Analyzing The Impact Of Community Intervention And Promotora On Childhood Obesity Prevention Among Latino Children

Overview of a Research Article on Latino Childhood Obesity Prevention Study

The purpose of this article is to describe the importance of healthy eating and physical activity in order to control the obesity of Latino children. The issue of obesity of school children is really alarming and appropriate preventive measures should be taken as early as possible. This paper broadly explains the analysis for the prevention of childhood obesity proceeds with explaining the impact of the interventions conducted by the Community Health Advisor (also known as promotora) for controlling and preventing the childhood overweight and obesity (Rossen, 2014). These interventions are performed by randomizing some elementary schools in Latin America. This type of intervention is known as promotora based behavioral intervention. According to a study, about 25% of Mexican children of age group 6 to 11 years are suffered by the problem of obesity where 19% of non-Hispanic Blacks and 19% of non-Hispanic Whites also suffer from obesity and overweight issues. The proposed study proceeds through the application of statistical analysis that is, factorial designs to evaluate the effect of the single-level and multi-level interventions for modifying the ill practices causing obesity. The statistical measure z-scores of the BMI (Basal Metabolic Rate) rates are calculated and then findings are analyzed to accomplish the purpose of the study.

The study aims to provide solutions to control and prevent overweight and obesity related issue of Latino children of 6-11 years of age in order to reduce the chance of life-threatening diseases in future. Childhood obesity persists till the early adulthood period and the obesity is also a cause of developing diabetes in the body of Latino children (Ndisang, Vannacci & Rastogi, 2014). Moreover, Latino children are more prone to the risk of diabetes than non-Hispanic Blacks and Whites (Hoelscher et al., 2015). Therefore, the given study comes with a social cause to provide solutions to prevent obesity. In addition to this, it will help to promote the healthy habits that claim to reduce the obesity of the Latino children.

The study provides the significance of promotora based intervention on multi-level using the factorial design experiment. For that purpose, it is required to collect the samples to conduct the statistical analysis and therefore the data were collected. The study has used primary data sources. Thus, the data were collected from thirteen schools selected randomly. The thirteen schools were randomly gone through the Aventuras para Niños (APN) study which is a 3-year, 2 × 2 factorial design randomized controlled community subgroup trial and the schools were divided into four sub-groups namely, Family (Fam-only), Community (Comm-only), Combination of Family and Community (Fam+Comm), and subgroup of no-treatment control (Ayeni et al., 2013). Initially, twenty-five schools were sorted out. Among the, five were not eligible and seven were not interested for the study. 808 Latino parents and their children participated in the study who were enrolled to the 2^nd grade of the schools. The quantity that has been taken into account, is the BMI rate of the parents and their children. The main targeted variable is the parenting variable that causes the change in intervention of the Fam-only subgroup. This statistical analysis utilizes the BMI z-scores (Flegaldi & Cole, 2013) to interpret the multi-level promotora based impact.  The schools had to satisfy certain eligibility criteria which are as follows:

1. Minimum 70% enrollment of the Latino children.
2. An attendance criteria has been defined.

• Teachers should not undergo any physical training program in the past four years.

Significance of Healthy Eating and Physical Activities in Childhood Obesity Prevention

The subgroups are discussed below-

1. Family subgroup:Eight advisors conducted 22 hours of training to the participants on parenting, obesity and healthy food intake. Eating lots of fruits, vegetables, and drinking lot of water will help to control the weight.
2. Community subgroup: A continuous training was provided to parents for 2 to 8 hours on the maintenance of physical structures and social structures.

• Control conditions: Instead of conducting any treatment measurement on the children, this subgroup was constrained to maintain the normal lifestyle along with following the rules.

The Fam-only subgroup condition had accounted for the home environment and the Comm-only subgroup had the condition to account for school and community based on the physical and social transformation. Firstly, the factorial design is used to measure the multi-level impact of the promotora by calculating the BMI (Kuzawa et al., 2014) z-scores and secondly, the participants were instructed to increase physical activity and nutritious food intake for controlling the over-weight issue.

The primary data for this study were collected at four different time points which are as follows:

1. M1 which is Baseline
2. M2 that indicates one-year post intervention
3. M3 that represents one-year follow-up
4. M4 which shows two-year follow-up

The parents were undergone to a self-administered survey. Moreover, the participants’ height and weight were also measured and analyzed.

1. Parent and child BMI: The height and weight of the parents and children were measured on a stadiometer and a digital scale using anthropometric process. The anthropometric measure was 97-99%. The BMI rate was measured in kg/m² The z-score and percentile were calculated for child BMI for age and gender on the basis of 2000 CDC growth charts.
2. Physical activity of children: The children’s physical activities were measured on a 5 point scale and the categorical outputs were recorded by putting values from 1 to 5 where 1 indicates “much less than others” and 5 indicates “much more than others”.  

