NEW UPDATE: Completed trials may now upload and register supplementary documents (e.g. null results reports, populated pre-analysis plans, or post-trial results reports) in the Post Trial section under Reports, Papers, & Other Materials.
Mitigating the Rise of Child Overweight in Urban Vietnam: Evidence from an Information Intervention
Initial registration date
January 14, 2019
January 15, 2019 2:20 AM EST
Erasmus University Rotterdam
Other Primary Investigator(s)
University of Economics, Ho Chi Minh City
Economic Research Service, U.S. Department of Agriculture
Erasmus University Rotterdam
Additional Trial Information
This impact evaluation examines the effectiveness of a nutrition and health information intervention for parents coupled with goal-setting and soft commitment. The aim of the intervention is to increase nutrition knowledge, promote healthy eating/exercise behavior and, ultimately, reduce overweight and obesity rates among schoolchildren (second to fourth grade, aged 7-10) in urban Vietnam, a prime example of a country in the health and nutrition transition that comes with rapid economic development. While stunting rates of children under 5 have fallen from 64% to 19% between 1984 and 2013 according to the World Health Organization, childhood obesity rates have increased rapidly over the last decade. According to the Center of Nutrition in Ho Chi Minh City (HCMC), in 2013, 38.5% of HCMC’s primary schoolchildren were either overweight or obese (Tuoi Tre, 2013).
High rates of obesity are driven by various socio-economic and environmental factors (Sahoo et al., 2015). Vietnam may be characterized as having a population with low levels of nutrition literacy while, at the same time, facing rapid changes in consumption and physical activity patterns. In particular, consumption of sugary drinks is high among schoolchildren in HCMC. The recent surge in obesity can have severe long-run repercussions. Overweight children are more likely to remain overweight as adults (Weiss et al. 2013). Obese adults are at increased risk of developing health conditions such as diabetes and heart disease. Taken together, childhood obesity may have important effects on an economy’s long-term productivity and economic development. In order to minimize future societal costs of obesity, prevention of childhood obesity should be a focus of policy actions, especially in countries where undernutrition is still a (costly) concern.
Nam, Pham et al. 2019. "Mitigating the Rise of Child Overweight in Urban Vietnam: Evidence from an Information Intervention." AEA RCT Registry. January 15.
A recent review of the obesity literature finds that multi-faceted policies are more likely than singular policies to be effective in reducing obesity rates (Cawley 2016). Following the literature, we adopt a multi-faceted treatment approach.
We test one multi-faceted treatment arm against one control arm, where assignment to the treatment arm is randomized at the individual level (children). In the control arm, we will give a pamphlet to parents about the importance of healthy eating behavior, an active lifestyle, and healthy body weight. The treatment package includes the following components:
(1) Healthy Eating and Exercise Conversation with a Trained Nutritionist by Telephone
This conversation centers around what a healthy weight for the child is, how to achieve it (e.g limiting sugary drinks and western-style foods), and why an unhealthy weight is bad (e.g. it raises the risk of adult obesity). The goal of the conversation is to provide the necessary information for parents to understand the importance of a healthy weight and to plan (if desired) effective actions to reduce their kid’s weight based on the nutritionist’s recommendations (e.g. encouraging more playtime after school, walking instead of taking the bus to school, reducing sugary drink consumption, eating out less at western-style restaurants, etc). The conversation is based on a script.
(2) Goal-Setting and Soft Commitment
At the end of the conversation, parents and the nutritionist agree upon a set of goals. Parents will be reminded of these goals after the call and will be asked to commit to them and to sign a document to that effect.
(3) Follow Up
SMS messages will be sent to all treated parents after the training to remind them about the nutrition and/or physical activity plans. The SMS will be personalized (name of child and specific goals). (4) Weight Scale This will help parents to monitor their kid’s weight and to see how much their actions are helping to reduce weight over time. (5) BMI-for-Age Report Card This gives information—based on body weight measured at school. In isolation, each of these is likely to have little to no effect or only short-term effects, but together, they will hopefully have a sizeable, long-lasting effect.
Intervention Start Date
Intervention End Date
Primary Outcomes (end points)
Expected impacts in brackets:
Primary outcome family 1 - Anthropometrics (measured) (-)
Primary outcome 2 - Healthy food choice (experimental measure) (+)
Primary outcome family 3 - Parental Nutrition Knowledge (+)
Primary outcome family 4 - Current Unhealthy Diet of Child (-)
Primary Outcomes (explanation)
Primary outcome family 1 details:
a) BMI-for-age z-score based on WHO reference (-)
b) Overweight: >1SD WHO reference (-)
c) Obesity: >2SD WHO reference (-)
d) Waist circumference (in cm) (-)
e) Measured sagittal abdominal diameter (in cm) (-)
f) Measured body fat (%) (-)
Standardized Index (a to f) (-)
Primary outcome 2 details:
We will perform a simple snack choice experiment (List and Samek 2015a) without providing an incentive or educational message. The outcome is a binary indicator based on choice between mango and a cookie. The procedure is as follows:
•A fruit (mango or apple) and cookie are placed on a tray. •Enumerator asks the child to either choose one fruit or one cookie that she or he wants to eat at this moment. •Enumerator tells the child that she or he should not keep the fruit or cookie for themselves or share it with other kids. The child should eat it before the session ends.
•Enumerator records the choice of the child. •Gives fruit or cookie to the child.
•Observes and records whether or not the food item is consumed.
Primary outcome family 3 details:
a) Agreement with: “Some children are born to be fat and some thin; there is not much you can do to change this.”?
