Video Abstract

Video Abstract

Close modal
BACKGROUND AND OBJECTIVES:

Malnutrition is a significant contributor to child morbidity and mortality globally. Egg consumption has been associated with improved child nutrition yet is rare in rural, resource-poor settings. We test the effects of a culturally tailored behavior change intervention to increase child egg consumption.

METHODS:

A 3-arm cluster randomized controlled trial was conducted in rural Burkina Faso with 260 mother-child dyads. Children aged 4 to 17 months from 18 villages were included; those with reported history of malnutrition or egg allergy were excluded. Each child in the full intervention arm received 4 chickens, and mothers received the 10-month behavior change package. Participants in the partial intervention arm received only the behavior change package.

RESULTS:

In this analysis of 250 children, the full (β = 4.3; P = 6.6 × 10−12) and the partial (β = 1.0; P = .02) interventions significantly increased egg consumption. The full intervention also significantly increased poultry production (β = 11.6; 95% confidence interval 8.3–15; P = 1.1 × 10−5) and women’s decision-making about eggs (β = .66; P = .02), and significantly decreased wasting (β = .58; P = .03) and underweight (β = .47; P = .02).

CONCLUSIONS:

The culturally tailored behavior change package significantly increased child egg consumption. When coupled with the gift of chickens, the behavior change intervention yielded a greater increase in egg consumption and significantly reduced wasting and underweight. Behavior change strategies to increase egg consumption should be considered among nutrition and health programs in resource-poor settings where poultry is available.

What’s Known on This Subject:

Where dietary diversity is low, consumption of animal-source foods can provide important missing nutritional elements required for full child growth and development. Consumption of animal-source foods is low in many resource-poor settings because of complex socioeconomic and cultural constraints.

What This Study Adds:

Culturally tailored behavior change interventions can significantly increase child egg consumption, even in resource-limited settings where egg consumption is low. When combined with the gifting of chickens, the behavior change intervention and livestock asset can change child growth trajectories.

Stunting, underweight, and wasting (3 forms of undernutrition) are important contributors to child mortality and poor health globally.1  Poor nutrition and health in early life has been associated with increased morbidity, poverty, and malnutrition and decreased economic and cognitive productivity in adulthood.2  Wasting and stunting contribute to increased susceptibility to infectious disease, physical and cognitive deficits, and high rates of infant mortality.3,4  The process of stunting typically begins in utero and continues through at least the first 2 years of life.2,5,6  Consequently, interventions to improve nutrition and health should target early critical windows of development with quantity and quality of nutrient-rich, safe foods.7 

Dietary diversity is limited among rural households in low- and middle-income countries (LMIC), resulting in insufficient intake of nutrients.8,9  In areas where food security is a primary concern, diets often rely on grains and may not include animal-source foods (ASFs) because of lack of availability or cost.1012  Consumption of ASFs has been associated with nutritional sufficiency and improved growth,1114  yet findings are inconclusive in a number of recent reviews and meta-analyses, at least in part because of poor quality of evidence.1214  Eggs are 1 of 3 types of ASFs included in most metrics of dietary diversity.15  Eggs are an excellent, renewable source of essential fatty acids, proteins, choline, vitamin A, vitamin B12, selenium, and other nutrients critical for child growth and development, yet their consumption remains low in many resource-poor settings.16  A 2017 trial in Ecuador revealed significant reduction in wasting and stunting among infants 6 to 9 months when eggs were provided to participating families for the child’s consumption,17  although nutritional improvements were not sustained after 1 year,18  and findings were not consistent in Malawi, where the study was replicated.19  A culturally tailored behavior change study in Ethiopia, targeting children 6 to 12 months of age, achieved a significant increase in child egg consumption20  and a significant reduction in the prevalence of underweight in targeted children.21 

Women are often the decision-maker about child diet and health care22,23  and tend to be more engaged in poultry production than in other livestock domains.24  Despite women’s role in poultry production and decision-making surrounding child diet, many barriers impede egg consumption in LMIC. These barriers exist at the individual, household, and community level and include the cost relative to other food items,8,25  cultural stigmas,16  food taboos,26  gender bias,27  and ceremonial use of chickens.28  Consequently, interventions to improve diets of infants and young children (IYC) through increased ASF consumption must take a socioecological approach to behavior change, including culturally tailored strategies to increase women’s decision-making capabilities.

In Burkina Faso, a small land-locked country in West Africa, up to 80% of families own poultry, but only 3% of children aged <2 consume eggs on a regular basis,29  and the prevalence of stunting (21%) and wasting (10%) remains high.30  Historical qualitative research described the Mossi of Burkina Faso as having sociocultural norms and understandings of nature and community resources that prohibit egg consumption among children.31  Narratives about children who eat eggs becoming thieves perpetuate to date (A. N'Diaye Wereme, PhD, personal communication, 2017). The Un Enfant, Un Oeuf, Par Jour (One Child, One Egg, Each Day), or Un Oeuf study, presented here, tested a culturally tailored behavior change communication (BCC) intervention to increase egg consumption and improve child nutrition by increasing poultry production and women’s decision-making. The primary outcome of the study was egg consumption. Child nutritional status (stunting, wasting, and underweight), women’s decision-making, and poultry production are also reported here.

