, HbA1c is an accurate measure of patient outcomes in NIDD patients.

                Managed care organizations benefit not only from the improved patient outcomes, but also from reduced patient costs. On the surface, providers may be ordering more laboratory testing and seeing more specialists, but the cost savings come in from reducing the long-term complications such as nephropathy and serious cardiac issues. CarePatterns, the diabetes disease management program institute by Caremark, Inc. reduced medical expenditures of participants by 10% after 2 years (Goldstein & Wordell, 2002).

            Current research on the topic of NIDD and disease management has been focusing on novel techniques of administering programs such as delivery methods and provider training. The overall success of disease management programs in the application of diabetes has led researchers to inquire if better results can be reached with various administrations.

            However, little research has focused specifically on the middle-aged NIDD patient. This age range, 30 to 65 years, has the second highest prevalence rate of diabetes in the United States (Harris et al., 1998). Though some research has included patients in this age range, it has included older patients who are in a different disease pathology stage. Further, these middle-aged patients are more likely to benefit from a disease management program since they are more likely to be newly diagnosed and therefore less likely to already have long-term complications. With the older-adult and elderly patients, though NIDD may be more prevalent, they are more likely to already have irreversible complications and therefore have limited gains to make from a disease management program.

            This review will provide an analysis of the available literature on the role of disease management programs in middle-aged NIDD patients. This review was guided by the question "In adult male and female patients, ages 30 to 65, with non-insulin dependent diabetes, does participation in disease management programs significantly improve patient outcomes as measured by HbA1c.”

 

Methods

            The search for relevant articles was conducted using OVID and the following databases: Cochrane Database of Systematic Reviews, ACP Journal Club, Database of Abstract of Reviews of Effects, Cochrane Register of Controlled Trials, and Medline 1966 through present. Terms used in the search included non-insulin dependent diabetes mellitus, patient compliance, disease management, glycosylated hemoglobin A, and blood glucose mapped to their MESH terms including subheadings. Boolean “OR” was used to combine patient compliance and disease management as different authors could use them interchangeably. Boolean “OR” was used to combine glycosylated hemoglobin A and blood glucose as they too could be used interchangeably. Since both the categories of NNIDD and disease management/patient compliance are rather broad and extensively studied, Boolean “AND” was used to combine non-insulin dependent diabetes to these two searches. The search was further narrowed by age, limiting studies including adult (19 to 44) and middle age (45 to 64). Restrictions were placed on articles only in the English language. Articles not available at the UMDNJ libraries were reviewed only if available for a cost of $10.00 or less per article and received by the submission date of this review.

The articles were then manually screened for additional criteria. The OVID age limit function did not allow for the narrower age range of this review: 30 to 65, so articles were further eliminated based on participant ages. The measured intervention had to be at least one component of a disease management program aimed at improving patients with NIDD. However, the actual component could be any component the researchers identified as a disease management effort. Additionally, blood glucose end points had to be measured by HbA1c.

The table below represents the number of articles found at various levels of filtering. Due to the limited number randomized controlled trials (RCTs) with a true control group and adequate follow-up, this review contains other articles as well as long as all other criteria were met. From the original 153 retrieved articles, there are 8 articles reviewed: 4 RCTs, 1 retrospective analysis, 1 prospective design, and 1 cross-sectional study.

Initial articles retrieved via unfiltered search

153

Articles retrieved after filtering for RCTs

46

“RCT filtered” articles with true control group and meeting above inclusion criteria

8

“RCT filtered” articles with true control group, meeting inclusion criteria, and adequate follow-up

4

 

Literature Review

            The articles to be discussed in this review are broken down into two categories: moderate intervention and extreme intervention. This classification is based on the intensity of disease management intervention the patients in the study underwent, as can be seen in the table below. Within each category, articles are organized from strongest evidence to weakest.

