The teacher-delivered Hip-Hop to Health Jr. Obesity Prevention Effectiveness Trial is based on the National Institutes of Health's efficacy trial, Hip-Hop to Health Jr. (20-22). That trial compared a teacher-delivered weight control intervention with a teacher-delivered general health intervention targeting children aged 3 to 5 years enrolled in 18 Head Start programs administered through the Chicago Public Schools. We present the baseline data collected in the fall of 2007 for the 16 schools (n = 539 students) that agreed to participate in the nutrition component of this trial. The participating schools varied in number of students and grade levels taught; however, because all Head Start programs were designed to be separate from the rest of the school, these variations affected the preschool atmosphere minimally. Children were excluded from the study if they had no medical clearance to participate in the program's physical activity component or if they required a special diet. Children were also excluded if dietary intake data (at preschool or outside of preschool; n = 23) were missing. Data were analyzed on 516 participants at these 16 schools. The University of Illinois at Chicago's institutional review board approved the protocol, and written informed consent was obtained from each participant's parent or legal guardian.

Student characteristics

Parents reported their child's date of birth, sex, and race. If the parent could not be interviewed, the child's teacher reported the date of birth. Sex and race were determined by observation. Research staff measured height by using a Seca 214 portable stadiometer (Seca Corp, Hanover, Maryland). Weight was measured by using a Tanita BWB-800 digital scale (Tanita Corporation of America, Inc, Arlington Heights, Illinois). Children removed their shoes and any heavy outer clothing for the anthropometric measurements. Both height and weight were measured twice to the nearest 0.1 cm and 0.1 kg, respectively. BMI z scores and BMI percentiles for age and sex were calculated for each child based on the 2000 Centers for Disease Control and Prevention (CDC) growth charts (23) by using CDC's program for SAS (SAS Institute, Inc, Cary, North Carolina) (24).

Measurement of nutrient content of served preschool meals

Research staff with undergraduate degrees in nutrition observed food items served at each meal for 3 to 5 days at all preschools. The nutrient data and preparation techniques for each food item served were available from the Chicago Public Schools to determine energy and macronutrients in the food. The mean daily energy and macronutrients served were calculated for each preschool.

Measurement of participants' dietary intake

Intake at preschool. Research staff unobtrusively documented the amount of food both served and consumed for all meals for 3 to 5 children per classroom each day for 3 to 8 days. The food and beverages consumed by each child were determined by calculating the difference between the amount of food served and the amount remaining on their tray at the end of each meal. All remaining food on the trays was measured with measuring spoons and liquid measuring cups as appropriate. The mean nutrient intake at each preschool was calculated from the children's average amount of energy and micronutrients consumed.

Intake outside of preschool. A food record recall method was used to obtain diet intake outside of school. The day before a child was to be observed, a packet was provided to the child's parent or guardian containing a food record form for recording all food consumed outside of preschool and a food portion visual guide to assist in estimating the amount of foods eaten. Research staff instructed the parents on completion of the food records and asked parents to record all food and drinks consumed by their child outside of school (ie, that afternoon and evening and the following morning before school). Each record was reviewed by the research staff for completeness and clarification by using the 5-step multiple-pass method (25) on the morning of the child's classroom observation. To maximize participation, parents who did not return food record forms before their child's scheduled observation were contacted by telephone and the 5-step multiple-pass method was used to determine the foods consumed by the child during the previous evening and before coming to school that day. The mean outside-of-preschool intake was calculated from the children's average energy and micronutrient intake at each school.

24-Hour intake. Total daily intake for each school was calculated by adding food intake inside and outside of preschool over a 24-hour interval; intake was assessed only on school days. Each child contributed one 24-hour intake. The mean energy intake for each preschool was compared with the daily recommended energy allowance (REA) based on the group's average age and assuming an active physical activity level (26). In this group of children, the estimated daily REA was 1,656 calories (90 kcal/kg). The dietary intake data were processed with Nutrition Data System for Research (University of Minnesota Nutrition Coordinating Center, Minneapolis, Minnesota).