A 1-year cluster randomized controlled trial was conducted with mother-child dyads in 18 rural villages in Burkina Faso. Eighteen rural villages were randomly selected and assigned to 1 of 3 intervention arms; all participants in a village received the same treatment. Ethical clearance was obtained from the University of Florida Institutional Review Board and the Committee of Ethics of the Government of Burkina Faso. Mothers provided oral consent for participation. Any child participating in the study who was found to have severe acute malnutrition (SAM) (weight-for-height z score of <−3, middle upper arm circumference of <115 mm, visible severe wasting, or the presence of nutritional edema) was referred to the nearest health clinic. The trial is registered at clinicaltrials.gov: NCT04135625. A full account of the study protocol, including design, methods, and baseline findings, is published elsewhere.32 

The study was conducted in Kaya Department among a vulnerable population of rural smallholder farmers. Households are largely dependent on rain-fed agriculture (sorghum or millet), and most practice a mix of crop- and livestock-based farming. Malnutrition rates are high, food security and dietary diversity are low, and although chickens are nearly ubiquitous in households, egg consumption among children is rare.32 

Urban communities in Kaya Department (n = 7) were excluded. From a list of the remaining 70 rural villages, by using a randomly generated number (59 413, generated May 13, 2018, on numbergenerator.org by 2 US-based researchers) and a running population total, 18 villages were randomly selected. Villages were systematically and randomly assigned to 1 of 3 intervention arms when a research team member in the US drew the village name from a box and sequentially assigned it to an intervention arm using a 1:1:1 ratio. The research arms were (1) full intervention, which included the gifting of 4 chickens and a culturally tailored BCC package; (2) partial intervention, which included only the culturally tailored BCC package (no chickens); and (3) control, which included no intervention. These research arm assignments were made under observation and with the assistance of a person with no ties to the trial. Village intervention assignment was not shared with the field team until after enumeration was complete.

Children were enumerated for eligibility in May and June 2018 by a field coordinator in collaboration with community health workers in each village. Exclusion criteria included a history of egg allergy or severe malnutrition, as reported by the mother or as visually observed by the community health worker, and sharing a household with another otherwise-eligible child. No child was reported to have had a history of severe malnutrition or egg allergies or presented with visual signs of severe malnutrition; thus no exclusions occurred for these reasons. The intervention targeted 15 children aged 6 to 12 months in each village. Because not enough age-eligible children were available in the selected villages during enumeration, age inclusion criteria were expanded to children aged 4.5 to 18 months (age at intervention). In villages that still had ≤15 eligible children (n = 5), all eligible children were selected for enrollment, including otherwise-eligible children from the same household (3 in the full intervention arm and 1 each in the partial intervention and control arms). By using the expanded inclusion criteria, 260 from 289 eligible children were randomly selected for recruitment and enrollment in each village (4 infants were <6 months old at the start of the intervention; caregivers received training, including instructions not to feed children eggs, until they were 6 months).

Ultimately, 260 IYC were enrolled by the field coordinator, with 83 children in the full intervention arm, 89 in the partial intervention arm, and 88 in the control arm (see Fig 1). For this analysis, 10 participating children were missing data at either baseline or end line for primary or secondary outcomes and were not included for analysis; thus, 250 children are included in this analysis.

FIGURE 1

Village and participant recruitment, enrollment, exclusion, and analysis flow diagram for the analysis of the study outcomes.

FIGURE 1

Village and participant recruitment, enrollment, exclusion, and analysis flow diagram for the analysis of the study outcomes.

Close modal

Each child in the full intervention arm received 4 chickens: 3 were gifted to the child by a community champion (although purchased by the project) designating the child as the beneficiary of the flock and its egg production,20  and 1 additional chicken was provided by the child’s family. Mother-child dyads in the full and partial intervention arms received the BCC package, which included monthly integrated nutrition and agriculture trainings; individual counseling of mothers during monthly monitoring sessions; distribution of a culturally appropriate, picture-based flipbook that reinforced key messages; development and fostered use of a project jingle (ie, “One child, one egg, each day!”); and engagement of strategically targeted community leaders as champions of the project. In a One Health approach, community health workers and agricultural extension workers collaboratively conducted monthly integrated nutrition and agriculture trainings for 10 months, beginning in July 2018 and ending in May 2019. Topics included the benefits of egg consumption, household hygiene and sanitation, infant and child feeding practices, and chicken husbandry practices. Individual monthly counseling sessions were provided during monitoring visits to caregivers receiving the full and partial interventions. During these visits, data collectors used women’s flipbooks, composed of culturally tailored images, to reinforce topics and provide support to mothers.

A household survey was implemented at baseline, midline, and end line to collect demographic, socioeconomic, livelihood, health, nutrition, poultry production, and decision-making information; in all other months, a shorter survey was implemented to monitor child egg consumption, health, and growth as well as poultry production and egg availability.32 

Child egg consumption at end line was the primary outcome of the study design. Egg consumption is defined as the number of eggs consumed by the child in the week before data collection, as reported by mothers.

Poultry production and women’s decision-making about eggs at end line, both identified pathways to improving nutrition, were secondary outcomes. Poultry production was defined as the number of chickens in the household. Women’s decision-making about eggs is indicated by answers to a survey question about who makes decisions about what to do with eggs, which was recoded as a binary variable (other = 0, self = 1). Both variables were assessed at baseline and end line, although end line status constitutes the outcome variable for each.

