1. In a randomized crossover trial, use of a smartphone-based tool that predicts the etiology of acute diarrhea in children did not significantly reduce antibiotic prescription rates.
2. Among more than 900 children treated for diarrhea at multiple centers in Bangladesh and Mali, about 70% were prescribed antibiotics.
Evidence Rating Level: 1 (Excellent)
Study Rundown: Acute diarrhea in children is often inappropriately treated with antibiotics, which are indicated for some causes of bacterial enterocolitis and suspected cholera but not in most other cases. This study aimed to study the use of a previously studied algorithm, which takes inputs such as patient history, symptoms, and geographic location and creates a prediction of diarrheal etiology. The algorithm was implemented as a smartphone tool for physicians in two low- and middle-income countries, Bangladesh and Mali. Thirty physicians were randomized to use the tool or practice as usual for 4 weeks, then the arms crossed over for another 4 weeks. Among more than 900 patients treated for diarrhea, 69.8% of children in the intervention group were prescribed antibiotics compared to 76.5% in the control group. There was no statistically significant difference in the rate of antibiotic prescription between groups. An additional analysis accounting for the algorithm-predicted probability of viral diarrhea showed a significant, 5.6% decrease in antibiotic prescriptions with use of the smartphone tool. Interpretation of this finding is limited by the fact that the analysis was post hoc. This study successfully implemented a smartphone-based prediction algorithm into international, low-resource care settings, but the utility of the tool remains in question. Decreasing unnecessary antibiotic prescriptions for pediatric diarrhea is certainly a worthy goal, especially given the high rate of antibiotic prescriptions in both study groups.
Click to read the study in JAMA Pediatrics
Relevant Reading: Antibiotic treatment of acute gastroenteritis in children
In-Depth [randomized controlled trial]: Patients aged 2-5 were drawn from three hospitals in Bangladesh and four centers in Mali. Children with infectious comorbidities and severe malnutrition were excluded. Patients whose families did not have access to a cell phone for follow-up were also excluded. Only 19 children had bloody diarrhea. A one-week washout period was observed between the two 4-week study periods. The primary analysis was a linear mixed-effects model including the intervention, period, and their interaction along with a random effect reflecting site and physician. The post hoc analysis added an interaction term for study arm and algorithm-predicted probability of viral-only diarrheal etiology. The risk difference based on this analysis was -0.056, with a 95% confidence interval of -0.128 to -0.010. Physicians who were given a 10% higher predicted probability of viral diarrhea by the algorithm were 14% less likely to prescribe antibiotics.
Image: PD
©2022 2 Minute Medicine, Inc. All rights reserved. No works may be reproduced without expressed written consent from 2 Minute Medicine, Inc. Inquire about licensing here. No article should be construed as medical advice and is not intended as such by the authors or by 2 Minute Medicine, Inc.
