Can Regular Antibiotic Doses Protect Babies in West Africa? Weighing Lives Saved Against Antibiotic Resistance

What the LAKANA trial in Mali (Haidara et al., 2025; Adubra et al., 2026) and the AVENIR trial in Niger (Doan et al., 2026) found when the antibiotic azithromycin was given to babies aged 1 to 11 months

Large studies in Mali and Niger tested whether giving babies aged 1 to 11 months a dose of the antibiotic azithromycin every few months could prevent deaths in places where many young children die. Treating only babies did not save lives or help them grow, but it also did not noticeably raise antibiotic resistance. Treating all children under five saved lives in Niger but slightly increased resistance genes in gut bacteria, a trade-off that health planners must now weigh.

If your baby is in a neonatal unit, or has recently come home, it is worth saying at the start that these studies were about community health programmes in rural West Africa, and that no baby younger than one month was given the medicine. Still, they matter to anyone who cares about how babies fare after they leave hospital, because in many parts of the world the months after the newborn period remain dangerous.

The three reports behind this article

This article brings together three research reports from two very large studies. The first, from Mali, is called the Large-scale Assessment of the Key health-promoting Activities of two New mass drug administration regimens with Azithromycin (LAKANA) trial. It took place in 1,151 villages in the Kayes, Kita and Koulikoro regions of Mali, led by the Center for Vaccine Development–Mali and Tampere University in Finland, and it asked whether the medicine reduced baby deaths [1]. The second LAKANA report looked at how babies grew in 59 of those villages, in the Kita region [2]. The third comes from Niger, from the Azithromycine pour la Vie des Enfants au Niger: Implémentation et Recherche (AVENIR; in English, Azithromycin for Child Survival in Niger: Implementation and Research) trial, and looked at antibiotic resistance in 150 communities of the Dosso region [3]. Resistance means that bacteria have learned to survive an antibiotic, so the medicine stops working as well when someone really needs it.

Why doctors hoped a simple antibiotic dose could save babies

The idea grew out of trachoma control. Azithromycin was given to whole communities to control trachoma, an eye infection, and researchers noticed that fewer children seemed to die afterwards [1]. To test this properly, a study called Macrolides Oraux pour Réduire les Décès avec un Oeil sur la Résistance (MORDOR), French for "oral macrolides to reduce deaths with an eye on resistance", gave the medicine twice a year to all children aged 1 to 59 months in 1,533 communities in Malawi, Niger and Tanzania. Deaths fell by 13.5% overall, and by about a quarter among babies aged 1 to 5 months [4]. The protection seemed strongest in the three months after each dose [5], and largest where more children were dying to begin with [6]. But there was a catch. In treated communities in Niger, researchers found more bacteria that could resist azithromycin [7], and after four years gut bacteria in treated communities carried about seven times as many azithromycin-resistance genes, along with more genes for resistance to other types of antibiotic [8].

So in 2020 the World Health Organization (WHO) published guidance [9]. As the LAKANA researchers describe it, the guidance suggested that countries with high child death rates could consider giving the medicine only to babies aged 1 to 11 months, to keep most of the benefit while limiting resistance [1]. It was a sensible compromise, but nobody had yet tested whether it actually worked [10]. The LAKANA and AVENIR trials were designed to find out.

How the LAKANA study in Mali worked

Because the medicine might protect a whole community and not just the child who swallows it, the researchers compared villages rather than individual babies. This kind of study is called a cluster randomized trial: each village (a "cluster") is assigned to a group by chance, so the groups are alike, on average, in every way except the medicine. The Mali villages were divided by lottery into three groups. Every three months, trained workers visited each village, weighed every baby aged 1 to 11 months, and gave a liquid dose measured to the baby's weight. In one group the liquid was always a placebo, a look-alike with no medicine in it. In another, babies received azithromycin twice a year, at the visits between January and June, so that it did not overlap with the malaria-prevention medicine that Mali gives children from July to December. In the third, they received it every three months. Neither families nor staff knew which group a village was in, which keeps the comparison fair. Babies had to be at least one month old and weigh at least 3 kg, a deliberate safety step because macrolide antibiotics such as azithromycin are linked to a rare stomach problem in newborns [1]. The study's plans were published in advance [11].