• Sports participation of children: This has been recorded as categorical output by interrogating the parents about their child’s participation in sports.

1. Parental support for physical activity of children: The support was measured based on three responses. Number of days the parents provide encouragement, transportation, and parents’ participation with child in physical activity.
2. Dietary intake of children: The parents were asked about the frequency of the intake of 49 food item in a 10 point scale where 1 stands for never and 10 stands for five or more times per day.
3. Intake of fat: The assessment of the parent’s action to reduce fat intake was measured on a 30-item scale and αvalues were noted. The α= 0.73 for 19 items, α=0.76 for 11 items.

• Family meals: The parents were asked to name the family meal that the entire family eat together thrice or more in a week.
• Away-from-home eating: The dietary behavior was measured on a 5-point scale where 1 represents never and 5 indicates five to seven times per week. The responses for eating in relatives’ and/or neighbors’ homes or in restaurants are noted down and dichotomized to check the change in BMI for eating outside homes.

1. Demographic variables: Open and close ended questions (household income, marital status, parents an child’s age, educational status, employment status, and the generation status of child and parent) were asked to parents and the responses were recorded.

The optimal statistical analysis of the data is provided by factorial design. The approach used here is called intention-to-treat approach. The software tools used to perform the statistical analysis is SAS. The SAS command “Proc Mixed” (Mistler, 2013) has been used to examine the normal outcomes of mixed effects models and “Proc Glimmix” (Ene et al., 2015) command was used for analyzing non-normal outcomes. The link functions along with an exact error distribution were chosen for non-normal outcomes. Then, a logistic model (Hilbe, 2017) was built to calculate the binomial error (Reed, Jenouvrier & Visser, 2013) and the logit link. The counting outcomes like intake of number of fruits or snacks per day are modeled using either Poisson or Negative Binomial distribution checking the best fit for the particular outcomes. For M2, M3, and M4, the models are accounted with repeated measures. The adjustment is done for the baseline level. The models are adjusted for the responses on demographic variables. The intraclass correlation (ICC) coefficients (Landers, 2015) range from 0 to 0.019 only for one exception on the number of snacks intake per day having ICC value of 0.095.

The output of the 2 × 2 factorial design for both the Fam-only and Comm-only subgroups are “yes” and “no”. The model is built including the interaction time-by-Fam-by-Comm along with all the lower terms. The elimination is done for non-significant terms having p> 0.05.

The mediation analyses are done by hypothesizing the intervention for changing the mediator. Each dependent variable is examined for the intervention effect and then that effect is calculated for the parent mediator. Three regression models are fitted here for the useful parameters estimation. The mediated effect is the product of the unstandardized regression coefficient of the second regression model of the intervention effect.

Methodology of Latino Childhood Obesity Prevention Study

The sample size for the two-year follow up (M4) is 441 at year three in spite of the inclusion of all the participants contributing either for M2 or M4. The effect sizes are calculated based on the BMI data of Latino children of the age group 5 to 8 years. The main effect of Fam-only intervention is estimated to 1.33 k/m² and that for the Comm-only intervention is 0.66 k/m². The estimated standard deviation is 2.02 k/m² and the standardized effect size for Comm-only intervention main effect is 0.33.

The ICC for the clustering (Jacques & Preda, 2014) of schools are calculated within the range from 0 to 0.016 which are actually the BMI z-scores. At 5% level of significance for the two sided test, the power of detecting the main effect of the Comm-only intervention is 80% and that for the Fam-only intervention is 97%.

The study is developed on the basis of the examination of the intervention of the family and the community environments. The hypothesis of test is that, the combined intervention of family and community has more impact on the BMI z-score of children than single impact of family intervention and community intervention (Bates et al., 2014). The findings explains that significant changes are not present in any of the intervention subgroups. The probable explanations for the null hypothesis have a wide variability. The significant behavioral changes of the parents and children are measured as a degree product of the facts and the outcomes are measured based on the values of the height variable of children and parents. There has been heterogeneity in the BMI z-scores of the participants at M1 (Wang, Zhang & Fu, 2016). The community intervention is heterogeneous in nature and it contrasts with the clinical intervention that target the risk levels. The BMI z-scores range from 41% to 52% within the four sub-groups. The primary analyses are performed in the light of data collected from M1 to M4. The overall retention rate is shown as 55% in the analyses and the condition based retention rates are 48% for Fam-only, for Comm-only it is 59%. In case of the combination Fam+Comm, the rate is 50% and lastly 59% for the control intervention.  