Moderate Intervention
Intense Intervention

Relatively little daily, weekly, or monthly time commitment (i.e. short walks or quickly checking blood sugar)

Relatively high daily, weekly, or monthly time commitment (i.e. long exercise regimens or monitoring several health factors)

Patients did not have to leave house or, if they did, visits were not in addition to normal office visits

Patients had to leave house and/or make extra visits in addition to their normal office visits

Visits with consults 20 minutes or less (or only one time if longer)

Visits with consults greater than 20 minutes

 

Moderate Intervention

                Nurse practitioners and physician assistants are taking increasing larger roles in the healthcare delivery system since they are often have less demanding schedules and can spend more time with their patients than their physician counterparts. Litaker et al. (2003) compared the impact of nurse practitioner-physician team (NP-MD) administering a disease management program to physicians alone (MD only) administering it on HbA1c levels in a randomized controlled trial. Patients assigned to the NP-MD team received disease management established following guidelines from the American Diabetes Association and the Joint National Committee for the Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. Patients assigned to MD only received their usual care. This study included 157 subjects; 79 in the NP-MD and 78 in the MD only. Both men and women were in this study with an age range of 41 to 64 with all subjects accounted for at the end of the study. There was a high percentage of African-American patients (54% in the NP-MD; 64% in the MD only). Both groups had average school years completed just beyond high school. The two groups were similar on all important characteristics. Measurements were taken at baseline, 12 months after program enrollment, 6 months after program completion, and 12 months after program completion. For the entire study period, the NP-MD group had a significantly greater reduction in their HbA1c levels of 0.63 ± 1.5 versus 0.15 ± 1.0 for the MD only group (p=0.02). However, by study end, 12 months after completing the disease management program the NP-MD group’s HbA1c levels had returned to their baseline levels and comparable to the control group. This was a well-designed study because of the carefully selected patients; patients were specifically screened for pathology stages that were not too advanced to allow them to benefit from the disease management program. Additionally, a medical review committee ensured a consistent protocol was implemented between groups. These findings represent that though the extra time NPs can spend with their patients may translate into improved outcomes for NIDD patients, the effects of the program do not last after it is ceased.

Stress is a valid component of all aspects of health. Surwit et al. (2002) conducted a randomized controlled trial examining the effects of adding stress management to 5 weekly standard diabetes education courses on HbA1c levels. The study consisted of 108 subjects; 48 in the control and 60 in the intervention groups. Both men and women were included with an age range of 39 to 65 and all patients were accounted for at study end. Caucasian, African-American, and Asian were all represented. No information was reported on education level of subjects. The groups were equivalent on all characteristics except for baseline HbA1c levels. Measurements were taken at baseline, 2, 4, 6, and 12 months. By month 12, the control group had a slight increase in their HbA1c levels of 0.1 while the stress management group had a small, but statistically significant decrease of 0.5 (p=0.04). No confidence intervals or standard deviations were stated. The findings of this study are particularly interesting because the patients in the control group were still in a disease management program and one would have expected their glycemic control to improve, not worsen. Further, both groups had improvement until the 6-month mark, then the control group’s HbA1c levels suddenly deteriorated. The difference in baseline HbA1c levels is another area of concern. The intervention group started the study with higher values; actually, the control group was only 0.54 above what the American Diabetes Association defines as normal while the intervention group was 1.14 above normal (ADA, 2004). These findings loosely support the addition of stress management to general disease management courses for NIDD, they also call into question the benefit of general disease management courses. Since education level was not reported, it is possible level of the control group affected their ability to fully comprehend the disease management program.

