- Click here for explanation of classification scheme.

To determine the long-term effect of low glycemic index dietary advice on metabolic control and quality of life in children with type 1 diabetes.

1. 8-13 yr of age

2. Diagnosis of type 1 diabetes mellitus for more than1 yr

3. Regular attendance in the diabetic clinic

4. No additional dietary restrictions

5. No other immediate family members with diabetes.

6. No medications that would affect appetite

7. Family able to read and write English

Excluded if not included above.

Recruitment:  Pediatric patients in Melbourne's Royal Children's Hospital

Study design:  Randomized Prospective Trial

Intervention: 

Randomization and stratification to study protocol based on usual dietary regime as determined by 3-d food records:

• GG: instructed in low-GI diet before and during the study;
• CG: randomized to low- GI diet but had followed a carbohydrate counting diet before the study;
• GC: randomized to carbohydrate counting study but had followed a low-GI diet before the study.
• CC: instructed on carbohydrate counting diet before and during the study.

Blinding:  Data analysis and clinical outcome measures were performed by researchers unaware of treatment allocation.

Statistical Analysis:

• sample size allowed for a 15% dropout rate and provided 80% power to reject the null hypothesis: that mean HbA1C levels after 12 months were the same in the two groups if a 10% difference was found.
• significance set at 5%
• intent-to-treat analysis performed
• results were analyzed using multiple linear regression or Spearman's correlation for continuous variables or with a cobiation of logistic regression or Pearson's chi-square analysis for categorical data
• nonnormal data were analyzed using Wilcoxon's rank-sum test.
• clinical data were adjusted for baseline values where specified

 

 

Timing of Measurements:  0, 3, 6, and 12 months

Dependent Variables

HbA1c

weight/height

dietary intake

incidence of hypoglycemia (<3.5 mmol/L or <65 mg/dL) or hyperglycemia (>15 mmol/L or 270 mg/dL) as determined by preprandial breakfast, dinner and supper

quality of life questionnaire completed independently by subject and parent.

Independent Variables

• different written materials given to subjects to reinforce counseling
• 3-day food diaries kept at 1, 3, 6, and 12 months
• phone calls made 2 weeks before clinic visit to ensure compliance

• Research RD instructed patient and parent at baseline with follow-up at each clinic visit.

• GI, relative to a standard glucose value of 100, was calculated by summing the following:  [grams of carbohydrate from the food item/total daily carbohydrate x 100 x GI of food item] of 284 carbohydrate-containing foods, 194 were assigned a known GI, but 90 were given "estimated" values based on the GI of similar foods

• height and weight
• physical activity

Control Variables

• baseline values
• insulin regimen, not different between groups
• reported physical activity, not significant between groups

Initial N:  104 subjects, 49 subjects were assigned to the carbohydrate counting (control) diet and 55 were assigned to low-GI diet.

Final N:   89; 11 dropped from carbohydrate exchange group and 4 dropped from GI group; (22% vs. 7%, P = 0.03).   The dropout rate was especially higher for the GC group who previously followed a low-GI diet but were in the carbohydrate-counting group during the study (39% dropout rate).

Age:  10-12 yrs

Ethnicity: not specified

Other relevant demographics: duration of diabetes: 3-4 years

Anthropometrics:  No differences in baseline measures or demographic data between study groups. 

Location:  Australia

 

HbA1c: 

Children in the low-GI group had significantly better HbA1c values than those in the carbohydrate counting group. (8.05 ± 0.95 vs. 8.61 ± 1.37%, P = 0.05).

Dietary Intake

Percent of kcal: Carbohydrate: 49%, fat: 34%, non-milk extrinsic sugars: 7%, and 21 g dietary fiber.

There appeared to be a high proportion of subjects who underreport food intake (53% vs. 46% in the carbohydrate counting vs. the low-GI group).  The dietary variables were reanalyzed, excluding underreporters but remained unchanged.

GI was not significantly different between the two groups, however at all time points the low GI study group had significantly greater proportion of subjects in the low GI range =, defined as a GI<55.

Incidence of Hyperglycemia

Rates of excessive hyperglycemic episodes (>15/month) were significantly lower in the low-GI group (35% vs. 66%, P=0.006)

Quality of Life Questionnaire

Both children and their parents reported that the low-GI diet was easier to follow, believed that the low-GI diet promoted better control of blood sugar and would choose to follow the low-GI diet over the carbohydrate counting diet.

There were no differences in insulin dose, hypoglycemic episodes, or dietary composition. Those in the low-GI diet group reported a better quality of life for both the child and parent.

Flexible dietary instruction based on the Food Guide Pyramid with an emphasis of low Glycemic Index foods improves HbA1C without increasing the risk for hypoglycemia and enhances the quality of life for children with diabetes.

Not-for-profit

0 Foundation associated with industry:

The dietary intervention was very similar for both study groups. The primary difference was that the low-GI groups were given information about foods with a low-GI index and were instructed to use at least 1 low-GI food at each meal, to incorporate low-GI foods in recipes and to eat a low-GI food when hungry between meals. Those counting carbohydrates were instructed to limit intake based on the carbohydrate content of each food.