Conclusion

From the above study, it can be concluded that many essential improvisations have been achieved by interpreting the intervention effects which are obtained from the calculated BMI z-scores of the children, their percentiles and proportions of children belonging to the obese category. It has been found that BMI z-scores have increased always throughout the course of the analyses. Moreover, the proportion of obese children has also been increased. The effect of significant intervention of child BMI and parent BMI are negligible. Again, the interaction effect was absent in case of child gender. In addition to this, there has been no interaction effect for the intake of dietary food, child participation in any sports and many other factors that have been taken into account. The Famm-only intervention is responsible for the child’s environment. 

Statistical Analysis Used in Latino Childhood Obesity Prevention Study

Besides, promoting the healthy food intake through APN study and by organizing campaign of the name ‘Start with Salad’ is also an effective step to control the obesity. Increase of physical education classes, renovation of the schools’ park and playground for the enhancement of the physical activities are important outcomes in the course of this study. However, additional study is still required to assess the most efficient policy to boost healthy eating and physical activity enhancement. The study should primarily include the BMI z-scores and body fat measurements to curb the overweight problem and also to reduce the obesity rates.

References

Ayeni, A. O., Banerjee, S., Omoleye, J. A., Hymore, F. K., Giri, B. S., Deshmukh, S. C., … & Mudliar, S. N. (2013). Optimization of pretreatment conditions using full factorial design and enzymatic convertibility of shea tree sawdust. biomass and bioenergy, 48, 130-138.

Bates, D., Maechler, M., Bolker, B., & Walker, S. (2014). lme4: Linear mixed-effects models using Eigen and S4. R package version, 1(7), 1-23.

Ene, M., Leighton, E. A., Blue, G. L., & Bell, B. A. (2015, November). Multilevel models for categorical data using SAS® PROC GLIMMIX: the basics. In SAS Global Forum 2015 Proceedings (pp. 3430-2015).

Flegal, K. M., & Cole, T. J. (2013). Construction of LMS parameters for the Centers for Disease Control and Prevention 2000 growth charts (No. 63). US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics.

Hilbe, J. M. (2017). Logistic regression models. CRC press.

Hoelscher, D. M., Butte, N. F., Barlow, S., Vandewater, E. A., Sharma, S. V., Huang, T., … & Oluyomi, A. O. (2015). Incorporating primary and secondary prevention approaches to address childhood obesity prevention and treatment in a low-income, ethnically diverse population: study design and demographic data from the Texas Childhood Obesity Research Demonstration (TX CORD) study. Childhood obesity, 11(1), 71-91.

Jacques, J., & Preda, C. (2014). Model-based clustering for multivariate functional data. Computational Statistics & Data Analysis, 71, 92-106.

Kuzawa, C. W., Chugani, H. T., Grossman, L. I., Lipovich, L., Muzik, O., Hof, P. R., … & Lange, N. (2014). Metabolic costs and evolutionary implications of human brain development. Proceedings of the National Academy of Sciences, 111(36), 13010-13015.

Landers, R. (2015). Computing Intraclass Correlations (ICC) as Estimates of Interrater Reliability in SPSS, The Winnower 2: e143518. 81744, 2015, DOI: 10.15200/winn. 143518.81744 Landers This article is distributed under the terms of the Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and redistribution in any medium, provided that the original author and source are credited. Recently, a colleague of mine asked for some advice on how to compute interrater reliability for a coding task, and I discovered that there aren’t many resources online written in an easy-to-understand.

Mistler, S. A. (2013, April). A SAS macro for applying multiple imputation to multilevel data. In Proceedings of the SAS Global Forum (pp. 1-8). San Francisco, CA: Contributed Paper (Statistics and Data Analysis).

Ndisang, J. F., Vannacci, A., & Rastogi, S. (2014). Oxidative stress and inflammation in obesity, diabetes, hypertension, and related cardiometabolic complications. Oxidative medicine and cellular longevity, 2014.

Reed, T. E., Jenouvrier, S., & Visser, M. E. (2013). Phenological mismatch strongly affects individual fitness but not population demography in a woodland passerine. Journal of Animal Ecology, 82(1), 131-144.

Rossen, L. M. (2014). Neighbourhood economic deprivation explains racial/ethnic disparities in overweight and obesity among children and adolescents in the USA. J Epidemiol Community Health, 68(2), 123-129.

Wang, J. F., Zhang, T. L., & Fu, B. J. (2016). A measure of spatial stratified heterogeneity. Ecological Indicators, 67, 250-256.