Diabetes is characterized by insulin insufficiency, lack of insulin production, and/or resistance to insulin. Lack of glycemic control leads to hyperglycemia and is associated with a variety of serious complications, including retinopathy, nephropathy, neuropathy, and cardiovascular disease.

In 1992 alone, $45 billion were spent on the management of diabetes and treatment of its complications. There was an estimated additional cost of $47 billion for premature death, disability, and work loss, resulting in a total yearly cost of diabetes of $92 billion.¹

Many studies have been conducted regarding the effectiveness of diabetes education.^2-7 Although the methods used and outcomes observed have varied, these studies have clearly shown a beneficial effect of education on diabetes control and reduction of complications.

Of all the aspects of a diabetic regimen, patients have been least likely to comply with diet modifications. Travis⁸ reported that personal motivation, social support, continuity of care, and understanding of meal plans positively affect diabetes self-management. Emotional factors, a busy schedule, and holidays tend to negatively affect patients' self-care. Patients were more likely to follow meal plans when the importance of diet in glycemic control was stressed. Follow-up visits with a registered dietitian also increased compliance. There is, however, little documentation of the beneficial effects on diabetes control of medical nutrition therapy (MNT) given specifically by registered dietitians.

The purpose of this study was to determine the contribution of diabetes MNT and diabetes self-management education (DSME) by a dietitian in lowering hemoglobin A_1c (HbA_1c) values in individuals with type 1 or type 2 diabetes. A retrospective chart review was performed on patient medical records at three clinics that either had an American Diabetes Association-recognized education program or were in the process of applying for recognition. This study also conducted a correlation between patient self-perception of diabetes goal achievement and actual control as assessed by HbA_1c values.

The MNT provided to the participants was patterned after the American Dietetic Association/ Morrison Health Care protocols for type 1 and type 2 diabetes.⁹ Patients had a minimum of two visits with a dietitian, with the visits typically scheduled 2 weeks apart. Medical records were reviewed only for patients who had had diabetes for longer than 6 months (mean duration of diabetes 8.1 months) before the initial HbA_1c measurements were taken for this study. The initial HbA_1c was the measurement taken near the first visit with the dietitian, and the post-HbA_1c level was the measurement taken approximately 3 months after the initial contact with the dietitian. (mean time between initial and post-dietitian HbA_1c was 3.34 ± 1.1 months in 74% of the charts reviewed.)

Goal-setting to improve diabetes control or DSME was incorporated into the MNT provided. These goals were specific (e .g., exercise 3 times per week, consume 60 g carbohydrate per meal) and established by the patients. Patients rated their level of adherence to their goals on a 10-point Likert scale where one meant that the goal was not met at all and 10 indicated a high level of goal achievement.

Patients also rated their level of understanding of 13 topics that were included in the MNT sessions (sick-day rules, benefits of exercise, and so forth). The purpose of rating the understanding of diabetes topics was to help the educator direct future sessions and to determine whether any association existed between self-perceived mastery of a skill and diabetes control.

The procedure for this study was as follows:

1. Institutional approval was obtained.
   
2. A data collection tool was developed for ease of data analysis.
   
3. Data were retrieved from chart notes or telephone contacts for patients' self-rating of their knowledge of diabetes (specifically on 13 questions using a Likert scale from good to poor understanding) before and after nutritional education with the dietitian.
   
4. Data were retrieved from chart notes or telephone contact for patient goal achievement.
   
5. Charts were reviewed for pre- and post-education HbA_1c values, demographics, goals, anthropometric measurements, and level of diabetes understanding.

Improvement in HbA_1c was significant both for patients with type 1 diabetes (mean pre-education HbA_1c of 9.24% ± 1.75 to mean post-education HbA_1c of 7.97% ± 1.29 P < 0.005) and for those with type 2 diabetes (mean pre-education HbA_1c 9.35% ± 2.12 to mean post-education HbA_1c 7.70% ± 1.53, P < 0.000). Average time between pre- and post-education HbA_1c was 3.34 ± 1.1 months in 75 patients that made follow-up appointments close to the recommended time frame, and 4.18 ± 3.23 months when calculated for all 83 patients that reported a return visit. (Six patients had a first return visit 12­18 months after the initial visit.) Changes in HbA_1c did not differ based on gender, age, or perceived level of understanding.

Thirty-two patients set goals to improve diabetes control, of which 54% set goals related to diet, 55% set goals related to exercise, 42% set goals related to blood glucose monitoring, and 14% set goals related to weight loss. The percentage of men and women setting each type goal was not significantly different.

Perception of understanding of diabetes (as determined by pre- and post-education test scores), improved after education sessions by a mean of 11.14 points (t = 7.6, P = 0.000) (pre-education rank score of understanding = 21.5% ± 7.9, post-education rank score of understanding = 32.5 ± 4.5). There was no correlation between perceived control (assessed by subjects rating themselves on goal accomplishments) and actual control (r = 0.029, P = 0.85).

Weight remained stable in this population. Chart review indicated that 6 patients lost weight, 8 gained weight, and 88 had no change in weight. Only three patients reported a hospital visit in the past 5 years (two had two hospital visits each, and one had four hospital visits.)

Discussion and Conclusion
The cost for treating diabetes is high. Lengths of stay in the hospital per admission have decreased, and education related to special diets for patients is limited during hospitalization. This leads to an increased need for outpatient MNT.^10,11 Special diet education provided on an outpatient basis could prevent or postpone the expensive complications of this disease.^10,11 However, it has been estimated that only 58.6% of patients with type 1 diabetes, 48.9% of patients with insulin-treated type 2 diabetes, and 23.7% of patients with noninsulin-treated type 2 diabetes attended a class of program about diabetes at some time during the course of their disease.²

The findings of this study support previous studies on improving HbA_1c levels with diabetes education.^12-14 In addition, they support the contention that registered dietitians contribute to improvement in diabetes control as measured by HbA_1c. This study demonstrated a highly significant improvement in HbA_1c levels after MNT and DSME provided by a dietitian. The improvement in HbA_1c levels occurred regardless of gender, age, type of diabetes, level of education, or BMI.

It should be acknowledged that patients were receiving ongoing medical care from their physician in addition to nutrition education from the dietitians, and therefore improvements cannot be attributed solely to MNT. Patients had to have had diabetes for longer than 6 months to participate in this study because of the decrease in HbA_1c level that occurs when starting medications. The authors recognize that the improvement in HbA_1c could be multifactorial; however, excluding new-onset patients with diabetes and beginning the analysis at the first dietitian visit is a first step in collecting effectiveness data for MNT.

It should also be acknowledged that the study population was largely Caucasian. Therefore, these interventions need to be evaluated with more ethnically diverse populations.

The effect of nutrition education taught by a dietitians on HbA_1c (the accepted measurement of diabetes control) must be continually evaluated and reported to hospital administrators or clinic managers to reinforce the cost-effectiveness of dietitians. Future studies modeling this investigation with a prospective design could demonstrate significant cost savings secondary to nutrition education. We recommend that smaller clinics cooperate with each other and plan their data collection. A larger sample size and standardized care/goal forms and data collection sheets among facilities would provide more compelling data regarding cost-effectiveness of dietitians.