Nutritional outcomes were also secondary outcomes of the study and are indicated by z scores of child growth at end line. Anthropometry at baseline and end line was assessed for enrolled children, including weight to the nearest 0.1 kg (seca model 874 portable scale; seca, Birmingham, United Kingdom) and recumbent length to the nearest 0.1 cm (United Nations Children’s Fund height board). Measurements were repeated 3 times at collection to ensure accuracy; the median was used in the analysis.31  Wasting is defined by the weight-for-length z score (WLZs (WLZ < −2); underweight, the weight-for-age z score (WAZ) (WAZ < −2); and stunting, the length-for-age z score (LAZ) (LAZ < −2) according to the World Health Organization Child Growth Standards.33 

To determine sample size, a power analysis was conducted on the basis of differences in egg consumption observed in Omer’s study in Ethiopia (A.O., personal communication, May and October 2017). A sample size of 270 children (n = 90 per arm) was needed to detect at least a 13% difference between the percentage of children in the full intervention (13%) and control arms (0%) consuming ≥4 eggs per week by using a 2-sided z test (unpooled) with a P value of .05 and a power of 80%. An equal cluster size of 6 villages was assigned per treatment arm. An intracluster correlation of 0.02 and a design of effect of 1.28 were assumed.

Data collection, extraction, and cleaning was conducted in Research Electronic Data Capture, Excel, and R software.34  Anthropometric z score calculations were generated in R software by use of World Health Organization Child Growth Standards.33  Hypothesis testing for summary statistics was analyzed by using paired χ2 and analysis of variance tests, when appropriate, to compare summary statistics and project outcomes between intervention arms at baseline (Table 1). Hypothesis testing was performed by using 95% confidence intervals with 2-sided tests. The 6 outcomes for this study, defined above, consisted of 5 continuous regression models (egg consumption, poultry production, and 3 measures of child growth) and 1 logistic (logit) regression model (woman’s decision-making about what to do with household eggs). All regression models were constructed by using end line estimates for the outcome, with adjustment for baseline estimates. The regression models all controlled for fixed effects of the intervention groups as well as the random effects of village residence (cluster). The lme4 R package for cluster study designs was used to analyze the continuous outcomes, whereas the R geepack package was used to assess the logistic outcome.35,36 

TABLE 1

Baseline Characteristics and Outcomes of the Un Oeuf Study Population by Intervention Arm (N = 250)

