DESIGN: Retrospective chart review

SETTING: A Veterans Affairs Health Care System where patient care was provided either by traditional medical management or by a pharmacist or nurse practitioner in a diabetes disease management clinic.

PATIENTS: 81 veterans with type 2 diabetes mellitus managed with oral antidiabetic agents alone. These patients were divided into two groups: 1) those who were initiated on self-monitoring of blood glucose (SMBG) by their primary care provider (control group); and 2) those who began receiving care in the DDMC when they initiated SMBG (treatment group).

MAIN OUTCOME MEASURES: Change in total glycosylated hemoglobin value from baseline values.

RESULTS: The median baseline total glycosylated hemoglobin values were similar between the two groups (11.5% vs.11.4%, p=0.550). Patients enrolled in the DDMC achieved a significantly greater reduction in glycosylated hemoglobin compared to those patients not enrolled in the clinic (-1.6% vs. -0.4%, p=0.047). Patients who were not enrolled in DDMC also failed to achieve a significant reduction in glycosylated hemoglobin from baseline (p=0.231), whereas the follow-up glycosylated hemoglobin values in patients enrolled in the DDMC were significantly lower than baseline (p<0.001).

CONCLUSION: Participation in the DDMC appeared to be associated with improved glycemic control. Furthermore, patients who received SMBG alone did not benefit; only patients who were enrolled in clinic in addition to the initiation of SMBG achieved a reduction in total glycosylated hemoglobin.

KEYWORDS: Diabetes mellitus, Blood glucose, Self monitoring

J Managed Care Pharm 1999: 511-515

One tool used to help patients achieve improved glycemic control is the diabetes disease management clinic (DDMC). Current literature has shown that patients who attend diabetes clinics have improved glycemic control,^6-8 as well as improved blood pressure control, fewer diabetic complications, and better weight control.⁶ However, because these studies predominately involved patients with type 1 diabetes, the impact of a diabetes clinic on patients with type 2 diabetes mellitus is less well established.

Pharmacists have been increasingly involved in the management of patients with diabetes since the early 1970s.⁹ In a large survey of American acute-care hospitals, diabetes clinics were the most common clinics to involve pharmacists in providing care.10 Despite these survey results, few published studies exist that evaluate the effectiveness of pharmacist-managed diabetes clinics.^11-14 The few reports available were published in the 1970s and 1980s, and describe programs in detail with regard to structure and the type of services provided. Some examine the effectiveness of the programs by measuring outcomes such as patient compliance, subjective and objective complaints of diabetes, and fasting blood sugar. The knowledge of diabetes and the role of pharmacists in the clinics have been evolving over the last 20 years, and results of these studies may no longer be applicable. For example, none of these studies assess the impact of diabetes clinics on total glycosylated hemoglobin or diabetic complications, two measures more recently recognized as standard methods of assessing diabetes care. This retrospective study assesses the impact of a diabetes disease management clinic (DDMC) on the total glycosylated hemoglobin (GHb) of patients with type 2 diabetes mellitus.

METHODS

The DDMC provides care on three half-days per week and averages approximately 120 patient visits per month. Primary care providers refer patients diagnosed with diabetes mellitus to the clinic for comprehensive disease state management, which includes both education and diabetes medication management. Education is provided on disease state, medication usage, and the importance of diet and exercise. In addition, patients receive extensive training in the use of a blood glucose testing device. Under protocol, the clinic practitioners modify the patient's antidiabetic regimen following consultation with the patient's primary care provider. The DDMC practitioners also play a vital role in ensuring each patient receives the proper specialized follow-up care from other providers such as dietitians, podiatrists, and ophthalmologists.

Traditional medical management was delivered at the discretion of the patient's primary care provider. Patient education could involve nurses, nurse practitioners, or medical residents. No formal educational process or algorithm was used.

Inclusion Criteria
Patients included in this study were those diagnosed with type 2 diabetes mellitus and receiving treatment with oral anti-diabetic agent(s). These patients were divided into two groups: 1) those who were initiated on self-monitoring of blood glucose (SMBG) by their primary care provider (Group I); and 2) those who began receiving care in the DDMC when they initiated SMBG (Group II).

Inclusion also required either two total glycosylated hemoglobin (GHb) values -- a baseline value taken no more than 12 months before beginning SMBG +/- clinic enrollment, and a follow-up value taken between three and 18 months following this intervention -- and a documented initiation date of SMBG. In addition, patients receiving insulin were excluded, as were those with questionable medication refill histories, defined as pharmacy profile antidiabetic medication(s) refill requests which were less than 90% of the projected usage rate over time.

Patient Accession
From the Decentralized Hospital Computer Program (DHCP), four patient databases were obtained: 1) patients receiving oral antidiabetic agent(s); 2) patients receiving blood glucose strips; 3) patients receiving insulin; and 4) patients enrolled in the DDMC. The four databases were cross-referenced using a relational database program (Paradox 7.0, ©Borland Intl. Inc.) to obtain the two groups of study patients. Patients were randomly selected from the list of patients generated by this program.

Patient Demographics
General characteristics, social characteristics, evidence of diabetic complications, lipid panel, and total glycosylated hemoglobin values were determined through retrospective chart review, DHCP computer outpatient pharmacy profiles, and the DHCP computer lab database (see Table 1).

General characteristics examined included age, sex, ethnicity, weight, medication(s) dosage, and duration of disease. Due to the retrospective nature of this study and the limitations of available databases, more accurate assessment of obesity, such as body mass index or percent of ideal body weight, could not be determined.

Social characteristics, including alcohol use, tobacco use, diet, and exercise, were noted because evidence has demonstrated that patients who modify their life styles can dramatically improve glycemic control.¹⁵ All social demographic data was self-reported by patients and documented in medical records.