Exercise is an important component of diabetes management. Natarajan, Clyburn, and Brown (2002) conducted a cross-sectional design evaluating the relationship between various exercise stages and glycemic control. Three levels of exercise were used as the treatment groups: exercising less than 6 months, exercising 6 months to 5 years, and exercising greater than 5 years. Three levels of controls were used: no exercise, considering exercise, preparing to exercise. HbA1c levels were compared between all groups. 74 subjects participated in the study. Both men and women were represented, however nearly three-fourths of the participants were women. The age range was 30 to 65. Caucasian and African-American races were represented, with over three-fourths of the subjects African-American. Education level ranged from less than high school to more than high school, with the majority not completing high school. The different groups were similar on all characteristics. Patients responded to questionnaires on exercise levels as follows: no exercise (31%), considering exercise (9%), preparing to exercise (24%), exercising less than 6 months (8%), exercising 6 months to 5 years (22%), and exercising greater than 5 years (3%). The highest HbA1c values were seen with the patients in the “no exercise” (9.4%) group and the lowest in the 6 months and more groups (7.3% and 7.5%). HbA1c levels for patients in the “no exercise” and “consider exercise” were significantly higher than levels for patients who had been exercising for 6 months and longer (p<0.05). These data support that regular exercise, as part of a disease management program will help to control blood glucose levels over extended periods of time. However, this was a non-interventional design and has inherent weaknesses because of it. Relying on patients to report their exercise stage accurately seriously limits the validity of the study. The higher percentage of African-Americans should not have affected the ability to generalize these results to other populations since African-Americans are generally at higher risk for NIDD and harder to treat. Thus, if anything, other populations may respond even better. The higher percentage of females may limit the generalizability of the findings; however, it is unclear whether gender-specific responses would occur.

 

Intense Intervention

            Many disease management programs focus their efforts on education patients on the risk factors of the disease and how to monitor and prevent them. Rachmani, Levi, Slavachevski, Avin, and Ravid (2002) used a randomized prospective study to evaluate the role of teaching patients to monitor their risk factors on HbA1c levels. Patients were randomly assigned to either standard consultation (SC) or patient participation (PP). Patients assigned to the PP group were given responsibility for complying with their medications, monitoring treatment effects, requesting treatment changes, and achieving/maintaining targeting blood pressure, cholesterol, HbA1c, and weight through a teaching session. PP subjects also had the option to initiate a visit or telephone conversation with a diabetic consultant when they thought necessary. Both groups received an annual consultation with a diabetic consultant and physician for four years. 141 subjects participated in the study; 70 in the SC and 71 in the PP groups. Both men and women were included with an age range of 45 to 65 and all patients were accounted for at study end. No information was reported on education level or race of participants. Both groups had similar baseline characteristics. Measurements were taken at baseline and annually through year four. Patients in the PP group reduced their HbA1c levels by 0.8 by year one while the SC has only reduced their by 0.4 (p<0.05), the decrease continued the year four with the PP group down a cumulative decrease of 1.3 ± 1.5 and the SC group down 0.7 ± 1.2 (p<0.05). There were several key strengths to this study, in addition to its randomized design. The greatest strength is it length of follow-up with the patients; few studies have followed patients for multiple years to examine long-term outcomes. This study provides solid evidence that strongly encouraging patients to monitor their own health factors and explaining the associated health risks versus the physician simply monitoring them and dictating a change in therapy can result in a significant improvement in diabetic outcomes.

            Pharmacists are important members of the disease management delivery team. Rothman, Malone, Bryant, Shintani, et al. (2005) studied the impact of an intense disease management program on patients HbA1c levels in a randomized controlled trial. The intervention group received educational sessions with clinical pharmacists based on evidence-based algorithms independent of appointments with their primary care physicians every 2 to 4 weeks for the duration of the 12-month trial. The control group received one 1-hour session with the disease management team and received standard care from their physicians. 217 subjects were in this study; 105 in the control and 112 in the intervention groups. Both men and women were included in the study with an age range of 46 to 64 and all patients were accounted for at then conclusion of the study. African-American patients represented 59% of the control group and 70% on the intervention group (p>0.20). 79% of subjects in the control group and 80% in the intervention group had less than a high school education. Both groups had similar baselines characteristics. Measurements were taken at baseline, 6 months, and 12 months. By 12 months the intervention group had a significantly greater reduction in HbA1c level than the control group; 2.5 versus 1.6, respectively, a difference between the groups of 0.8 (95% CI: 0 to 1.7; p=0.02). Since the intervention group received their disease management education separately from standard care, these findings are most limited by the fact that the control group did not receive any kind of sham visits with the disease management team other than the initial one-hour session. This makes it impossible to attribute the intervention group’s improvement in glycemic control to the actual disease management program or just the attention they received. The generalizability of these findings should not be limited by the high percentage of African-Americans in the sample as discussed previously, though the very high percentage of undereducated may impact its generalizability.

