Poster Topical Area: Global Nutrition

Location: Hall D

Poster Board Number: 579

P12-077 - Determinants of Stunting among Preschool Children in the 2015-2016 Malawi Micronutrient Survey

Monday, Jun 11
8:00 AM – 3:00 PM

Objectives:

The 2015-16 Malawi Micronutrient Survey assessed the prevalence of nutritional indicators among preschool children 6-59 months (PSC) and found that approximately 1 in 3 were stunted. We aimed to identify multiple levels of predictors of stunting to inform targeted program development and implementation.


Methods:


PSC were stratified into 4 age categories: 6-11, 12-23, 24-35 and 36-59 months. We modeled bivariate associations between predictors and stunting (defined as height for age z-score < -2 SD) using logistic regression accounting for complex survey design. Multivariate modeling was performed using block stepwise regression. The primary predictor across each age group model was inflammation (either alpha-1-acid glycoprotein > 1g/L or C-reactive protein > 5mg/L), controlling for age (in months) and sex. Other blocks were grouped based on theorized proximity to stunting and included: 1) morbidity (diarrhea, respiratory infections, malaria, fever in the last 2 weeks); 2) individual micronutrient status; 3) infant and young child feeding practices; 4) exposure to unimproved sanitation and drinking water sources; and
5) household characteristics (socioeconomic status, rural/urban and household hunger). Variables were removed iteratively based on statistical significance.


Results:


A total of 34% of PSC were stunted; inflammation was highly prevalent at 57% overall. Inflammation was significantly associated with stunting among PSC aged 6-11 mo (OR: 13.0, 95% CI: 3.8-44.5) and 12-23 mo (OR: 2.2, 95% CI: 1.3-3.7). Vitamin A deficiency was uncommon with a prevalence of only 3.6% overall; however VAD was associated with stunting among PSC 12-23 mo (OR: 6.5, 95% CI: 1.1-38.2), and 36-59 mo (OR: 10.5, 95% CI: 3.5-31.6). Additionally, unimproved drinking water was associated with increased odds of stunting among children 6-11 mo (OR: 6.4, 95% CI: 1.2-33.6). For children 12-23 mo, fever reported in the last two weeks, (OR: 4.6, 95%CI: 1.9-11.5) was associated with stunting. For children 24-35 mo, ferritin deficiency predicted stunting (OR: 4.4, 95% CI: 1.8-11.2). Finally, for children 36-59 mo, moderate to severe household hunger was associated with stunting (OR 2.00, 95% CI: 1.2-3.5).


Conclusion:


The predictors of stunting differ by age among PSC. Inflammation was an important predictor of stunting in this population, particularly in the first 2 years of life. Efforts to target inflammation and its causes are likely needed to reduce the high prevalence of stunting, along with diverse nutrition-specific and nutrition-sensitive interventions.



Table 1 (part 1) Characteristics of children 6-59 months in the Malawi Micronutrient survey by age

Table 1 (part 2) Characteristics of children 6-59 months in the Malawi Micronutrient survey by age

Table 2 (part 1) Determinants of stunting in PSC using multiple logistic regression models stratified by age category

Table 2 (part 2) Determinants of stunting in PSC using multiple logistic regression models stratified by age category

CoAuthors: Parminder Suchdev, MD, MPH, FAAP – Emory University; Elizabeth Rhodes – Emory University; Anne Williams, PhD – Emory University

Rukshan Mehta

PhD Student
Emory University
Atlanta, Georgia