What happened to survival and growth

About 149,000 babies took part, and sadly 968 died during the study. Researchers counted deaths per "person-year", meaning one baby followed for one year. In all three groups the rate was almost the same, about 12 deaths for every 1,000 baby-years: 11.9 with placebo, 11.8 with twice-yearly azithromycin and 11.3 with azithromycin every three months [1]. These differences are small enough to be due to chance. Older children in the same villages, aged 1 to 4 years, who were not treated, did not benefit either. Encouragingly, no serious side effects linked to the medicine were reported.

The growth report measured the weight, length and arm size of 1,789 children aged 6 to 8 months or 12 to 14 months. Children from azithromycin villages grew just like children from placebo villages [2]. For example, stunting, which means being much shorter than expected for one's age, was found in roughly 27% to 33% of measurements at 12 to 14 months in every group, and the small differences between groups could easily have been due to chance. Another report from the AVENIR trial in Niger likewise found no difference in babies' weight [12].

What the AVENIR study in Niger found about resistance

In Niger, communities were also divided by lottery into three groups: azithromycin twice a year for all children under five, azithromycin for babies only (with placebo for older children), or placebo for everyone. An earlier AVENIR report had found 14% fewer child deaths when all under-fives were treated, while treating babies alone made no clear difference: 22.3 compared with 23.9 baby deaths per 1,000 baby-years [10].

For the new report, researchers visited 150 communities six months after the fourth round of doses and gently swabbed about 30 children in each, collecting more than 4,000 samples from the bottom and more than 4,000 from the back of the nose [3]. The samples from each community were mixed together and searched for resistance genes. Where all under-fives had been treated, gut bacteria carried about 16% more genes for resistance to azithromycin and related antibiotics than with placebo. Where only babies had been treated, there was no detectable increase. Reassuringly, there was no rise in resistance to other antibiotic families, such as the penicillin-type medicines used for many infections, and no clear change in the nose. Because samples were pooled, these results describe whole communities, not individual children.

Why giving medicine only to babies may not be enough

The LAKANA researchers point out that far fewer babies died than they had expected, which made any benefit harder to detect [1]. The LAKANA researchers also suggest, in both of their reports, that older children may carry and spread the infections that kill babies, so treating them may protect babies indirectly [1][2]. A further Niger analysis fits this idea: babies who lived with older brothers or sisters seemed to do better when those siblings were treated, although the finding was not certain [13].

What this means for families

For families, the message is calm. These studies do not change how sick babies are cared for, and they do not suggest that healthy babies should be given antibiotics. A study in Burkina Faso that gave a single dose to newborns aged 8 to 27 days found no difference in growth by six months [14], and another that gave a dose at routine baby check-ups at 5 to 12 weeks did not reduce deaths [15]. The best protection for a baby remains good everyday care and prompt treatment when a baby is unwell. If you want to read more about why antibiotics are used carefully in newborn units, see https://inneosight.org/en-US/articles/neonatal-amr-transmission-to-stewardship-2026.

What researchers are working on next

Health planners now face a difficult choice. Treating only babies keeps resistance low but does not seem to save lives; treating all young children saves lives but encourages resistance. Some experts argue that the benefits can justify the antibiotic use if resistance is carefully tracked [16], while others call for rethinking who should be eligible, with a fairer sharing of benefits and burdens between generations [17]. The AVENIR team says close monitoring of resistance should be part of any such programme [3]. Researchers in Niger are still testing samples to see whether resistance has spread to older children and adults, and whether the bacteria themselves, not just their genes, have become harder to treat [3].

The bottom line

Giving azithromycin only to babies aged 1 to 11 months, twice a year or every three months, appeared safe but did not save lives or improve growth in Mali, and in Niger treating only babies did not detectably increase resistance genes. Treating all young children can save lives but slightly increases resistance, so any future programme will need to watch resistance closely.