            Mease et al. (2000) examined the role of delivering a disease management program via telemedicine technology on patients’ HbA1c levels. The intervention group had a telemedicine system installed in their homes, which allowed for testing of blood pressure as well as voice and visual communication with a disease management consultant to review glucose levels, blood pressure, and treatment objectives. Patients in the intervention group received weekly visits from a consultant and monthly visits from a physician via the telemedicine equipment throughout the 3-month study period. The control group received standard care from their physicians and was encouraged to seek disease management education. 28 subjects participated in this study; 13 in the control group and 15 in the intervention group. Both men and women were included in the study with an age range of 32 to 63 and all patients were accounted for at the conclusion of the study. No information was given on educational attainment and race. Patients were selected from an Army medical center. Both group had similar baseline characteristics at the commencement of the study. Measurements were taken at baseline and 3 months. The intervention group had a reduction in HbA1c of 1.3 of the three months while the control group had a reduction of 0.9 (p<0.05). No confidence intervals or standard deviations were reported. Though these findings support the use of telemedicine to implement disease management programs, there are several limitations to consider. First, the population coming from an Army hospital is one issue. The “military culture” of these patients may limit these findings’ generalizability to other populations. Of greater issue though is the level of intervention. In probably no other setting is it feasible or cost effective to install personal computers in individual homes and have physicians and diabetic consultants make personal one-on-one calls. Whether or not the findings are valid, they may never happen in any other setting.

            Another pharmacist-led intervention was conducted by Rothman, Malone, Bryant, Horlen, et al. (2003). This retrospective cohort design analyzed the impact of a pharmacist-led disease management program on patients’ HbA1c levels compared to their baseline levels prior to beginning the program. Clinical pharmacists provided a diabetes management program consisting of diabetes education, limited physical examinations (foot exams), initial treatment recommendations, algorithms for titrating insulin and metformin, and frequent patient follow-up in session up to one hour long. 138 non-insulin dependent diabetics participated in the study with ages ranging from 31 to 57 years, both men and women. 69% of participants were African-American and 56% had not completed high school. Baseline measurements were determined from patient records at time of enrollment in the diabetes management program and then again after 6 months in the program. One year after the diabetes program was instituted, the researchers retrospectively analyzed the impact of the program on HbA1c. No comparison group was used in the retrospective analysis. However, to demonstrate there was no regression to the mean, lab values before the baseline HbA1c were also analyzed. Blood pressure changes were calculated to see if HbA1c were from the program or from a general treatment effect (since blood pressure was not addressed in the management program). There was a significant reduction in HbA1c 6 months into the diabetes management program of 1.9 (95% CI: 1.5 to 2.3; p<0.0001). Further, there was no regression to the mean and no change in blood pressure. The findings of this study are seriously limited by the inherent threats to internal validity of a retrospective design. This design is susceptible to a history threat; there may have been some public health event during the same time as their disease management program that would also explain improved glycemic control. A maturation threat could also explain why patients’ HbA1c levels improved; over time learning to manage their diabetes better. The inclusion of a control group would have helped with both of these issues. There is also an issue with subject selection; selecting subjects retrospectively, as in this design, and not by random selection made it more likely that individuals were selected who were more motivated to successfully complete a disease management program in the first place since the subjects enrolled in the program on their own accord and not recruited for a research study. This design was very weak, but its findings are consistent with most other disease management research.