Full Intervention (n = 77)Partial Intervention (n = 86)Control Intervention (n = 87)P value (N = 250)
Child     
 Age, mo, mean (SD) 9.93 (3.05) 9.51 (2.94) 9.63 (3.25) 0.667 
 Sex, female, n (%) 32 (41.6) 47 (54.7) 40 (46.0) 0.230 
Maternal     
 Age, y, mean (SD) 28.4 (6.42) 27.0 (6.29) 26.1 (6.87) 0.080 
 No formal education, n (%) 58 (75.3) 65 (75.6) 76 (87.4) 0.396 
 Not literate, n (%) 64 (83.1) 73 (84.9) 76 (87.4) 0.622 
Household     
 Household size, mean (SD) 18.7 (11.0) 17.9 (15.1) 17.7 (14.0) 0.894 
 Household livelihood, n (%)     
  Crop production 75 (97.4) 75 (97.4) 82 (94.3) 0.685 
  Animal husbandry 30 (39.0) 44 (51.2) 40 (46.0) 0.294 
  Other 1 (1.3) 2 (2.4) 4 (4.6) — 
Religion, n (%)    0.582 
 Muslim 58 (75.3) 69 (80.2) 64 (73.6) — 
 Christian 19 (24.7) 17 (19.8) 23 (26.4) — 
Ethnicity, n (%)     
 Mossi 73 (94.8) 79 (91.9) 83 (95.4) 0.600 
 Other 4 (5.2) 7 (8.1) 4 (4.6) — 
Household livestock ownership, n (%)     
 Any livestock 76 (98.7) 81 (94.2) 82 (94.3) 0.366 
 Cow 37 (48.1) 49 (57.0) 48 (55.2) 0.494 
 Sheep 58 (75.3) 58 (67.4) 65 (74.7) 0.502 
 Goat 55 (71.4) 51 (59.3) 61 (70.1) 0.206 
 Chicken 68 (88.3) 66 (76.7) 69 (79.3) 0.133 
 Donkey 57 (74.0) 59 (68.6) 62 (71.3) 0.736 
Water and sanitation, n (%)     
 Drinking source type    0.002 
  Improved, other 75 (97.4) 73 (84.9) 77 (88.5) — 
  Improved, piped 0 (0) 12 (14.0) 6 (6.9) — 
  Unimproved 2 (2.6) 1 (1.2) 3 (3.4) — 
 Latrine     
  Improved latrine 46 (59.7) 63 (73.3) 60 (69.0) 0.168 
 Practice open defecation 24 (31.2) 15 (17.4) 19 (21.8) 0.124 
Poultry production     
 No. chickens, median (range) 4.00 (0.00–100) 2.00 (0.00–100) 3.00 (0.00–60.0) 0.211 
Women’s empowerment, n (%)     
 Woman makes decisions about eggs 21 (27.3) 32 (37.2) 26 (29.9) 0.361 
Egg consumption     
 Child consumed eggs past week, n (%) 3 (3.9) 1 (1.2) 6 (6.9) — 
 No. of eggs child consumed in past wk, mean (SD) 0.117 (0.811) 0.023 (0.216) 0.195 (0.833) 0.459 
 Child consumed 4+ eggs in past wk, n (%) 1 (1.3) 0 (0.0) 3 (3.4) — 
Anthropometric status     
 WLZ     
  All wasting, mean (SD) −0.964 (1.31) −0.837 (1.14) −0.552 (0.957) 0.058 
  None, % ≥−2 SDs, n (%) 62 (80.5) 78 (90.7) 83 (95.4) — 
  Moderate, % <−2 SDs to ≥−3 SDs, n (%) 9 (11.7) 3 (3.5) 3 (3.4) — 
  Severe, % < −3 SDs, n (%) 6 (7.8) 5 (5.8) 1 (1.1) — 
 WAZ     
  All underweight, mean (SD) −1.24 (1.09) −1.30 (1.00) −1.05 (0.934) 0.254 
  None, % ≥ −2 SDs, n (%) 59 (76.6) 66 (76.7) 73 (83.9) — 
  Moderate, % < −2 SDs to ≥−3 SDs, n (%) 24 (15.6) 16 (18.6) 11 (12.6) — 
  Severe, % < −3 SDs, n (%) 6 (7.8) 4 (4.7) 3 (3.4) — 
 LAZ     
  All stunting, mean (SD) −1.04 (1.12) −1.25 (1.05) −1.19 (1.09) 0.448 
  None, % ≥ −2 SDs, n (%) 64 (83.1) 65 (75.6) 67 (77.0) — 
  Moderate, % < −2 SDs to ≥−3 SDs, n (%) 10 (13.0) 17 (19.8) 13 (14.9) — 
  Severe, % < −3 SDs, n (%) 3 (3.9) 4 (4.7) 7 (8.0) — 
Full Intervention (n = 77)Partial Intervention (n = 86)Control Intervention (n = 87)P value (N = 250)
Child     
 Age, mo, mean (SD) 9.93 (3.05) 9.51 (2.94) 9.63 (3.25) 0.667 
 Sex, female, n (%) 32 (41.6) 47 (54.7) 40 (46.0) 0.230 
Maternal     
 Age, y, mean (SD) 28.4 (6.42) 27.0 (6.29) 26.1 (6.87) 0.080 
 No formal education, n (%) 58 (75.3) 65 (75.6) 76 (87.4) 0.396 
 Not literate, n (%) 64 (83.1) 73 (84.9) 76 (87.4) 0.622 
Household     
 Household size, mean (SD) 18.7 (11.0) 17.9 (15.1) 17.7 (14.0) 0.894 
 Household livelihood, n (%)     
  Crop production 75 (97.4) 75 (97.4) 82 (94.3) 0.685 
  Animal husbandry 30 (39.0) 44 (51.2) 40 (46.0) 0.294 
  Other 1 (1.3) 2 (2.4) 4 (4.6) — 
Religion, n (%)    0.582 
 Muslim 58 (75.3) 69 (80.2) 64 (73.6) — 
 Christian 19 (24.7) 17 (19.8) 23 (26.4) — 
Ethnicity, n (%)     
 Mossi 73 (94.8) 79 (91.9) 83 (95.4) 0.600 
 Other 4 (5.2) 7 (8.1) 4 (4.6) — 
Household livestock ownership, n (%)     
 Any livestock 76 (98.7) 81 (94.2) 82 (94.3) 0.366 
 Cow 37 (48.1) 49 (57.0) 48 (55.2) 0.494 
 Sheep 58 (75.3) 58 (67.4) 65 (74.7) 0.502 
 Goat 55 (71.4) 51 (59.3) 61 (70.1) 0.206 
 Chicken 68 (88.3) 66 (76.7) 69 (79.3) 0.133 
 Donkey 57 (74.0) 59 (68.6) 62 (71.3) 0.736 
Water and sanitation, n (%)     
 Drinking source type    0.002 
  Improved, other 75 (97.4) 73 (84.9) 77 (88.5) — 
  Improved, piped 0 (0) 12 (14.0) 6 (6.9) — 
  Unimproved 2 (2.6) 1 (1.2) 3 (3.4) — 
 Latrine     
  Improved latrine 46 (59.7) 63 (73.3) 60 (69.0) 0.168 
 Practice open defecation 24 (31.2) 15 (17.4) 19 (21.8) 0.124 
Poultry production     
 No. chickens, median (range) 4.00 (0.00–100) 2.00 (0.00–100) 3.00 (0.00–60.0) 0.211 
Women’s empowerment, n (%)     
 Woman makes decisions about eggs 21 (27.3) 32 (37.2) 26 (29.9) 0.361 
Egg consumption     
 Child consumed eggs past week, n (%) 3 (3.9) 1 (1.2) 6 (6.9) — 
 No. of eggs child consumed in past wk, mean (SD) 0.117 (0.811) 0.023 (0.216) 0.195 (0.833) 0.459 
 Child consumed 4+ eggs in past wk, n (%) 1 (1.3) 0 (0.0) 3 (3.4) — 
Anthropometric status     
 WLZ     
  All wasting, mean (SD) −0.964 (1.31) −0.837 (1.14) −0.552 (0.957) 0.058 
  None, % ≥−2 SDs, n (%) 62 (80.5) 78 (90.7) 83 (95.4) — 
  Moderate, % <−2 SDs to ≥−3 SDs, n (%) 9 (11.7) 3 (3.5) 3 (3.4) — 
  Severe, % < −3 SDs, n (%) 6 (7.8) 5 (5.8) 1 (1.1) — 
 WAZ     
  All underweight, mean (SD) −1.24 (1.09) −1.30 (1.00) −1.05 (0.934) 0.254 
  None, % ≥ −2 SDs, n (%) 59 (76.6) 66 (76.7) 73 (83.9) — 
  Moderate, % < −2 SDs to ≥−3 SDs, n (%) 24 (15.6) 16 (18.6) 11 (12.6) — 
  Severe, % < −3 SDs, n (%) 6 (7.8) 4 (4.7) 3 (3.4) — 
 LAZ     
  All stunting, mean (SD) −1.04 (1.12) −1.25 (1.05) −1.19 (1.09) 0.448 
  None, % ≥ −2 SDs, n (%) 64 (83.1) 65 (75.6) 67 (77.0) — 
  Moderate, % < −2 SDs to ≥−3 SDs, n (%) 10 (13.0) 17 (19.8) 13 (14.9) — 
  Severe, % < −3 SDs, n (%) 3 (3.9) 4 (4.7) 7 (8.0) — 

—, not applicable.

