American Diabetes Association Guide to Nutrition Therapy for Diabetes. Marion J. Franz

been published in the literature but are not reported in the summary of evidence because of high dropout rates or incomplete data.

      Providing hospitalized patients with nutritionist visits and education can also be highly cost-effective for the health care system. In an evaluation of different types of educational visits for patients with diabetes (n = 18,404) at eight Philadelphia Health Care Centers, a total of 31,657 hospitalizations were recorded for 7,839 patients in the cohort. For patients who had at least one type of educational visit, the hospitalization rate was 34% lower than for patients who had no educational visit. Patients who had at least one visit with a nutritionist had hospitalization rates 45% lower than the rate of patients who had no educational visit. The average annual hospital charges for patients who received any educational visit were 39% less than the per-year average for patients who had no such visits (Robbins 2008).

      Randomized controlled trials and other outcome studies of MNT document mean decreases in A1C of ~1–2% (up to ~3% in newly diagnosed patients), depending on the type and duration of diabetes and at what time point outcomes are reported. The evidence suggests that MNT is most beneficial at initial diagnosis, but is effective at any time during the disease process, and that ongoing evaluation and intervention are essential. Outcomes resulting from nutrition interventions are generally known in 6 weeks to 3 months, and evaluation should be performed at these times. At 3 months, if no clinical improvement has been seen in metabolic outcomes (glucose, lipids, blood pressure), usually a change in medication(s) is needed. Type 2 diabetes is a progressive disease, and as b-cell function decreases, glucose-lowering medication(s), including insulin, must be combined with MNT to achieve target goals.

      Examples of Type 2 Diabetes Studies

      The U.K. Prospective Diabetes Study (UKPDS) was a randomized controlled trial that involved 3,044 newly diagnosed patients with type 2 diabetes at 15 centers. All treatment and control subjects received nutrition counseling, usually from a dietitian on study entry until 3 months, at which time they were randomized into intensive or conventional therapy. During the initial period when nutrition counseling was the primary intervention, the mean A1C decreased by 1.9% (from ~9 to ~7%), and there were average weight losses of 4.5 kg (UKPDS 7 1990; UKPDS 2000). UKPDS researchers concluded that, for improved glycemia, a reduction in energy intake was at least as important, if not more important, than the actual weight lost. At 2 years, the conventional group, whose primary therapy was diet, maintained an A1C of ~7%, and even at study end, the A1C was still slightly less than at diagnosis. However, because of the progressive deterioration of diabetes control, the majority of patients needed multiple therapies to attain glycemic target levels in the longer term.

      Also in the U.K. and in newly diagnosed individuals with type 2 diabetes (n = 593), the Early ACTID (Early Activity in Diabetes) trial compared usual care (initial dietitian consultation and follow-up every 6 months; control group) to an intensive nutrition intervention (dietitian consultation every 3 months with monthly nurse support) or to the latter plus a pedometer-based activity program (Andrews 2011). Baseline A1C levels were 6.7, 6.6, and 6.7%, respectively. At 6 months, A1C had not improved in the usual care group but had improved in the two intensive nutrition intervention groups (–0.3%). These differences persisted to 12 months despite the use of fewer diabetes drugs. Improvements were also seen in body weight and insulin resistance between the intervention and control groups. Of interest, adding the physical activity program created no additional benefit.

      In individuals with an average duration of diabetes of 4 years, intensive nutrition therapy provided by RDs resulted in a decrease in A1C of 0.9% (8.3 to 7.4%) and in subjects with a duration of diabetes <1 year of 1.9% (8.8 to 6.9%) (Franz 1995). By 6 weeks to 3 months, it was known if nutrition intervention had achieved target goals; if it had not, the RD notified the referral source that changes in medication were needed. A1C values were maintained to 6 months.