 

Implications and Unresolved Questions

Intensity

            The results of the moderate intensity and extreme intensity disease management programs show that more intense programs can have significant better outcomes as measured by HbA1c. The four studies outlined in the Intense Intervention section showed reductions in patients’ HbA1c levels that were consistently greater than what was seen in the three studies outlined in the Moderate Intensity section.

            Nonetheless, though outcomes are better with more intense disease management programs, some of these studies did not measure the patients’ likelihood of staying with the program. Consider the example of Mease et al. (2000), where sophisticated computer equipment was installed in patients’ homes. The study only evaluated the program for a 3-month period. After the fad of the new computer system wore off, the benefits of the disease management program may as well. This is possibly true of more intense programs in general, patients’ interest and participation may wean as time goes on. Therefore, reductions in HbA1c levels may be greater with more intense programs, but patients and providers need to balance this with what patients are more likely to stay with long term when choosing a disease management program. A smaller reduction may be better if the patient can maintain that reduction versus a greater reduction that eventually returns to their baseline level.

 Reductions in HbA1c

            An issue that is often not addressed in the literature is the lack of patients achieving goal HbA1c levels. The American Diabetes Association’s goal for NIDD is a HbA1c level of 7.0% or less (2004). Some guidelines are set even stricter. Nonetheless, not a single study in this review had patients reach this goal. It is hard to determine if this is indicative of inadequate disease management programs or the difficulty of treating NIDD.

            Every article in this review, both the moderate and intense intensities, had statistically significant reductions in HbA1c levels. In fact, the clinical impact is significant as well. The UKPDS Group demonstrated that even a relatively slight reduction in HbA1c of 11% has been correlated with a 16% reduction in heart attacks and a 25% reduction in microvascular complications (1998). To put these figures in perspective, an individual with a HbA1c of 10.0% would need a point reduction of 1.1 to receive these same microvascular and cardiovascular benefits.

The smallest point reduction in this review was 0.5 in Surwit et al. (2002), for patients with a baseline of 7.6%. Therefore, all of the patients in this review would have improved their microvascular and cardiovascular risk profile. Yet, they still did not reach the ADA goal. This is a large issue for the diabetes community. Drastic changes are needed in order to control patients’ management of their diabetes and reduce their complications. As discussed above, more intense disease management still does not provide enough reduction in HbA1c.

 

Combination of Programs

            A potential solution for getting patients to goal that has been loosely explored in the literature is comprehensive disease management that combines components that have been shown to be successful independently. This has only been explored in the literature to the extent of combinations offered in traditional disease management programs such as using patient education with treatment algorithms as study by Rothman, Malone, Bryant, Shintani, et al. (2005). What has not been explored is adding in non-traditional components with traditional ones.

Natarajan, Clyburn, and Brown (2002) demonstrated that exercise is correlated with lower HbA1c levels. Consider combining the clinical pharmacist program with an exercise regimen. Possibly patients working with clinical pharmacists and on a set exercise regimen would have greater reductions in HbA1c than with either program alone. Different combinations of disease management programs need to be explored further in the literature.

 

Conclusions

The articles summarized in this review demonstrate that participation in disease management programs can improve outcomes in male and female age 30 to 65 NIDD patients. Further, there are various specific components that can contribute to the improved outcomes. Nonetheless, none of these programs have improved diabetic outcomes to the standards established by national guidelines. It is unknown if combining various components of disease management studied in these articles would be successful in bringing patients to the established HbA1c goals and warrants further research.

 

References

American Diabetes Association. (2004). Standards of medical care in diabetes.  Diabetes Care. Retrieved April 17, 2005, from www.diabetes.org.