Evidence of diabetic complications was extracted from chart review or the computer problem list, or through lab data. Definitions were adapted from previous studies.^2,16 Nephropathy was defined as the presence of proteinuria or serum creatinine persistently greater or equal to 1.5 mg/dL. Peripheral neuropathy, retinopathy, and peripheral vascular disease (PVD) were documented in the medical record.

Lipid panel values were obtained because evidence exists to show that lipid control is associated with the level of glycemic control.¹⁷

Total glycosylated hemoglobin value range at this institution is 4%�8%. At the time of this study, the laboratory used total glycosylated hemoglobin as the measurement. The percent HgbA1c equals (%total GHB)(0.6%) + 2.10.

Statistical Analysis
Because the data was not normally distributed, results are presented as medians and nonparametric statistic tests were employed where applicable. Patient demographic data were analyzed using the Mann-Whitney Rank-Sum test for interval data and Chi Square test with Yates correction for nominal data comparisons. The Wilcoxin Signed-Rank test was used to compare baseline versus follow-up total glycosylated hemoglobin values for each group. To compare the change in total glycosylated hemoglobin values between the two groups, the Mann-Whitney Rank-Sum test was used. Statistical tests were performed using Sigma Stat (Version 2, ©Jandel, Sausalito, CA).

RESULTS

Patient Demographics
Nearly all patients enrolled were male, consistent with the Veterans Affairs Health Care System patient population. Base-line patient demographics were similar in both groups, except that the control patients were slightly younger (median age 66 vs. 68, p<0.005), more likely to use tobacco (42% vs. 16%, p<0.005), and more likely to have evidence of nephropathy (50% vs. 26%, p<0.005).

Total glycosylated hemoglobin
The median baseline total glycosylated hemoglobin values were similar between the two groups, 11.5% and 11.4% for group I and group II, respectively (p=0.550). On follow-up, a significant difference existed in the median glycosylated hemoglobin reduction between the two groups (-0.4% in group I vs. -1.6% in group II, p=0.047) (see Figure 1). In group I, the median follow-up glycosylated hemoglobin did not significantly differ from baseline (p=0.231). In contrast, the median follow-up glycosylated hemoglobin in group II significantly decreased from baseline (p<0.001). Thus, only patients who were enrolled in the DDMC clinic in addition to the initiation of SMBG achieved a reduction in total glycosylated hemoglobin.

DISCUSSION

The reduction of glycosylated hemoglobin in DDMC patients was not only found to be statistically significant, it is also of clinical relevance. In the DDMC patients, the median baseline glycosylated hemoglobin was 11.4%, which on follow up declined 1.6% to a value of 9.8%. The Wisconsin epidemiologic study results suggest that a glycosylated hemoglobin change of this magnitude would predict for a decrease in the incidence of proliferative diabetic retinopathy from approximately 40% to 10%.⁴ Unfortunately, the retrospective nature of this study did not allow for the assessment of progression of retinopathy in patients.

The retrospective nature of this study also did not allow us to determine precisely why DDMC patients had improved control in total glycosylated hemoglobin. Neither patient selection nor referral bias appear to be responsible for this result, because the baseline GHb was strikingly similar in both groups. Several possible explanations exist and should be examined in prospective evaluation.

The education that DDMC patients received may be partially responsible for those patients achieving greater reduction in total glycosylated hemoglobin. Because our clinic is a specialized disease state management clinic, patients underwent comprehensive education on the various aspect of diabetes, such as disease states, medication usage, behavior modification (e.g. diet, exercise), self-care skills, and blood glucose device training. Previous studies suggest that systematic intensive education itself can have a positive effect on indicators of glycemic control, including fasting blood glucose and glycosylated hemoglobin, in patients with diabetes mellitus.^18-19 Un-fortunately, due to the retrospective nature of this study, we were unable to accurately document actual changes in patient behavior (e.g. increased compliance with diet, exercise, or blood glucose device use). Another possible explanation for the greater reduction in the total glycosylated hemoglobin in DDMC patients was the close follow-up these patients received. Depending on the level of glycemic control, some patients were seen as frequently as once every one to two weeks, with telephone follow-up for those patients who required closer monitoring. In addition, clinic staff also consulted with dietary, podiatry, and ophthalmology services to ensure patients received specialized care when necessary. This type of close follow-up could allow clinic staff to detect poor glycemic control early and take appropriate action.

Successful management of diabetes involves numerous psychological factors that determine whether patients will comply with treatment. Patient perception of symptoms, fear, unawareness of symptoms, attitudes, and control issues are all important factors.20 A recent study on the psychological needs of patients with diabetes reports that 23% of newly diagnosed diabetic patients desire more emotional support.21 The psychological impact of the DDMC on patients may also have contributed to improved glycemic control in clinic patients. In this DDMC, patients and staff members work as a team to manage the illness. Patients are encouraged to ask questions or express concerns regarding their disease. Many patients have developed trust in the DDMC staff, which may have helped to convince these patients that the education they receive at the DDMC is highly relevant to their well-being.

The secondary finding in this study demonstrated that the addition of SMBG to an oral antidiabetic regimen did not improve glycemic control in patients receiving traditional medical care. Although the American Diabetes Association recommends the use of SMBG and as many as 60% of the general practitioners routinely recommend SMBG in patients with type 2 diabetes mellitus, the benefits of SMBG in these patients remain unclear.^{16, 22-25} This study's findings correlated well with results of previous studies that showed that the addition of SMBG does not affect glycemic control in patients with type 2 diabetes.^16,24 However, this study was not designed to measure the usefulness of SMBG, and prospective controlled studies will be helpful in defining the role of SMBG in these patients.

SUMMARY

  References