Related Experiment Video
Updated: Sep 17, 2025

Quantifying Vibrio cholerae Colonization and Diarrhea in the Adult Zebrafish Model
Published on: July 12, 2018
Personalized azithromycin treatment rules for children with watery diarrhea using machine learning
Sara S Kim1, Allison Codi2, James A Platts-Mills3
1Department of Epidemiology, Rollins School of Public Health, Emory University, Atlanta, GA, USA. sara.kim2@emory.edu.
Insights
Machine learning identifies children most likely to benefit from azithromycin for watery diarrhea. Personalized rules target treatment, reducing diarrhea risk and improving outcomes, potentially without pathogen diagnostics.
Area of Science:
- Computational biology and bioinformatics
- Pediatric infectious diseases
- Clinical trial data analysis
Background:
- Azithromycin is used to treat watery diarrhea in children, but optimal targeting strategies are needed.
- Personalized medicine approaches can improve treatment efficacy by identifying subgroups most likely to benefit.
- Machine learning offers powerful tools for analyzing complex clinical data to develop predictive treatment rules.
Purpose of the Study:
- To develop and validate personalized treatment rules for azithromycin in children with watery diarrhea using machine learning.
- To identify child-specific characteristics that predict benefit from azithromycin treatment.
- To assess the impact of targeted azithromycin treatment on diarrhea duration, re-hospitalization, and mortality.
Main Methods:
- Utilized data from a randomized trial (NCT03130114) involving 6692 children with watery diarrhea.
- Employed an ensemble machine learning procedure to estimate child-level expected treatment benefit based on diagnostic, child, and clinical characteristics.
- Developed personalized treatment rules and estimated the proportion of children treated and average treatment benefits.
Main Results:
- Personalized treatment rules recommended azithromycin for approximately one-third of children.
- Targeted treatment reduced the risk of diarrhea on day 3 by 10.1% (NNT: 10) and day 90 re-hospitalization/death by 2.4% (NNT: 42).
- Host characteristics, rather than pathogen diagnostics, were stronger predictors for benefits related to re-hospitalization and death.
Conclusions:
- Machine learning can effectively identify children who will benefit most from azithromycin for watery diarrhea.
- Personalized targeting of azithromycin can significantly improve clinical outcomes and reduce healthcare utilization.
- Effective targeting for severe outcomes may be achievable without relying on pathogen diagnostics, simplifying treatment strategies.
Abstract:
We use machine learning to identify innovative strategies to target azithromycin to the children with watery diarrhea who are most likely to benefit. Using data from a randomized trial of azithromycin for watery diarrhea (NCT03130114), we develop personalized treatment rules given sets of diagnostic, child, and clinical characteristics, employing a robust ensemble machine learning-based procedure. This procedure estimates the child-level expected benefit for a given set of covariates by combining predictions from a library of statistical models. For each rule, we estimate the proportion treated under the rule and the average benefits of treatment. Among 6692 children, treatment under the most comprehensive rule is recommended on average for one third of children. The risk of diarrhea on day 3 is 10.1% lower (95% CI: 5.4, 14.9) with azithromycin compared to placebo among children recommended for treatment (NNT: 10). For day 90 re-hospitalization and death, risk is 2.4% lower (95% CI: 0.6, 4.1; NNT: 42). While pathogen diagnostics are strong determinants of azithromycin effects on diarrhea duration, host characteristics may better predict benefits for re-hospitalization or death. This suggests that targeting antibiotic treatment for severe outcomes among children with watery diarrhea may be possible without access to pathogen diagnostics.
More Related Videos
10:50Design and Use of a Low Cost, Automated Morbidostat for Adaptive Evolution of Bacteria Under Antibiotic Drug Selection
Published on: September 27, 2016
08:33Quantitative Polymerase Chain Reaction-based Analyses of Murine Intestinal Microbiota After Oral Antibiotic Treatment
Published on: November 17, 2018
Related Concept Videos
Drugs Affecting GI Tract Motility: Antimicrobials as Antidiarrheal Agents
Drugs for Treatment of Diarrhea-Predominant IBS
Two specific drugs used in the treatment are alosetron (Lotronex) and eluxadoline (Viberzi). Alosetron, a 5-HT3 antagonist, works by slowing the movement of stools in the gut, reducing bowel...
Drugs Affecting GI Tract Motility: Adsorbents as Antidiarrheal Agents
Adsorbents...