Sign in Subscribe
Machine learning

Personalized Azithromycin Treatment Rules for Children With Watery Diarrhea

Strategic Insights

— 1 min read

Researchers at Emory University have developed a machine learning-based procedure to identify innovative strategies to target azithromycin treatment to children with watery diarrhea who are most likely to benefit. This approach estimates the child-level expected benefit for a given set of covariates by combining predictions from a library of statistical models. The research, funded by the Bill and Melinda Gates Foundation, CGIAR, and the NIH National Institute of Allergy & Infectious Diseases (NIAID), aims to reduce the risk of diarrhea and re-hospitalization or death among children with watery diarrhea.

Key Takeaways:

  • The researchers used data from a randomized trial of azithromycin for watery diarrhea (NCT03130114) to develop personalized treatment rules based on diagnostic, child, and clinical characteristics.
  • The machine learning-based procedure estimated the proportion of children treated under each rule and the average benefits of treatment, with treatment recommended for one third of children on average.
  • Among 6692 children, treatment with azithromycin compared to placebo resulted in a 10.1% lower risk of diarrhea on day 3 (95% CI: 5.4, 14.9; NNT: 10).
  • For day 90 re-hospitalization and death, the risk was 2.4% lower (95% CI: 0.6, 4.1; NNT: 42) among children recommended for treatment.
  • Pathogen diagnostics were found to be strong determinants of azithromycin effects on diarrhea duration, but host characteristics may better predict benefits for re-hospitalization or death.
  • The study suggests that targeting antibiotic treatment for severe outcomes among children with watery diarrhea may be possible without access to pathogen diagnostics.

Statistics:

  • 6692 children were included in the analyses.
  • 1/3 of children were recommended for treatment under the most comprehensive rule.
  • Azithromycin treatment reduced the risk of diarrhea on day 3 by 10.1% (95% CI: 5.4, 14.9).
  • The NNT (number needed to treat) for azithromycin treatment was 10.
  • Azithromycin treatment reduced the risk of re-hospitalization or death on day 90 by 2.4% (95% CI: 0.6, 4.1).
  • The NNT for azithromycin treatment was 42.

Sources:

  • NewsRx. Researchers from Emory University Report Details of New Studies and Findings in the Area of Personalized Medicine (Personalized Azithromycin Treatment Rules for Children With Watery Diarrhea Using Machine Learning). Health & Medicine Week. August 8, 2025; p 514.
  • Personalized Azithromycin Treatment Rules for Children With Watery Diarrhea Using Machine Learning. Nature Communications, 2025;16(1). Nature Communications can be contacted at: Nature Portfolio, Heidelberger Platz 3, Berlin, 14197, Germany.

Read more