- Click here for explanation of classification scheme.

Primary objective:

1. to determine whether a low glycemic index diet (LGI) was a feasible dietary intervention for general pediatricians to employ in the treatment of overweight children in their practices, in which there are often constraints of limited time and lack of specialized personnel such as registered dietitians

Secondary objectives:

1. to determine whether parents of overweight children could implement the LGI in order to lower the glycemic index (GI) of their child's diet after receiving brief instructions and a handout from their primary care pediatrician
2. to determine parental satisfaction with the use of the LGI diet and the impact of the LGI on their child's body mass index (BMI)

Hypothesis tested:

Following a brief description (5 minutes or less) of LGI diet provided by a primary care pediatrician, combined with a descriptive handout, parents of overweight children would be able to lower the GI of their child's diet.

Participants were recruited from patients aged 5-12 years who had visited one of five primary care practices (members of the Utah Pediatric Research Network), usually for well-child care.

Children who had a BMI greater than the 95th percentile for age and gender (using CDC charts) were identified by their pediatricians. After brief description of study to parents of eligible overweight children, interested parents signed informed consent forms.

Children were excluded from study if:

1. they had medical conditions known to be associated with being overweight such as Prader-Willi syndrome, or
2. they were on medications known to be associated with weight gain such as corticosteroid therapy,

Blinding used (if applicable)--N/A

Intervention (if applicable)

After brief description of LGI diet, parents of the overweight children were provided with a dietary handout titled the "stoplight diet" categorizing foods by their GI (differs from the "Traffic Light" diet employed by Epstein et al. that classifies foods according to their fat content as well as simple carbohydrate content with "red light" foods being those with the highest fat content). The LGI dietary approach emphasized the GI content of foods rather than the calorie content of foods. 

For example, foods with a high GI such as potatoes, soda, cereals with added sugar, and candy were identified as "red light" foods to be eaten once a week or less. "Yellow light" foods such as pasta, peanut  butter, and whole fruit juices could be eaten 2-3 times per week. Foods with low GI, "green light" foods, included most fruits and vegetables, meats, dairy products, and legumes and could be eaten every day. A GI score was derived from standard GI tables. Two weeks of sample menus were also provided.

At entry, parents completed a 24-hour recall and a food frequency questionnaire (FFQ).

At 3, 6, and 9 weeks parents completed a mailed 3-day food record (2 weekdays and 1 weekend day).

At 12 weeks, completed a second FFQ.

Other: 

At entry, parents of the overweight children completed a questionnaire that included demographic and medical family information regarding the presence of overweight, type 2 diabetes, hypertension, hyperlipidemia, heart disease, or sleep apnea. In addition, parents were asked: "Do you think your child's weight is causing him/her any health or psychological problems?"

Statistical Analysis -- Food-hour recall and 3-day food records were analyzed using the Nutritionist IV nutrient analysis software. The GI on the initial and final FFQs were compared using a scoring system based on the frequency of red, yellow, and green light foods.

BMI Z-scores were calculated using the Epi Info programs (CDC).

Between group comparisons were performed using t-tests for continuous variables and chi-square for categorical ones. Between subject comparisons of measurements at baseline and completion of study were performed using the Wilcoxon signed rank test.

P<0.05 was established as significant.

Measures: 

• BMI (height and weight measured at entry and at 12 weeks);
• 24-hour recall (at entry),
• food frequency questionnaire (at entry and at 12 weeks ),
• 3 day food records (at 3, 6, and 9 weeks),
• demographic and family medical history questionnaire,
• end-of-study questionnaire about the child's and family's experiences with the diet. 

A GI score was determined using standard GI tables.

Dependent Variables

1. BMI z-score--the difference between the subject's BMI and the 50th percentile for their age and gender and divided by the standard deviation; this allowed comparison between the children of different ages and genders
2. excess BMI score--difference between the subjects' BMI and the 95th percentile for their age and gender
3. change in BMI (weight loss)
4. dietary--change in mean FFQ GI score, mean daily carbohydrate intake and mean daily caloric intake over 12-week study period

Independent Variables

1. brief instructions and dietary education handout from primary pediatricians; parent's understanding of LGI diet and "red," "yellow," and "green" light foods; ability to alter child's diet to LGI diet; willingness of child to follow diet 
2. LGI diet intervention

Control Variables

• age
• gender
• initial degree of BMI (measured by BMI z-score)

Initial N: 34 met enrollment criteria; 10 were nonparticipants.