Baseline characteristics of the Un Oeuf study population are presented in Table 1 (N = 250). Just less than half of children were female (47.6%), and mean age at enrollment was 9.68 (SD = 3.1) months. The mean maternal age was 27 (SD = 6.6) years. Most caregivers were illiterate 213 (85%) and had no formal education 199 (79.6%). Household livelihoods primarily included crop production 239 (95.6%) and animal husbandry 136 (54.4%). Almost all households owned livestock (95.6%), with the majority owning chickens 203 (81.2%). Importantly, because of random chance of village assignment, baseline differences between intervention arms existed for maternal age, drinking water source, and WLZs (Table 1).

Egg consumption among children at end line was significantly higher in both the full (β = 5.64; P = 3.0 × 10−14) and partial (β = 1.74.6; P = 1.8 × 10−6) intervention groups compared with the control group, after controlling for baseline egg consumption and random effects of village clustering (Table 2). As seen in Fig 2, egg consumption in the past week increased from a mean of 0.1 and 0.0 in the full and partial intervention groups, respectively, to a mean of 6.3 eggs and 2.4 eggs at end line.

TABLE 2

Regression Models for Outcomes of the Un Oeuf Study Population (N = 250)

Model TypeOutcomeβ (Reference = Control)P (<.05)R2cMaximum Likelihood EstimationICC
Continuous Egg consumption Full: 5.64 Full: 3.0 × 10−14 0.723 RMEL: 898.5 Adjusted: 0.010 
  Partial: 1.74 Partial: 1.8 × 10−6 — — Conditional: 0.003 
Continuous Poultry production Full: 11.7 Full: 1.1 × 10−5 0.42 RMEL: 1661 — 
  Partial: 3.1 Partial: .11 — — — 
Continuous Wasting Full: .58 Full: .03 0.47 RMEL: 613 — 
  Partial: .35 Partial: .19 — — — 
Continuous Underweight Full: .47 Full: .025 0.67 RMEL: 470 — 
  Partial: .32 Partial: .110 — — — 
Continuous Stunting Full: .02 Full: .9318 0.79 RMEL: 359 — 
  Partial: .05 Partial: .8079 — — — 
Logistic Women’s decision-making about eggs Full: 1.42 (RR: 4.1) Full: 3.65 × 10−5 0.01 AIC: 321.3 — 
  Partial: .78 (RR:2.2) Partial: .017 — — — 
Model TypeOutcomeβ (Reference = Control)P (<.05)R2cMaximum Likelihood EstimationICC
Continuous Egg consumption Full: 5.64 Full: 3.0 × 10−14 0.723 RMEL: 898.5 Adjusted: 0.010 
  Partial: 1.74 Partial: 1.8 × 10−6 — — Conditional: 0.003 
Continuous Poultry production Full: 11.7 Full: 1.1 × 10−5 0.42 RMEL: 1661 — 
  Partial: 3.1 Partial: .11 — — — 
Continuous Wasting Full: .58 Full: .03 0.47 RMEL: 613 — 
  Partial: .35 Partial: .19 — — — 
Continuous Underweight Full: .47 Full: .025 0.67 RMEL: 470 — 
  Partial: .32 Partial: .110 — — — 
Continuous Stunting Full: .02 Full: .9318 0.79 RMEL: 359 — 
  Partial: .05 Partial: .8079 — — — 
Logistic Women’s decision-making about eggs Full: 1.42 (RR: 4.1) Full: 3.65 × 10−5 0.01 AIC: 321.3 — 
  Partial: .78 (RR:2.2) Partial: .017 — — — 

In all regression models, the control group is used as the reference group. AIC, akaike information criterion; ICC, intraclass correlation coefficient; R2c, conditional R squared; RMEL, restricted maximum likelihood; —, not applicable.

FIGURE 2

Egg consumption by the child in the past week: baseline compared to end line by intervention arm.

FIGURE 2

Egg consumption by the child in the past week: baseline compared to end line by intervention arm.

Close modal

Poultry production significantly increased in the full intervention group (β = 11.7; P = 1.1 × 10−5) compared with the control group after controlling for baseline poultry production and random effects of village clustering (Table 2).

Decision-making by women (self) regarding eggs significantly increased in both the full (β = 1.42; relative risk = 4.1; P = 3.7 × 10−5) and partial (β = .78; P = .02) intervention groups compared with the control group after adjusting for baseline decision-making and random effects of village clustering (Table 2).

The full intervention significantly affected wasting (WLZ; β = .58; P = .03) as well as underweight (WAZ; β = .47; P = .02) after adjusting for baseline z scores and controlling for random effects of village clustering (Table 2). Whereas wasting increased among children in the control arm, wasting was slightly, but significantly, reduced among children who received the full intervention. Similarly, underweight increased among those in the control group but was slowed among children in the full intervention group. The intervention revealed no statistically significant effect on child stunting (LAZ; Table 2). Distributions of wasting, stunting, and underweight at baseline and end line are presented for each research arm in Fig 3.

FIGURE 3

Distribution of child z scores. A, Wasting by intervention arm: baseline compared to end line. B, Underweight by intervention arm: baseline compared to end line. C, Stunting by intervention arm: baseline compared to end line.

FIGURE 3

Distribution of child z scores. A, Wasting by intervention arm: baseline compared to end line. B, Underweight by intervention arm: baseline compared to end line. C, Stunting by intervention arm: baseline compared to end line.

