Date of Award

January 2021

Document Type


Degree Name

Master of Public Health (MPH)


School of Public Health

First Advisor

Melinda M. Pettigrew


Background: Antibiotic-associated diarrhea (AAD), defined as three or more loose stools in a day since starting antibiotics, can last up to 8 weeks. It is the most common side-effect of antibiotic use and a significant contributor to antibiotic therapy nonadherence. Pediatric AAD is associated with gastrointestinal dysbiosis. Few studies have compared temporal microbiome dynamics in children who do and do not develop AAD. To describe microbiome characteristics associated with AAD, we compared temporal dynamics of the gastrointestinal microbiome among children treated with beta-lactam antibiotics for community-acquired pneumonia.

Methods: Stool samples were collected from children, ages 6-71 months, enrolled in the Short-Course Outpatient Therapy for Community-Acquired Pneumonia (SCOUT-CAP) trial. Children were classified as experiencing AAD if they had one or more days of diarrhea during the study. Samples were collected at enrollment (study day 1), Outcome Assessment Visit 1 (day 6-10), and at the end of the study Outcome Assessment Visit 2 (day 19-25). Samples were analyzed using 16s-rRNA gene and shotgun metagenomic sequencing. We evaluated associations between patient characteristics, antibiotic treatment duration, microbiome characteristics (e.g., diversity, composition, and antibiotic resistance gene abundance), and AAD (yes/no).

Results: Among 67 children evaluated, 19 (28%) developed AAD. We observed significant taxonomic compositional differences between the two groups (PERMANOVA, P < 0.03) and across visits (P < 0.001). Children with higher relative abundances of Bacteroides and Bifidobacterium at enrollment were less likely to experience AAD and, while high Blautia proportions were more likely to experience AAD. Analyses of Outcome Assessment Visit 2 samples showed that children in the AAD group experienced prolonged dysbiosis and had significantly higher levels of aminoglycoside and multidrug efflux pump antibiotic resistance genes (P < 0.05 for each).

Conclusions: The gastrointestinal microbiome differed in children with and without AAD. The identified potential protective taxa suggest new targeted preventive approach against AAD.


This thesis is restricted to Yale network users only. This thesis is permanently embargoed from public release.