Attrition (final N): There were 9 dropouts (completed initial data but incomplete follow-up data). 15 completed the full 12-week study.

Age: mean age 9.6 years of age, 53% male.

Ethnicity: not provided

SES: not specified

Location: Utah

The primary end point of the study was a comparison of the FFQ GI scores at the beginning and end of the study as a reflection of whether the parents were able to change the GI of their overweight child's diet.

Results are expressed as mean ± standard deviation (SD)

Dietary

14 of the 15 children who completed the 12-week study lowered their FFQ GI score (mean initial score = 26.7 ± 5.7; mean score at 12 weeks = 15.6 ± 5.2; p<0.0001).

Mean daily carbohydrate intake decreased by 73 g over the 12-week study period (initial: 256.1 ± 52 g; 12-week: 183.2 ± 30.2 g; p<0.02).

Mean calorie intake decreased by 292 kcal over the 12-week study period (initial:1919.2 ± 229 kcal; 12-week: 1627.2 ± 284 kcal).

BMI

The overall mean BMI for the group did change (initial: 25.3 ± 2.9; 12-week: 24.8 ± 2.8); however, the change was not statistically significant (p=0.07). BMI decreased in 12 of the 15 children (8 children lost weight; 4 childrens' BMI decreased because of linear growth during 12-week period). The mean BMI z-score (which adjusts for age and gender) and mean excess BMI measure did change significantly (initial BMI z-score: 2.18 ± 0.4; 12-week: 2.05 ± 0.3, p<0.02, initial excess BMI score: 3.22 ± 1.9; 12-week: 2.25 ± 1.85, p<0.01).

Parental Response

Of the 15 parents, all reported that it was "very easy" or "fairly easy" to understand the stoplight concept of which foods were "red," "yellow," and "green" (high, medium, and low GI) light foods; 10 reported that their child had been "successful" in following the LGI diet; 14 said they were "very" or "fairly" likely to continue the LGI diet; and 12 of their children were "very willing" or "fairly willing" to follow the LGI diet.

The results of this pilot study indicate that after hearing a brief discussion of the LGI diet and receiving a handout, many parents were able to lower the GI of their overweight child's diet.

Of the parents who completed the study, all reported that the stoplight concept was easy to understand; 14 lowered the GI of the child's diet (decrease in mean FFQ GI score); and 14 of 15 stated that they were very or fairly likely to continue the diet.

The authors indicated that the BMI of 12 of the 15 children decreased over the 12-week study period; 8 children lost weight, and in the others, BMI decreased because linear growth exceeded weight gain over the course of the study.

"We believe our findings suggest that it is feasible for general pediatricians to implement the LGI diet within the constraints of a busy practice. Further study on a larger scale with a control (comparison) group is needed to determine its efficacy as an accepted component of a primary care-based treatment program for overweight children and, possibly, their families." 

Not-for-profit

0

Limitations:

1. High attrition rate (56%; 19 invited and expressed willingness to participate did not completed study; only 15 of 34 eligible overweight children and parents completed study). In addition, the authors did not study families who dropped out or did not participate. Those who completed the study may have been more motivated and may not be representative of parents of many overweight children.
2. Overweight children were not randomly invited to participate. It is possible that some of the pediatricians chose patients who they thought had a good chance of complying; thus possible selection bias.
3. There was no control or comparison group. Although 12 of 15 children had a decrease in BMI over the 12-week study period, without a control group one cannot be sure that the BMIs would not have decreased with no treatment or if they had been given a different intervention.
4. Did not control for other possible factors that may have attributed to changes in BMI such as physical activity.
5. 12-weeks may not have been long enough to assess long term treatment with LGI diet.
6. Limited generalizability; study children were aged 5-12 years and parents were relatively well-educated (mother's highest grade: 15.5 years and father's highest grade: 16.3 years).