References

  1. Haidara FC, Adubra L, Abdou M, et al. Mass Administration of Azithromycin to Infants in Mali to Reduce Mortality. N Engl J Med. 2025;393(15):1498-1508. doi:10.1056/NEJMoa2504644 ↩
  2. Adubra L, Luoma J, Fan YM, et al. Infant Growth After Mass Administration of Azithromycin: Secondary Outcomes of a Cluster Randomized Clinical Trial. JAMA Netw Open. 2026;9(6):e2617425. doi:10.1001/jamanetworkopen.2026.17425 ↩
  3. Doan T, Yan D, Arzika AM, et al. Mass azithromycin distribution and antibiotic resistance in the gut and nasopharynx: a cluster-randomized trial. Nat Med. 2026;32(3):859-868. doi:10.1038/s41591-026-04217-9 ↩
  4. Keenan JD, Bailey RL, West SK, et al. Azithromycin to Reduce Childhood Mortality in Sub-Saharan Africa. N Engl J Med. 2018;378(17):1583-1592. doi:10.1056/NEJMoa1715474 ↩
  5. Porco TC, Hart J, Arzika AM, et al. Mass Oral Azithromycin for Childhood Mortality: Timing of Death After Distribution in the MORDOR Trial. Clin Infect Dis. 2019;68(12):2114-2116. doi:10.1093/cid/ciy973 ↩
  6. Oron AP, Burstein R, Mercer LD, et al. Effect Modification by Baseline Mortality in the MORDOR Azithromycin Trial. Am J Trop Med Hyg. 2020;103(3):1295-1300. doi:10.4269/ajtmh.18-1004 ↩
  7. Doan T, Arzika AM, Hinterwirth A, et al. Macrolide Resistance in MORDOR I - A Cluster-Randomized Trial in Niger. N Engl J Med. 2019;380(23):2271-2273. doi:10.1056/NEJMc1901535 ↩
  8. Doan T, Worden L, Hinterwirth A, et al. Macrolide and Nonmacrolide Resistance with Mass Azithromycin Distribution. N Engl J Med. 2020;383(20):1941-1950. doi:10.1056/NEJMoa2002606 ↩
  9. World Health Organization. WHO Guideline on Mass Drug Administration of Azithromycin to Children under Five Years of Age to Promote Child Survival. Geneva: World Health Organization; 2020. ISBN 9789240009585. https://www.who.int/publications/i/item/9789240009585 ↩
  10. O'Brien KS, Arzika AM, Amza A, et al. Azithromycin to Reduce Mortality - An Adaptive Cluster-Randomized Trial. N Engl J Med. 2024;391(8):699-709. doi:10.1056/NEJMoa2312093 ↩
  11. Adubra L, Alber D, Ashorn P, et al. Testing the effects of mass drug administration of azithromycin on mortality and other outcomes among 1-11-month-old infants in Mali (LAKANA): study protocol for a cluster-randomized, placebo-controlled, double-blinded, parallel-group, three-arm clinical trial. Trials. 2023;24(1):5. doi:10.1186/s13063-022-06966-7 ↩
  12. Peterson B, Arzika A, Maliki R, et al. Impact of biannual azithromycin on weight-for-age z-score among infants in the AVENIR cluster-randomised trial. BMJ Paediatr Open. 2025;9(1):e004037. doi:10.1136/bmjpo-2025-004037 ↩
  13. Arzika AM, Amza A, Maliki R, et al. Spillover of Azithromycin Mass Drug Administration and Child Survival: A Secondary Analysis of a Cluster-Randomized Clinical Trial. JAMA Netw Open. 2025;8(7):e2519693. doi:10.1001/jamanetworkopen.2025.19693 ↩
  14. Sie A, Bountogo M, Zakane A, et al. Neonatal Azithromycin Administration and Growth during Infancy: A Randomized Controlled Trial. Am J Trop Med Hyg. 2023;108(5):1063-1070. doi:10.4269/ajtmh.22-0763 ↩
  15. Sié A, Ouattara M, Bountogo M, et al. Azithromycin during Routine Well-Infant Visits to Prevent Death. N Engl J Med. 2024;390(3):221-229. doi:10.1056/NEJMoa2309495 ↩
  16. Kahn R, Eyal N, Sow SO, Lipsitch M. Mass drug administration of azithromycin: an analysis. Clin Microbiol Infect. 2023;29(3):326-331. doi:10.1016/j.cmi.2022.10.022 ↩
  17. Goh M, Viens AM, Abdool Karim S, Kesselheim AS, Outterson K. Whose burden, whose benefit? Revisiting ethical trade-offs in the WHO guidelines on scaling up mass azithromycin administration. PLoS Med. 2025;22(9):e1004736. doi:10.1371/journal.pmed.1004736 ↩