The effect of an intervention to reduce aflatoxin consumption from 6 to 18 mo of age on length-for-age z-scores in rural Tanzania: a cluster-randomized trial.
Erica Phillips, Francis M Ngure, Neema Kassim, Paul C Turner, Edna Makule et al. ·
Summary
Written by AI, not yet AICS's view · claude-haiku-4-5-20251001
This study tested whether providing low-aflatoxin maize and groundnut flour to infants aged 6-18 months in rural Tanzania would improve their growth. Aflatoxins are toxic substances found in some foods that laboratory studies suggest may stunt children's growth. The intervention did not reduce aflatoxin exposure or improve growth outcomes, partly because aflatoxin levels were unexpectedly low during the trial period due to favorable weather conditions.
Key points
The intervention provided low-aflatoxin maize and groundnut flour monthly to 2,842 maternal-infant dyads in 52 health facilities over 12 months
Mean length-for-age z-scores at 18 months were -1.83 in the intervention group versus -1.90 in the control group (no significant difference)
Aflatoxin blood biomarker (serum AF-albumin) prevalence decreased from 54% at baseline to 36% at 18 months in both groups equally
Drought and favorable agricultural conditions during the trial resulted in low levels of aflatoxin exposure in both arms, preventing a meaningful test of the intervention
The authors conclude that while stunting and intermittent aflatoxin exposure remain problems in the region, the trial could not demonstrate whether reducing aflatoxin improves growth
Limitations
The trial could not adequately test the intervention hypothesis because aflatoxin exposure levels were unexpectedly low in both groups due to favorable weather conditions. The results do not tell us whether aflatoxin reduction would work under conditions of higher contamination, which are common in the region during other seasons or years.
Abstract
BACKGROUND: Linear growth faltering continues to negatively affect children in low- and middle-income countries and is associated with poor cognitive, developmental, and educational outcomes. Laboratory and observational data suggest that aflatoxin (AF) is a contributor to stunting. OBJECTIVE: The Mycotoxin Mitigation Trial was a cluster-randomized, community-based 2-group trial conducted in Kongwa District, Tanzania, between 2018 and 2020. The trial assessed whether a 12-mo intervention to reduce AF consumption increased length-for-age z-scores (LAZ) at 18 mo. METHODS: Low-AF maize and groundnut flours were provided to the intervention group each month; skin lotion was distributed to the control group monthly. Infant and young child feeding education was delivered equally in 52 health facilities (clusters). Anthropometry and the AF blood biomarker serum AF-albumin (AF-alb) were assessed at 6, 12, and 18 mo of age. Outcomes were analyzed as intention-to-treat and per-protocol using linear mixed-effects models. RESULTS: Two thousand eight hundred forty-two maternal-infant dyads were recruited into the study. The intervention did not create a contrast in AF-alb. At 18 mo, 36% (n = 186/520) of infants had detectable levels of AF-alb compared with 54% (n = 195/364) at baseline, with no difference between groups. Mean LAZ in the intervention group at 18 mo was -1.83 (n = 1231, 95% CI: -1.93, -1.73) compared to -1.90 (n = 1287, 95% CI: -1.99, -1.82) in the control group (P = 0.28). CONCLUSIONS: An intervention to reduce AF exposure did not alter AF-alb nor anthropometric measures between treatment groups. Drought and agricultural data indicated less favorable conditions for toxin production, resulting in low levels of exposure in both trial arms. Annual, seasonal, and geographic heterogeneity of AF contamination make it difficult to study in an ethically designed trial. Our formative research and early trial data indicate that stunting and intermittent AF exposure continue to be a problem in this region. However, the low-AF exposure levels during the trial led to inconclusive results. TRIAL REGISTRATION NUMBER: NCT03940547 (ClinicalTrials.org).
Findings the Hub has extracted
Each claim below is labelled by its type of design and links back to this report.
