Wasting, Stunting, and Anemia in Angolan Children after Deworming with Albendazole or a Test-and-Treat Approach for Intestinal Parasites: Binary Longitudinal Models with Temporal Structure in a Four-Arm Randomized Trial

Abstract

Undernutrition, anemia, and intestinal parasitic infections are public health problems in Angola, especially in pre-school children. We analyzed binary data from a longitudinal four-arm randomized parallel trial conducted in Bengo Province, northern Angola, over the course of two years, with seven follow-up assessments to explore the effects of four interventions (deworming and a test-and-treat approach for intestinal parasites, at both the individual and household levels) on wasting and stunting, and to understand their indirect benefits for anemia, malaria, diarrhea, and vomiting. A total of 121 children with intestinal parasitic infections received baseline treatment, and were allocated to the four arms (1:1:1:1). Using continuous outcome variables of height-for-age (HAZ) and weight-for-height (WHZ) statistical approaches did not reveal a clear benefit of any particular arm (Pathogens 2021, 10, 309). Next, HAZ and WHZ were transformed into binary variables of stunting and wasting, respectively, considering their mild-to-severe (Z-score < −1) and moderate-to-severe degrees (Z-score < −2). Original clinical data (on anemia, diarrhea, vomiting, and malaria) were also analyzed. From a binary longitudinal analysis with different dependence structures, using the R package bild, fitted models revealed the potential benefit of a test-and-treat approach at the individual level for wasting compared with annual albendazole at the individual level, especially considering mild-to-severe forms (ORadj = 0.27; p = 0.007). All arms showed similar effects on stunting, compared with annual albendazole, at a 5% significance level. Time and age at baseline presented favorable effects in the percentage of stunting using both severity degrees. Results showed a decreased chance of having anemia and diarrhea over time, although with no significant differences between arms. Data from longitudinal studies are essential to study the direct and indirect effects of interventions, such as deworming, and to explore additional approaches aiming at better understanding the temporal structure of nutrition and health outcomes in children.

Keywords: Angola; anemia; deworming; intestinal parasites; longitudinal; malnutrition; mild-to-severe; moderate-to-severe; stunting; test-and-treat; wasting.

Figure A1

Residual analysis for model M8 considering mild-to-severe stunting.

Figure 1

Randomization, allocation, and follow-up time of participants.

Figure 2

Baseline sociodemographic characteristics, nutritional status, and health conditions of the participants (n = 121, t = 0).

Figure 3

Boxplots for age at baseline by (a) arms: 1 to 4 and (b) sex: male or female.

Figure 4

Observed stunting and wasting profiles by arm: (a) mild-to-severe stunting; (b) moderate-to-severe stunting; (c) mild-to-severe wasting; (d) moderate-to-severe wasting.

Figure 5

Observed stunting and wasting profiles by sex: (a) mild-to-severe stunting; (b) moderate-to-severe stunting; (c) mild-to-severe wasting; (d) moderate-to-severe wasting.

Figure 6

Stunting profiles by age and follow-up (t = 0,1,2,3,4,5,6): (a) mild-to-severe stunting; (b) moderate-to-severe stunting.