[Coding somewhat and strongly disagree to 1, zero otherwise] (+)
b) Grams of vegetables needed each day for good health [log transformed] (+)
c) Grams of fruits needed each day for good health [log transformed] (+)
d) Calories needed per day to maintain current weight [log transformed] (-) [reverse coded for standardized index impact] Standardized Index (a to d) (+)
Primary outcome family 4 details:
a) In the past 30 days, number of meals in “western style” fast-food or pizza place (-)
b) In the past 30 days, ready to eat food meals from grocery store (-)
c) In the past 30 days, frozen meals/pizzas (-)
d) In the past 7 days, days consumed soft drinks at home (-)
e) In the past 7 days, days consumed fruit at home (+) [Reverse coded for index]
f) In the past 7 days, days consumed green vegetables at home (+) [Reverse coded for standardized index impact]
Standardized Index (a to f) (-)
Secondary Outcomes (end points)
Secondary outcome family 1 - Parental Nutrition Perception (~)
Secondary outcome family 2 - Self-reported Household Unhealthy Food Expenditures (-) Secondary outcome family 3 - Exercise over the last 7 days (+)
Secondary Outcomes (explanation)
Secondary outcome family 1 details:
a) Perceived healthiness of child’s overall diet? [Excellent and Very good coded as 1, zero otherwise] (~)
b) Perceived healthiness of current weight [1=Yes; 0=No] (~)
Standardized Index (a to b) (~)
Secondary outcome family 2 details:
Self-reported spending over the last 30 days, in logs (or hyperbolic sine transformation in the case of zeros):
a) Spending on sweet snacks (candy, chocolate, doughnuts, cookies, cake, and ice cream etc.) (-)
b) Spending on salty snacks (chips, pretzels, and crackers etc.) (-)
c) Spending on soft drinks (Coke, Pepsi, 7-Up, or Sprite etc.) (-)
Standardized Index (a to c) (-)
Secondary outcome family 3 details:
Over the last 7 days:
a) hours watching TV, playing video games, or using electronic devices (-) [reverse coded for standardized index impact] b) days physically active for a total of at least 20 minutes per day (+)
c) days walking or biking for at least 10 minutes (+) Standardized Index (a to c) (+)
The study uses a randomized control design at the child/parent level based on children in 7 schools (6 schools with treated and untreated children, and 1 pure control school). Target children (second to fourth grade, aged 7-10) are overweight or obese at baseline. Schools were identified at the outset for inclusion in the study from a set of schools that participated in an earlier child nutrition survey.
The statistical model will be a linear regression for all outcomes to estimate differences in means (intent-to-treat). We will present unconditional differences in means (model 1), as well as differences adjusting for baseline differences of a given outcome, as well as pre-determined child (age in months, gender), parental (survey respondent relationship to child, education, age, work status, height) and household (family size, home ownership, number of bedrooms) covariates (model 2). Robust standard errors will be presented. Attrition:
Balance of missing outcome data will be presented by treatment status. If balanced by treatment status and if attrition is small (<10%), no adjustments will be made. If missingness is systematic and substantial, we will use bounds (such as Lee Bounds). Heterogeneity:
Subgroup analysis will include separate linear regressions by male and female children. We will also run quantile regressions [0.1;0.9] in case of anthropometric outcomes.
To address potential issues of spill-overs, we will use the following two-fold strategy:
(i) We will ask treated and control children about their 5 best friends at school in the baseline survey. We will also ask parents about the top 5 parents they interact with. In a robustness check we will then adjust for connections between control and treatment individuals or exclude heavily connected subjects.
(ii) We will also have a “pure control” school in which no kids receive the treatment package. This will allow us to conduct additional robustness checks and sensitivity analyses regarding the importance of treatment spillover effects. We will compare outcomes between control and pure control children and gauge (qualitatively) the sensitivity of baseline estimates to excluding and including pure control children. Multiple hypothesis testing:
We will calculate p-values and false discovery rate corrected q-values (Benjamini & Hochberg,1995) by family of indicators.
Average standardized effects:
We will give average effects on standardized indicators (Kling, Liebman, and Katz 2007) by family of indicators.
Experimental Design Details
Child level randomization after collection of baseline data. Balancing checks will be performed on baseline outcomes and background covariates characteristics (see list above).
Was the treatment clustered?
Sample size: planned number of clusters
6 schools with treated and control children and 1 pure control school.
Sample size: planned number of observations
We will survey 360 control and 360 treatment units (children in second to fourth grade, aged 7-10) across 6 schools in HCMC at baseline and after 6 months again. The 6 schools constitute the sampling frame to access children. 80 additional children will be sampled in a 7th pure control school. The sample will consist of overweight or obese children at baseline.
Sample size (or number of clusters) by treatment arms
Treatment: 360 children (in 6 schools)
Control: 440 children (360 in the 6 schools with also treated children present, 80 in one additonal, pure control school)
Sample size is 800 in both baseline and follow-up survey after 6 months. Total sample size is 1600.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Sample size was determined based on the power calculation for an individual RCT with small Cohen’s D of 0.21, power=0.8, significance=0.05. In R-Cran 3.1.2 using library(pwr) and command pwr.t.test(d = 0.21 , sig.level =0.05, power =0.8 ) gave a total sample size of 714, which we rounded up to 720. An additional 80 children were added from 1 pure control school.
INSTITUTIONAL REVIEW BOARDS (IRBs)
Research Ethics Committee at the International Institute of Social Studies of Erasmus University Rotterdam
IRB Approval Date
IRB Approval Number
Post Trial Information
Is the intervention completed?
Is data collection complete?