- Click here for explanation of classification scheme.

1. To report a study of biocultural aspects of obesity in a sample of 6 to 12 year old middle-class Mexican children,
2. To identify general socio-economic features that may explain the particularly high rates of middle childhood obesity in a Mexican population characterized by relatively high SES,
3. To investigate how individual socioeconomic variation (such as differences in exercise, diet, family, household, parental, and demographic factors) help explain within-population variations in risk of obesity in middle childhood in Mexico, and
4. To consider how secular trends associated with such broad scale social change changes at a more proximate level.

• 6 to 12 year old children attending a single state elementary school in Jalapa (represents approximately half of the total study body).
• Obese: BMI at or above the 95^th percentile

Not stated

Design

Data sets were collected for each participating child. Interviews took place in a private room at school and were conducted by student teachers. Observational studies were made in school settings and in homes. Investigators observed foods being consumed during recess (n=1095 observations, in which 435 instances of food consumption were recorded). To make these observations, each child in the sample was located on 5 different occasions during recess, and whatever they were eating or drinking when first observed was recorded. Parent & child reports of number of soda drinks consumed in the preceding 7 days were also provided.

Statistical Analysis

T-tests (identify significant gender variations in mean anthropometric values), binary logistic regression (describing and testing hypothesis between obese and not obese groups and multiple continuous or categorical predictor variables).

Dependent Variables

• Body mass index

• Body fat (skinfold thickness and upper arm circumference)

Independent Variables

• Demographic variables including gender, birth weight, infant
• child feeding practices (age at weaning, soda consumption, sweet snack consumption),
• parental factors (mother’s and father’s age at birth, mother’s and father’s education level, mother working or at home),
• biobehavioral factors (hours of television viewing, computer use, reading, vigorous exercise),
• family and household variables (size, membership, birth order, SES).

Family functioning is not analyzed separately by gender and strictness is rated by parent’s score on a 0-10 point disciplinary attitudes scale

Health variables including parent, teacher and child’s self reporting of anxiety, depression and self esteem, parent reports of child suffering from allergies or asthma, absence from school, visits to doctor, and parent, teacher, and classmate reports of popularity among peers and happiness

Control Variables

Not specified.

Initial N: 219 (110 girls, 109 boys)

Attrition (final N): 219 (110 girls, 109 boys)

Age: 6 to 12 years

Ethnicity: Mexican

SES: Middle-class

Anthropometrics: 24.2% of all children in this sample were obese at baseline (29.4% boys and 19.1% girls).

Location: Jalapa, Mexico

Parents of these obese children classified them as having a weight problem only 58.9% of the time (63.6% boys, 52.9% girls). Just under 18% of all parents classified their children as having a weight problem.

Snacking frequency observations: Fruit is not described, but vegetables are called “vegetable snacks” and included jicama with chili powder, which was the most popular (n=11); “cookies and cakes” snacks (n=26); the most common candy items is lollipops (n=48 out of 92 candy observations); other snacks observed include bagged fried snacks are popcorn, Cheetos, and potato chips (n=51); savory snacks are high fat and include enchiladas, tamales, gorditas, hamburgers, hot dogs, and chicharrones (n=81).

Children spent an estimated average of 11.48 hour per week watching television, 2.76 hours on computers outside of school, 2.52 hours reading for pleasure, and 5.88 hours spent at vigorous exercise.

The majority of children were breastfed (84.6%). Mean age at weaning was 7.74 months.

Logistic regression model included three predictors explaining children’s classification as obese or non-obese:

1. gender
2. number of other children in the household
3. parental strictness in disciplinary strategy

The log odds of a child being obese is negatively related to more children in the household (p<0.05), negatively related to gender--i.e., boys more likely than girls to be obese (p<0.05), and negatively related to the degree of strictness in the parental disciplinary strategy (p<0.05). The odds of a boy being obese are 2.33 times greater than the odds of a girl being obese.

Obese children showed no differences in social, mental, or physical aspects of well-being than their non-obese peers.

The findings of this study allow some recognition of how large-scale secular trends in body size are mediated at the individual level. The finding that there are crucial microniche dynamics underlying gender differences in obesity risk indicated the importance of including a focus on gender as a socioeconomic construct in human biological studies of child growth, development and nutrition.

University/Hospital:

University of Georgia

Limitations:

Control variables not clearly indicated.

Rate of obesity is higher than other populations/regions. May not be generalizable to other regions, countries, or races.

Other Comments:

Cultural attitude: “chubby children are healthy children.”

Highest BMI was 31.45; only 1.5% had BMI > 30; no extreme degrees of obesity.