Determinants of undernutrition among children aged 6 to 59 months in rural Zambézia Province, Mozambique: Results of two population-based serial cross-sectional surveys.

Background: While many countries are transitioning from epidemics of undernutrition to overnutrition, Mozambique's very high 44% prevalence of stunting in children under age 5 years is cause for serious concern.

Methods: We conducted two population-based cross-sectional surveys of ~4000 female heads of households each in Zambézia Province, Mozambique from August-September 2010 (Baseline) and April-May 2014 (Endline) as part of the USAID funded Strengthening Communities through Integrated Programs (SCIP) grant. Anthropometric measurements were collected on 560 children aged 6-59 months at Baseline and 912 children at Endline and classified as: "stunted," a height-for-age z-score less than -2; "wasted," weight-for-height z-score less than -2; and "underweight," weight-for-age z-score less than -2. Descriptive statistics and logistic regression using Stata 13.1 were used to examine factors associated with undernutrition.

Results: Of children under age five years, 43% were undernourished in 2010 and 55% in 2014. The most common form of undernutrition was stunting (39% in 2010, 51% in 2014), followed by underweight (13% in both 2010 and 2014), and wasting (7% in 2010, 5% in 2014). Child's age was found to have a non-linear association with stunting. Vitamin A supplementation was associated with a 31% (p=0.04) decreased odds of stunting. Children who were exclusively breastfed for at least six months had an 80% (p=0.02) lower odds of wasting in 2014 and 57% (p=0.05) decreased odds of being underweight in 2014. Introducing other foods after age six months was associated with a five-fold increased odds of wasting in 2014 (p=0.02); household food insecurity was associated with wasting (OR=2.08; p=0.03) and underweight in 2010 (OR=2.31; p=0.05). Children whose mother washed her hands with a cleaning agent had a 40% (p=0.05) decreased odds of being underweight. Surprisingly, per point increase in household dietary diversity score, children had 12% greater odds of being stunted in 2010 (p=0.01) but 9% decreased odds of being underweight in 2014 (p=0.02).

Conclusions: A combination of household and individual level factors was associated with undernutrition. As such, employment of multidimensional interventions should be considered to decrease undernutrition in children under five years old.