Data analyses were performed with SAS version 9.1. The primary outcomes of interest were average energy and macronutrients 1) served and consumed at preschool, 2) consumed outside of preschool, and 3) consumed during the total 24-hour period, by preschool program. Dietary data were pooled and analyzed at the preschool level because energy and macronutrients served were observed by preschool, not individual children. Descriptive statistics including the mean, standard deviation, and range were generated for the 16 preschool programs. Because the variability in energy and macronutrient intake between preschools may be due to characteristics of the group of children, χ ² tests were used to assess differences between the preschool programs by sex. Spearman correlation coefficients were used to test for associations between BMI and BMI z score and other variables. Mean age and mean BMI z scores of the groups of children at each preschool were compared by using the Wilcoxon rank sum test. Mean nutrient differences between preschools for 1) nutrients served, 2) at-preschool intake, 3) outside-of-preschool intake, and 4) total 24-hour intake by preschool program were determined by using analysis of covariance with mean BMI z score and sex as covariates. This adjustment was made to control for differences in preschools because the ability to regulate intake has been associated with weight in preschool children (27). BMI and age as covariates were also explored; however, the data are reported with sex and mean BMI z score, because results did not differ. Tukey's honestly significant difference (HSD) test was used for pairwise comparisons of means at the preschool level. Mean energy served at each preschool, energy intake at preschool, energy intake outside of preschool, and total 24-hour energy intake at the 16 preschool programs were compared with the average daily REAs based on mean age and weight-specific estimated energy requirements for the group (26). Data are presented as preschool group mean and standard deviation with statistical significance set at P < .05 for differences between the preschools.

Approximately 11 eating events (breakfasts, lunches, and snacks) were observed for each school. The mean energy served and percentage of REA ranged from 799 to 1,494 kcal and 48% to 90% of the REA (Table 2). Three preschools served more than the recommended 67% of REA for energy, and 4 preschools served less than 55% of the REA. One preschool (preschool 1) served significantly more mean energy than 3 other preschool sites (for preschools 2 and 7, P = .02, and for preschool 8, P = .01, Tukey's HSD test for pairwise comparison). Preschool 1 also served meals consisting of significantly higher mean grams of carbohydrate than all other schools. More than half of the schools exceeded the recommended calories from fat (<30%) (26); the range was 25% to 40% of calories. There was no significant difference in the mean grams of protein and fiber served by each school, and all the schools served more mean protein and less mean fiber than recommended (26).

On average, 32 students were observed at each preschool. The children were not required to consume any meal; thus the number of observed children for each eating event varies and reflects the number of children who consumed each meal or snack (Table 3).

The energy consumed at preschool averaged 33% (range, 25%-39%) of the REA (1,1,656 kcal/d) (Table 4). The high amount of energy (mean [SD] = 1,494 [82] kcal) served at preschool 1 resulted in significantly higher mean energy consumption (39% of REA). Children at this school consumed significantly greater energy (mean [SD] = 647 [221] kcal) while at school compared with children at 2 other schools (P = .001 for preschool 11 and P = .01 for preschool 12), where a lower amount of energy was served, adjusting for BMI z score and sex. The mean percentage of calories from fat consumed at each school ranged from 21% to 40% of calories; children at 11 of the schools consumed more than 30% of calories from fat.

The mean outside-of-preschool intake was 58% (range, 53%-76%) of the REA. This did not differ by preschool, adjusting for BMI z score and sex. The percentage of total calories from fat consumed outside of the preschool was more than 30% (mean, 35%; range, 31%-42%) for all of the preschools.

Despite the highly variable mean energy intake at school, the mean 24-hour energy intake did not differ by preschool, adjusting for BMI z score and sex (Table 4). Daily energy consumed was 95% (range, 85%-105%) of the average REA. The percentage of calories consumed as fat ranged from 26% to 37%, and 14 of the 16 of preschools had more than 30% of total daily mean calories consumed from fat. The mean 24-hour protein intake exceeded recommendations (range, 53-71 g), and mean fiber intake (range, 9-12 g) was below the adequate intake recommendation of 19 g (26).