Close modal

On the basis of previous research that underscore pathways by which agriculture can improve nutritional outcomes,17,20,3740  the research team designed a culturally tailored BCC intervention to increase egg consumption in rural villages of Burkina Faso. The project team aimed to facilitate women’s ability to include eggs in the diet of their children, thereby improving diet quality and consequent nutritional status of IYC. Results indicate that the BCC strategy was effective because poultry production, women’s decision-making surrounding eggs, and egg consumption all increased significantly. Child egg consumption in the full intervention arm increased from 0.1 to 6.3 eggs per week, with 95% of children ultimately eating ≥4 eggs per week. In the partial intervention group, in which only the BCC package was implemented, child egg consumption increased from 0.0 to 2.4 eggs per week. These findings contribute to previous research that has revealed that empowering women and providing education around health and nutrition are effective strategies to increase child dietary diversity22,41  and that high-quality foods, such as eggs, can be added through thoughtful BCC strategies.

Egg consumption increased in the partial intervention arm compared with the control arm but did not reach the level of egg consumption seen in the full intervention arm, in which participants received livestock assets in addition to the BCC strategy.20  These differences manifested in nutritional impact in the full intervention group, whereby both wasting and undernutrition were significantly improved. Although not statistically significant at P = .05, a lesser positive effect was also seen with the partial intervention. It is important to note that there were differences in the nutritional status of the research arms at baseline, despite randomization. Three villages (2 in the full intervention arm and 1 in the partial intervention arm) had children of significantly poorer nutritional status at baseline. These differences have been controlled for statistically in all models but are important when interpreting the findings. The full intervention group, in which wasting was worse at baseline, saw a small increase in median WLZs among children, whereas those in the control group, which was better off at baseline, worsened over time. A similar regression toward the mean is seen with the WAZs, although the full intervention only slowed the negative progression, compared to the control intervention. These findings indicate that although intervention made statistically significant nutritional impacts to both wasting and underweight, the most vulnerable, such as those with SAM (defined as a WAZ of < −3), were most affected by intervention. Given the significant impact of SAM on mortality of children42,43  from various causes, this reduction is important for researchers and practitioners.

The intervention revealed no significant effect on child stunting. However, the study was not powered around child growth outcomes, and stunting increased less in the full intervention arm than in the partial or control arms, indicating a positive trend. These findings are different from those seen by Iannotti et al17  in Ecuador, where egg consumption significantly reduced stunting during the period of the study, but are in line with other studies in which researchers have sought to replicate those findings.19,21  Stunting is multifactorial and is associated with diet, disease, and gut health44,45  and accrues over time. Repeated periods of wasting can be an important driver of chronic malnutrition.46  This study underlines the importance of BCC in livestock programs that are aimed at improving the nutritional status of children. Additional research is needed to understand if, and under what conditions, egg consumption can reduce stunting in children.

The study was powered around egg consumption; thus, the sample size was smaller than if it had been designed to test the intervention’s effect on child wasting, underweight, or stunting. This constrains the conclusion one can draw about the possible impact of egg consumption on stunting. Egg consumption data are based on maternal reporting and thus are subject to recall bias. In addition, women’s responses in the control arm are subject to demand bias and could reflect a change (or reported change) based on participation in the data collection process. Numerous approaches were undertaken to minimize effects of these limitations, including unannounced visits to monitor maternal and child behavior, branching logic in questionnaires (only when eggs were mentioned were additional questions asked), and qualitative interviews after the intervention to ensure research team understanding of survey reported data.

The BCC strategy to increase egg consumption among IYC significantly increased child egg consumption in a resource-poor setting, both with and without livestock asset distribution. The increase in egg consumption was significantly greater when the BCC strategy was coupled with a culturally tailored distribution of livestock assets, sufficient to reduce both wasting and underweight in children in the intervention compared with the control group. Efforts to increase ASF consumption among children in Burkina Faso and other LMIC should be focused on the combination of efforts to increase livestock assets with robust BCC strategies.

Dr McKune conceptualized and designed the study, coordinated and supervised data collection and analysis, and drafted the initial manuscript; Dr Stark conceptualized and designed the study, designed the data collection instruments, trained data collectors, and coordinated and supervised data collection; Mrs Sapp oversaw data management and conducted initial analyses; Dr Yang supervised and critically examined data analyses; Ms Slanzi designed data collection instruments and drafted the initial manuscript; Ms Moore oversaw data cleaning, contributed to data management, and contributed to data analyses; Mr Omer conceptualized and designed the study and trained data collectors; Dr Wereme N’Diaye conceptualized and designed the study and supported data collection; and all authors reviewed and revised the manuscript and approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

Deidentified individual participant data (including data dictionaries) will be made available through Dataverse (https://dataverse.org/) after September 30, 2020.

This trial has been registered at www.clinicaltrials.gov (identifier NCT04135625).

FUNDING: Funded in whole or part by the US Agency for International Development Bureau for Food Security under Agreement AID-OAA-L-15-00003 as part of Feed the Future Innovation Lab for Livestock Systems. Any opinions, findings, conclusions, or recommendations expressed here are those of the authors alone.

