Antibiotic Mass Treatment Fails to Improve Child Nutrition in Burkina Faso Trial

Twice-yearly mass distribution of the antibiotic azithromycin has become one of the most closely watched interventions in global child health, after large trials showed it can reduce deaths among young children in some of the world’s poorest settings. But a new secondary analysis from Burkina Faso adds an important caveat to the story: the drug does not appear to improve children’s nutritional status, a mechanism that researchers had suspected might partly explain its mortality benefits. The findings, published in PLOS Global Public Health, suggest that the life-saving effects of mass azithromycin are unlikely to be mediated through better growth or reduced undernutrition.

The study drew on the Child Health with Azithromycin Trial, known as CHAT, a placebo-controlled cluster randomized trial conducted in Burkina Faso in which entire communities were assigned, in a one-to-one ratio, to receive twice-yearly azithromycin or a matching placebo for children aged one to 59 months. The original trial was designed to test whether the antibiotic could prevent deaths from any cause, extending to a high-mortality African setting the earlier evidence from the landmark MORDOR trial in Niger, Malawi, and Tanzania, which had demonstrated a significant reduction in all-cause child mortality with mass drug administration.

The research team, led by investigators at the University of California, San Francisco and the Centre de Recherche en Santé de Nouna in Burkina Faso, asked a deceptively simple question: if azithromycin works by clearing infections, and if infections sap children’s growth, shouldn’t treated children grow better? To answer it, they tracked two standard measures of undernutrition. The first was mid-upper arm circumference, or MUAC, a tape measurement of the upper arm that is widely used in field settings to screen for acute malnutrition because it changes little with age in early childhood and requires no scales. The second was the weight-for-age Z-score, or WAZ, a statistical measure of how far a child’s weight deviates from the expected weight for a healthy child of the same age and sex, calculated here for infants aged one to eleven months.

The scale of the analysis was considerable. The researchers included 80,282 children aged one to 59 months, of whom 26,983 were infants in the one-to-eleven-month age range, drawn from 284 communities. Measurements were collected every six months over the course of the trial, with MUAC taken in all children and weight recorded in infants both to calculate weight-based drug dosing and to enable the nutritional analysis. Because communities rather than individuals were randomized, the statistical methods accounted for clustering, comparing outcomes between the azithromycin and placebo arms at the community level and also following individual children longitudinally over time.

The results were unambiguous in their null finding. At the community level, after 36 months of intervention there was no evidence of any difference in mean MUAC or mean WAZ between villages that had received azithromycin and those that had received placebo. Nor was there any divergence over time: the change in MUAC over the follow-up period showed no difference between arms, with a P-value for interaction of 0.94, and the change in WAZ similarly showed no difference, with a P-value of 0.18. In plain terms, children in azithromycin communities did not have measurably better arm circumference or weight gain than children in placebo communities, either on average or in their growth trajectories.

These findings matter because they speak directly to the leading hypothesis about how mass azithromycin saves lives. The drug, a long-acting macrolide antibiotic, is known to treat a broad range of bacterial infections and some intracellular pathogens, including subclinical infections that never announce themselves with fever or clinic visits. Researchers have plausibly argued that reducing this invisible infectious burden could free up calories and nutrients that would otherwise be diverted to immune defense, allowing children to grow better and thereby lowering their vulnerability to severe illness and death. Undernutrition and infection form a well-documented vicious cycle in early childhood: infection suppresses appetite and impairs nutrient absorption, while malnutrition weakens immune defenses and increases the severity of subsequent infections. Breaking that cycle at the infection end, the reasoning went, should improve the nutrition end.

The CHAT nutritional analysis suggests it did not, at least not in this setting. One interpretation is that the infections most responsible for azithromycin’s mortality benefit are acute, severe episodes rather than the chronic, growth-slowing burden that shapes anthropometric measures over months and years. Mortality in these settings is often driven by rapidly progressive infections such as pneumonia, sepsis, and malaria-associated bacterial coinfections, which can kill a child long before any nutritional pathway could come into play. If azithromycin primarily prevents these fulminant events, its effect on growth would be expected to be small or absent, exactly what the trial observed.

Another possibility is that the dominant drivers of undernutrition in Burkina Faso lie outside the reach of a twice-yearly antibiotic. Diarrheal disease from enteric pathogens, environmental enteric dysfunction driven by repeated exposure to fecal contamination, dietary insufficiency, and malaria are all entrenched in the region, and a macrolide given twice a year may not durably suppress the microbial ecology that stunts growth. Some of the pathogens that most strongly depress growth, such as certain enteroviruses and parasites, are not susceptible to azithromycin at all. The drug’s effects on the gut microbiome, moreover, are complex, and repeated exposure has been associated with rising antimicrobial resistance, a concern that weighs on any expansion of mass distribution programs.

For policymakers, the analysis does not undermine the case for mass azithromycin where mortality benefits have been demonstrated, but it does sharpen the picture of what the intervention can and cannot do. Countries weighing twice-yearly distribution should not expect it to serve as a nutrition program, and investments in the established tools of nutrition improvement, including infant and young child feeding, micronutrient supplementation, food security, and water, sanitation, and hygiene infrastructure, remain essential alongside any antibiotic strategy. The finding also redirects scientific attention toward identifying the actual biological mechanisms behind the mortality reduction, whether through targeted studies of severe infection incidence, biomarkers of systemic inflammation, or cause-of-death ascertainment in future trials.

The study also illustrates the value of embedding mechanistic questions within large mortality trials. Because CHAT already collected MUAC and weight measurements at six-month intervals across nearly three hundred communities, the researchers could test the nutritional hypothesis with statistical power that no standalone study of this size could easily achieve, while respecting the cluster randomized design that guards against contamination between intervention and comparison groups. The trial was registered as NCT03676764 on ClinicalTrials.gov, and the analysis protocol was specified in advance as a secondary objective, reducing the risk of post hoc interpretation. As mass azithromycin programs move from trials toward policy in high-mortality settings, studies like this one will be essential for calibrating expectations, ensuring that a genuinely promising mortality intervention is deployed for what it does, and not credited with benefits the evidence does not support.

Subject of Research: Effects of twice-yearly mass azithromycin distribution on nutritional status in young children

Article Title: Mass azithromycin distribution and nutritional status among children aged 1–59 months: A secondary analysis of a cluster randomized trial

Article References: Ouattara, M., Hu, H., Bountogo, M., Coulibaly, B., Boudo, V., Compaoré, G., Dah, C., Lebas, E., O’Brien, K. S., Arnold, B. F., Lietman, T. M., Sié, A., & Oldenburg, C. E. (2026). Mass azithromycin distribution and nutritional status among children aged 1–59 months: A secondary analysis of a cluster randomized trial. PLOS Global Public Health, 6(10), e0007473. https://doi.org/10.1371/journal.pgph.0007473

Image Credits: AI Generated

DOI: 10.1371/journal.pgph.0007473

Keywords: azithromycin, child mortality, malnutrition, MUAC, weight-for-age Z-score, Burkina Faso, cluster randomized trial, mass drug administration, global health, undernutrition, CHAT trial, PLOS Global Public Health