 

Beckman, J.A., Creager, M.A., & Libby, P. (2002). Diabetes and atherosclerosis – epidemiology, pathophysiology, and management. Journal of the American Medical Association, 287, 2570-2581.

 

Bootman, J.L. & Schulman, K.A. (2002). Disease Management. Roanoke, VA: The Certified Medical Representatives Institute, Inc.

 

Diabetes Research Working Group. (1999). Conquering Diabetes: A Strategic Plan for the 21st Century. Bethesda, MD: National Institutes of Health.

 

Dudl, J. & Christensen, K. (2005). Managing your diabetes. Your Health. Retrieved April 12, 2005, from www.kaiserpermenente.org.

 

Goldstein, B.J. & Wordell, C.W. (2002). Diabetes: Approaches in Disease Management. Roanoke, VA: The Certified Medical Representatives Institute, Inc.

 

Harris, M.I., Flegal, K.M., Cowie, C.C., Eberhardt, M.S., Goldstein, D.E., Little, R.R.,Wiedmeyer, H.M., & Byrd-Holt, D.D. (1998). Prevalence of diabetes, impaired fasting glucose, and impaired glucose tolerance in U.S. adults. Diabetes Care, 21, 518-524.

 

Khaodhiar, L., McCowen, K.C., & Blackburn, G.L. (1999). Obesity and its comorbid conditions. Clinical Cornerstones, 2, 12-31.

 

Litaker, D., Mion, L.C., Planavsky, L., Kippes, C., Mehta, N., & Frolkis, J. (2003). Physician-nurse practitioner teams in chronic disease management: the impact on costs, clinical effectiveness, and patients’ perception of care. Journal of Interprofessional Care, 12, 223 – 237.

 

Mease, A., Whitlock, W.L., Brown, A., Moore, K., Pavliscsak, H., Dingbaum, A., Lacefield, D., Buker, K., & Xenakis, S. (2000). Telemedicine improved diabetic management. Military Medicine, 165, 579-584.

 

Natarajan, S., Clyburn, E.B., & Brown, R.T. (2002). Association of exercise stages of change with glycemic control in individuals with type 2 diabetes. American Journal of Health Promotion, 17, 72-75.

 

Rachmani, R., Levi, Z., Slavachevski, I., Avin, M., & Ravid, M. (2002). Teaching patients to monitor their risk factors retards the progressions of vascular complications in high-risk patients with type 2 diabetes mellitus – a randomized prospective study. Diabetes Medicine, 19, 385-392.

 

Rothman, R., Malone, R., Bryant, B., Horlen, C., & Pignone, M. (2003). Pharmacist-led, primary care-based disease management improves hemoglobin A1c in high-risk patients with diabetes. American Journal of Medical Quality, 18, 51-58.

 

Rothman, R.L., Malone, R., Bryant, B., Shintani, A.K., Crigler, B., Dewalt, D.A., Dittus, R.S., Weinberger, M., & Pignone, M.P. (2005). A randomized trial of a primary care-base disease management program to improve cardiovascular risk factors and glycated hemoglobin levels in patients with diabetes. The American Journal of Medicin, 118, 276-284.

 

Saudek, C.D. (2000). The challenge of diabetes management. The American Journal of Managed Care, 6, S664-S667.

 

United Kingdom Prospective Diabetes Study Group (1998). Intensive blood-glucose control with sulphonyureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet, 352, 837-853.

 

Sursit, R.S., van Tilbur, M.A.L., Zucker, N., McCaskill, C.C., Parekh, P., Feinglos, M.N., Edwards, C.L., Williams, P., & Lane, J.D. (2002). Stress management improves long-term glycemic control in type 2 diabetes. Diabetes Care, 25, 30-34.

 

Wild, S., Roglic, G., Green, A., Sicree, R., & King, H. (2004). Global prevalence of diabetes: estimates for the year 200 and projections for 2030. Diabetes Care, 27, 47-53.