Molly Klarman, MPH
Research Coordinator
University of Florida, United States
Disclosure information not submitted.
Youseline Cajusma, RN
Head Nurse
MotoMeds (UF/Chrsitianville), United States
Disclosure information not submitted.
Lerby Exantus, MD
Ultrasound Master Trainer
The NYAGI-Caris Ultrasound Project, United States
Disclosure information not submitted.
Valery Madsen Beau De Rochars, MPH, MD
Assistant Professor of Health Services Research Management and Policy
University of Florida, United States
Disclosure information not submitted.
Chantale Baril, MD
Neonatologist
State University of Haiti, United States
Disclosure information not submitted.
Torben Becker, MD, PhD
Chief, Division of Critical Care Medicine
Shands Hospital - University of Florida
Gainesville, Florida
Disclosure information not submitted.
Title: A Pediatric Pre-Emergency Service for Resource-Limited Settings: A Pre-Pilot Study in Haiti
Introduction: Acute respiratory infection and diarrheal disease are the two leading causes of pediatric mortality between 1 month and 5 years of age globally. Low-cost treatment for these infectious diseases and other common childhood illnesses are well established. However, treatment is most effective when administered early, which is a challenge in resource-limited settings. MotoMeds, a pediatric telemedicine and medication delivery service, was launched in Gressier, Haiti as a pre-pilot feasibility study to address this challenge.
Methods: Parents of an acutely sick child call MotoMeds to connect with a nurse who uses clinical decision-support tools. Our team adapted these guidelines from WHO in-person guidelines. The call center nurse synthesizes information relayed over the phone to triage, assess and implement a guideline-specific treatment plan, which may include medication delivery. To evaluate the MotoMeds model, a clinical congruence study was conducted where a nurse accompanied each delivery to perform a parallel in-person exam. The primary outcome measure was congruence of illness severity between call center and household assessments. Cohen's Kappa statistic and sensitivity and specificity measurements were calculated.
Results: From September 2019 to January 2021 391 cases were enrolled. The most common chief complaints were fever (42%), respiratory illness (17%), and skin infection/irritation (14%). At the call center, 6.1% (24) of patients were traiged as severe and sent to the hospital and 89% (347) received a nurse visit. At the household visit, danger signs were identified in 1.5% (6) of patients. Agreement between mild cases was substantial: Kappa=0.66 (95%CI 0.53-0.78) with 95% (95%CI 92%-97%) sensitivity and 72% (95%CI 56%-85%) specificity. Agreement between moderate cases was similarly substantial: Kappa=0.64 (95%CI 0.50-0.77) with 95% (95%CI 91%-97%) sensitivity and 76% (95%CI 59%-89%) specificity.
Conclusions: These data indicate call-center assessments hold clinical value. Future analysis will evaluate congruence of all assessment components and explore rates of conversion to severe illness and seeking in-person care.