     
  • ASF

    animal-source food

  •  
  • BCC

    behavior change communication

  •  
  • IYC

    infants and young children

  •  
  • LAZ

    length-for-age z score

  •  
  • LMIC

    low- and middle-income countries

  •  
  • SAM

    severe acute malnutrition

  •  
  • WAZ

    weight-for-age z score

  •  
  • WLZ

    weight-for-length z score

1
Black
RE
,
Victora
CG
,
Walker
SP
, et al.;
Maternal and Child Nutrition Study Group
.
Maternal and child undernutrition and overweight in low-income and middle-income countries
.
Lancet
.
2013
;
382
(
9890
):
427
451
2
Dewey
KG
,
Begum
K
.
Long-term consequences of stunting in early life
.
Matern Child Nutr
.
2011
;
7
(
suppl 3
):
5
18
3
Grantham-McGregor
S
,
Cheung
YB
,
Cueto
S
,
Glewwe
P
,
Richter
L
,
Strupp
B
;
International Child Development Steering Group
.
Developmental potential in the first 5 years for children in developing countries
.
Lancet
.
2007
;
369
(
9555
):
60
70
4
Walker
SP
,
Wachs
TD
,
Grantham-McGregor
S
, et al
.
Inequality in early childhood: risk and protective factors for early child development
.
Lancet
.
2011
;
378
(
9799
):
1325
1338
5
Branca
F
,
Ferrari
M
.
Impact of micronutrient deficiencies on growth: the stunting syndrome
.
Ann Nutr Metab
.
2002
;
46
(
suppl 1
):
8
17
6
de Onis
M
,
Branca
F
.
Childhood stunting: a global perspective
.
Matern Child Nutr
.
2016
;
12
(
suppl 1
):
12
26
7
Prentice
AM
,
Ward
KA
,
Goldberg
GR
, et al
.
Critical windows for nutritional interventions against stunting
.
Am J Clin Nutr
.
2013
;
97
(
5
):
911
918
8
Morris
SS
,
Beesabathuni
K
,
Headey
D
.
An egg for everyone: pathways to universal access to one of nature’s most nutritious foods
.
Matern Child Nutr
.
2018
;
14
(
suppl 3
):
e12679
9
Ickowitz
A
,
Powell
B
,
Rowland
D
,
Jones
A
,
Sunderland
T
.
Agricultural intensification, dietary diversity, and markets in the global food security narrative
.
Glob Food Sec
.
2019
;
20
:
9
16
10
Allen
LH
.
Global dietary patterns and diets in childhood: implications for health outcomes
.
Ann Nutr Metab
.
2012
;
61
(
suppl 1
):
29
37
11
Mahfuz
M
,
Alam
MA
,
Das
S
, et al
.
Daily supplementation with egg, cow milk, and multiple micronutrients increases linear growth of young children with short stature
.
J Nutr
.
2020
;
150
(
2
):
394
403
12
Eaton
JC
,
Rothpletz-Puglia
P
,
Dreker
MR
, et al
. Effectiveness of Provision of Animal-Source Foods for Supporting Optimal Growth and Development in Children 6 to 59 Months of Age. In:
Cochrane Database Syst Rev
.
2019
:
CD012818
13
Shapiro
MJ
,
Downs
SM
,
Swartz
HJ
, et al
.
A systematic review investigating the relation between animal-source food consumption and stunting in children aged 6–60 months in low and middle-income countries
.
Adv Nutr
.
2019
;
10
(
5
):
827
847
14
Pimpin
L
,
Kranz
S
,
Liu
E
, et al
.
Effects of animal protein supplementation of mothers, preterm infants, and term infants on growth outcomes in childhood: a systematic review and meta-analysis of randomized trials
.
Am J Clin Nutr
.
2019
;
110
(
2
):
410
429
15
Workicho
A
,
Belachew
T
,
Feyissa
GT
, et al
.
Household dietary diversity and animal source food consumption in Ethiopia: evidence from the 2011 Welfare Monitoring Survey
.
BMC Public Health
.
2016
;
16
(
1
):
1192
16
Iannotti
LL
,
Lutter
CK
,
Bunn
DA
,
Stewart
CP
.
Eggs: the uncracked potential for improving maternal and young child nutrition among the world’s poor
.
Nutr Rev
.
2014
;
72
(
6
):
355
368
17
Iannotti
LL
,
Lutter
CK
,
Stewart
CP
, et al
.
Eggs in early complementary feeding and child growth: a randomized controlled trial
.
Pediatrics
.
2017
;
140
(
1
):
e20163459
18
Iannotti
LL
,
Chapnick
M
,
Nicholas
J
, et al
.
Egg intervention effect on linear growth no longer present after two years
.
Matern Child Nutr
.
2020
;
16
(
2
):
e12925
19
Stewart
CP
,
Caswell
B
,
Iannotti
L
, et al
.
The effect of eggs on early child growth in rural Malawi: the Mazira Project randomized controlled trial
.
Am J Clin Nutr
.
2019
;
110
(
4
):
1026
1033
20
Omer
A
,
Mulualem
D
,
Classen
H
,
Vatanparast
H
,
Whiting
SJ
.
A community poultry intervention to promote egg and eggshell powder consumption by young children in Halaba Special Woreda, SNNPR, Ethiopia
.
J Agric Sci
.
2018
;
10
(
5
):
1
12
21
Omer
A
,
Mulualem
D
,
Classen
H
,
Vatanparast
H
,
Whiting
SJ
.
Promotion of egg and eggshell powder consumption on the nutritional status of young children in Ethiopia
.
International Journal of Food Sciences and Nutrition Research
.
2019
;
1
(
1
):
1004
22
Olney
DK
,
Bliznashka
L
,
Pedehombga
A
,
Dillon
A
,
Ruel
MT
,
Heckert
J
.
A 2-year integrated agriculture and nutrition program targeted to mothers of young children in Burkina Faso reduces underweight among mothers and increases their empowerment: a cluster-randomized controlled trial
