Expanding antibiotic treatment in sub-Saharan Africa could save children’s lives

“We were very surprised that we were able to prove that spillover effect, that treating the older children actually helped the younger children,” says epidemiologist Thomas Lietman. “It seems to be the majority of [antibiotic benefit] it is indirect rather than direct.”

In 2018, a large trial in Niger, Malawi and Tanzania showed that a single dose of azithromycin twice a year for children under 5 could reduce mortality by almost 14 percent, or from about 165 annual deaths per 10,000 children in about 145. led the World Health Organization, or WHO, to the current recommendation, but they did not suggest that it be given to all children under the age of 5.

WHO probably limited their recommendation to infants because that group has higher mortality rates than preschoolers, says Lietman, of the University of California, San Francisco. But he and his team suspected that the babies wouldn’t fare as well if the older children weren’t also treated.

Working with a team of health care workers in Niger, the researchers conducted a follow-up trial from 2020 to 2023 with more than 380,000 children under the age of 5. Participants were assigned to one of three groups: infants 1 to 11 months receiving treatment, with older children receiving a placebo; all children under 5 receive treatment; or all children under 5 receiving a placebo. The results, published on August 21 at New England Journal of Medicineshow that treating all children under 5 reduced infant mortality by 17 percent, from approximately 220 annual deaths per 10,000 children to 185.

Lietman and study co-author Kieran O’Brien, an epidemiologist also at UC San Francisco, spoke to Scientific news about recent findings and their implications. This interview has been edited for length and clarity.

SN: What inspired the first test in 2018?

Lietman: It started with trachoma studies [in the early 2000s]. Trachoma is an eye disease caused by chlamydia [that] causes a lot of blindness (SN: 20/2/08). And the WHO has recommended mass treatment with azithromycin for all communities – not just for preschool children, but for all communities – once a year [to treat trachoma]. In one of the trachoma trials in Ethiopia, we discovered [that communities with widespread antibiotic treatment had lower childhood mortality rates than those that did not. This was surprising because it wasn’t known that widespread antibiotics could reduce childhood mortality until this trial.]

There is much more child mortality in West Africa (for example, Niger) than in East Africa (for example, Ethiopia). So the Gates Foundation decided to fund the MORDOR trial, which is the 2018 study. It found a 13.5 percent reduction in childhood mortality in communities where they were randomized to give azithromycin twice a year to preschool children.

SN: After the MORDOR trial, how did you feel about the WHO recommendation?

Lietman: The WHO actually made the guidelines quite quickly, I was very impressed. I’ve never seen anything like this happen so quickly. But they are very concerned about the balance between antibiotic resistance and childhood mortality (SN: 24.1.22). So they recommended limiting antibiotics to children 1 to 11 months rather than 1 to 59 months, which we would [studied] in MORDOR.

O’Brien: I think there was a general frustration with the release of the guidelines and seeing that limited age group. And part of the sentiment behind that was that there are far fewer children in the 1 to 11 month age group than the 1 to 59 month age group. And so, you end up saving far fewer lives if you only target that small age group.

SN: In this new trial, called AVENIR, treatment also worked for infants in the older children group NO treated?

O’Brien: We saw a 6 percent reduction in mortality in the 1- to 11-month group when they were treated alone [compared with 17 percent when all children received antibiotics]. But it was not statistically significant. Unfortunately, we [didn’t have enough participants] to be able to detect an effect as small as 6 percent.

We think that these mass drug administration interventions work in part through direct effects on the children taking the drug. But then also through these indirect effects, with the community-wide reduction of disease transmission. Younger children are likely to benefit from older children who are active in the community having this transmission reduced. So, [when the older kids also receive treatment]they are not being infected as much by the older children.

Lietman: We come from the world of trachoma, where perhaps half the children are infected with chlamydia [in their eyes] when you start your treatment program. And if you treat just one child, you’ll probably clear their chlamydia, but they’ll just get reinfected within a few weeks. So it’s actually irrelevant; you have to treat all the other children. In fact, it is more important that other children are treated than that you are treated. So we expected this indirect effect.

SN: What do you hope will change, given your findings?

O’Brien: I guess we will hope to see [the guidelines] updated to recommend treatment for children 1 to 59 months of age for whatever conditions made sense at the time the guidelines were created. Current guidelines already state that any implementation should be accompanied by resistance monitoring, and so I would fully expect this to be in place moving forward.

[Some] of the next questions that I think many researchers have focused on [are]: How do we determine when to stop? How much resistance is too much? What kind of impact does resistance have that would cause us to stop treatment? And also, how long do you need to be treated in order to see a sustained reduction in mortality? These are some of the questions that can contribute to better defining some thresholds around intervention.

Big picture, though, I don’t think anyone envisions it being a very long-term intervention for any regular area. [The hope is that the treatment could reduce transmission enough to eliminate a future need for mass distribution of antibiotics.]