      Of interest is a randomized controlled trial of individuals with an average duration of diabetes of 9 years who had A1C levels >7% despite optimized drug therapy (Coppell 2010). The intervention group received intensive nutrition therapy resulting in a difference in A1C between the intervention and control groups at 6 months (–0.5%). This difference was highly significant, as were changes in anthropometric measurements, documenting the effectiveness of nutrition therapy even in diabetes of long duration. Furthermore, the reduction in A1C is comparable with that seen in clinical trials when a new drug, often a third, is added to conventional agents.

      In another smaller randomized control trial, obese subjects receiving intensive nutrition interventions experienced a decrease in A1C of 0.6% every 2 months for up to 15 months (Laitinen 1993). Also reported was a decrease in A1C of 0.5% in patients ≥65 years of age after 10 weekly sessions with an RD emphasizing goal-setting and using learning and social cognitive theory (Miller 2002). In a study of patients with type 2 diabetes in rural Costa Rica, a decrease in A1C value of 1.8% at 3 months was reported after nutrition and exercise interventions (Goldhaber-Fiebert 2003). Also, in a study of urban African Americans, decreases in A1C at 6 months of 1.9% were shown from interventions using healthy food choices and exchange-based meal plans (Ziemer 2003). In a randomized controlled trial conducted in Taiwan, decreases of 0.7% in A1C were reported in subjects after quarterly sessions with an RD for 1 year compared to a routine care control group (Huang 2010).

      A study that monitored outcomes illustrates the effectiveness of nutrition interventions in clinical practice (Lemon 2004). Data were collected from 221 patients with type 2 diabetes who were referred for nutrition education/counseling to 59 RDs working in 31 outpatient settings in the state of Wisconsin. To minimize selection bias, the RD recruited the first two patients meeting inclusion criteria each day, up to six per week. Data were collected at baseline, 3 months, and 6 months. RDs spent an average of 111 ± 55 min with each subject, they met with subjects an average of 2.1 ± 1.0 times, and 33 intervention topics were reported. Clinical outcomes (A1C, lipids, blood pressure, weight) improved significantly between baseline and 3 or 6 months, while stabilizing between 3 and 6 months. A1C decreased by 1.4% over 3 months and by 1.7% at 6 months (54% of subjects were newly diagnosed).

      Examples of Type 1 Diabetes Studies

      The Dose Adjusted for Normal Eating (DAFNE) trial was another study conducted in Great Britain to evaluate whether a 5-day course teaching how to adjust mealtime insulin based on planned carbohydrate intake can improve both glycemia and quality of life in individuals with type 1 diabetes (DAFNE Study Group 2002). In this study, individuals using routinely prescribed insulin therapy, in which the insulin regimen is determined first and eating must then be consistent and matched to the time actions of insulin, were either immediately provided the skills needed to determine mealtime bolus insulin doses based on desired carbohydrate intake on a meal-to-meal basis or they attended the training 6 months later. In the group receiving the DAFNE training, A1C levels were significantly improved by 1%, with no significant increase in severe hypoglycemia, along with positive effects on quality of life, satisfaction with treatment, and psychological well-being. These results occurred despite an increase in the number of insulin injections (but not in total amount of insulin) and an increase in blood glucose monitoring compared with the control subjects who received the training later.

      A follow-up of original trial participants at a mean of 44 months documented a mean improvement in A1C from baseline of 0.4%, remaining significant but less than the 12-month levels. Improvements in quality of life seen at 12 months were well maintained over ~4 years (Speight 2010). Of interest is another follow-up report examining changes in food and eating practices in DAFNE trial participants after changing to flexible, intensive insulin therapy. Concern had originally been expressed that individuals with type 1 diabetes, if given the freedom to adjust insulin doses based on carbohydrate intake, would overeat or make unhealthy food choices. These concerns were unfounded, since individuals using flexible, intensive insulin therapy did not engage in more excessive or unhealthy eating. Instead, many of the participants reported making few eating changes and, in some cases, actually reported being more rigid in their eating habits (Lawton 2011).

      A group in Germany reported a 1.5% lower A1C level 1 year after a 5-day intensive training course (after which the DAFNE trial was modeled) teaching participants how to match insulin doses to their food choices while keeping their blood glucose level close to normal. The course was taught by specially trained dietitians