.
J Nutr
.
2016
;
146
(
5
):
1109
1117
23
Jin
M
,
Iannotti
LL
.
Livestock production, animal source food intake, and young child growth: the role of gender for ensuring nutrition impacts
.
Soc Sci Med
.
2014
;
105
:
16
21
24
Wong
JT
,
de Bruyn
J
,
Bagnol
B
, et al
.
Small-scale poultry and food security in resource-poor settings: a review
.
Glob Food Sec
.
2017
;
15
:
43
52
25
Headey
D
,
Hirvonen
K
,
Hoddinott
J
.
Animal sourced foods and child stunting
.
Am J Agric Econ
.
2018
;
100
(
5
):
1302
1319
26
Pachón
H
,
Simondon
KB
,
Fall
ST
, et al
.
Constraints on the delivery of animal-source foods to infants and young children: case studies from five countries
.
Food Nutr Bull
.
2007
;
28
(
2
):
215
229
27
Jayachandran
S
,
Pandi
R
.
Why are Indian children so short? The role of birth order and son preference
.
Am Econ Rev
.
2017
;
107
(
9
):
2600
2629
28
Kondombo
SR
,
Nianogo
AJ
,
Kwakkel
RP
,
Udo
HMY
,
Slingerland
M
.
Comparative analysis of village chicken production in two farming systems in Burkina Faso
.
Trop Anim Health Prod
.
2003
;
35
(
6
):
563
574
29
Gelli
A
,
Headey
D
,
Becquey
E
, et al
.
Poultry husbandry, water, sanitation, and hygiene practices, and child anthropometry in rural Burkina Faso
.
Matern Child Nutr
.
2019
;
15
(
4
):
e12818
30
US Agency for International Development
.
Burkina Faso: nutrition profile.
2018
. Available at: https://www.usaid.gov/sites/default/files/documents/1864/Burkina-Faso-Nutrition-Profile-Mar2018-508.pdf. Accessed February 8, 2020
31
Toussaint-Samat
M
.
A History of Food
, 2nd ed.
Hoboken, NJ
:
John Wiley & Sons, Inc
;
2009
32
Stark
H
,
Omer
A
,
Wereme N’Diaye
AW
,
Sapp
AC
,
Moore
EV
,
McKune
SL
.
The Un Oeuf study: design, methods and baseline data from a cluster randomised controlled trial to increase child egg consumption in Burkina Faso [published online ahead of print August 8, 2020]
.
Matern Child Nutr
. doi:
33
WHO Multicentre Growth Reference Study Group
.
Who Child Growth Standards: Length/Height-for-Age, Weight-for-Age, Weight-for-Length, Weight-for-Height and Body Mass Index-for-Age; Methods and Development
.
Geneva, Switzerland
:
World Health Organization
;
2006
34
Harris
PA
,
Taylor
R
,
Thielke
R
,
Payne
J
,
Gonzalez
N
,
Conde
JG
.
Research electronic data capture (REDCap)–a metadata-driven methodology and workflow process for providing translational research informatics support
.
J Biomed Inform
.
2009
;
42
(
2
):
377
381
35
Halekoh
U
,
Højsgaard
S
,
Yan
J
.
The R package geepack for generalized estimating equations
.
J Stat Softw
.
2006
;
15
(
2
):
10.18637/jss.v015.i02
36
Bates
D
,
Mächler
M
,
Bolker
B
,
Walker
S
.
Fitting linear mixed-effects models using lme4
.
J Stat Softw
.
2015
;
67
(
1
):
10.18637/jss.v067.i01
37
Ruel
MT
,
Quisumbing
AR
,
Balagamwala
M
.
Nutrition-sensitive agriculture: what have we learned so far?
Glob Food Sec
.
2018
;
17
:
128
153
38
Hawkes
C
,
Ruel
MT
.
From Agriculture to Nutrition: Pathways, Synergies and Outcomes
.
Washington, DC
:
World Bank Group
;
2008
39
Njuki
JM
,
Wyatt
A
,
Baltenweck
I, et al.
An exploratory study of dairying intensification, women’s decision making, and time use and implications for child nutrition in Kenya
.
Eur J Dev Res
.
2016
;
28
:
722
740
40
Price
M
,
Galie
A
,
Marshall
J
,
Agu
N
.
Elucidating linkages between women’s empowerment in livestock and nutrition: a qualitative study
.
Dev Pract
.
2018
;
28
(
4
):
510
524
41
Gelli
A
,
Becquey
E
,
Ganaba
R
, et al
.
Improving diets and nutrition through an integrated poultry value chain and nutrition intervention (SELEVER) in Burkina Faso: study protocol for a randomized trial
.
Trials
.
2017
;
18
(
1
):
412
42
Black
RE
,
Allen
LH
,
Bhutta
ZA
, et al.;
Maternal and Child Undernutrition Study Group
.
Maternal and child undernutrition: global and regional exposures and health consequences
.
Lancet
.
2008
;
371
(
9608
):
243
260
43
Pelletier
D L
.
The relationship between child anthropometry and mortality in developing countries: implications for policy, programs and future research
.
J Nutr
.
1994
;
124
(
10 Suppl
):
2047S
2081S
44
Prendergast
AJ
,
Humphrey
JH
.
The stunting syndrome in developing countries
.
Paediatr Int Child Health
.
2014
;
34
(
4
):
250
265
45
Aguayo
VM
,
Menon
P
.
Stop stunting: improving child feeding, women’s nutrition and household sanitation in South Asia
.
Matern Child Nutr
.
2016
;
12
(
suppl 1
):
3
11
46
Schoenbuchner
SM
,
Dolan
C
,
Mwangome
M
, et al
.
The relationship between wasting and stunting: a retrospective cohort analysis of longitudinal data in Gambian children from 1976 to 2016
.
Am J Clin Nutr
.
2019
;
110
(
2
):
498
507

Competing Interests

POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.

FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.