the pediatric sepsis data colaboratory - kenya paediatric
TRANSCRIPT
J Mark AnserminoThe University of British Columbia
The Centre International Child Health
BCCH Research Institute
Digital Health Innovation Lab
The Pediatric Sepsis Data CoLaboratory
Disclosure
• LionsGate Technologies – founder
• Medtronic – contract research
• NeuroWave Systems – consultant, contract
research
• Save the Children - consultant
Medical data = 30% total
90% in last 2 yearsGigabyte
Terabyte
Petabyte
Exabyte
Zettabyte
Yottabyte
Precision
Health
● Prediction
● Feedback
● Automation
Alaris TIVA
Propofol pump
NeuroSENSE
iControl-RP
Alaris TIVA Remifentanil pump
iControl-RP
What is the
Pediatric
Sepsis CoLab?
THE PROBLEM
• Worldwide, 7 million people die every year from sepsis
The majority in low and middle income countries
• Data collection for individual treatment decisions and the safety, quality, and efficiency in health facilities of under-resourced countries is challenging.
THE SOLUTION
• Facilitate partnerships to develop and validate data collection and sharing tools for quality improvements of care for sepsis patients in health facilities worldwide.
An international data sharing network for collaboration to address pediatric sepsis mortality and morbidity.
Patient and Facility Data
Develop tools for relevant, accurate, transferrable data
Mobile and Web Applications
Diagnose, advise, & plan using low-cost technologies
Modelling and Analytics
Aggregate data for better prediction and treatment
Advocacy and Prevention
Improved guidelines and practices on local to global scale
SEPSIS COLAB
FOUR AREAS OF FOCUS
A Model for Quality Improvement
1. What are we trying to accomplish?
2. How will we know a change is an
improvement?
3. What changes can we make that will
result in improvements?
Environmental Scan
➢ Hospital Statistics
➢ Service Offered
➢ Operational Flow
➢ Infrastructure
➢ Staffing
➢ ER Equipment
➢ Neonatal Care
➢ Paediatric Medication
➢ Paediatric Equipment
➢ Laboratory
➢ Job Aides
➢ Overall Standards
Smart Discharges to
Improve Sepsis OutcomesMatthew O Wiens, PhD
Senior Researcher, Center for International Child Health
University of British Columbia
Visiting Researcher, Faculty of Medicine
Mbarara University of Science and Technology
Precision Public Health: Smart
Discharges
Admission n=1307
Discharge n=1242
Post-discharge mortality n=61
AUC = 0.82
English et al, 2016. JMIR mHealth uHealth;4(1)e16.
English et al, 2016. JMIR mHealth uHealth;4(1)e16.
Precision Public Health: Smart Discharges
• Education for mother
• Post-discharge
vulnerability
• Health behavior
• Early recognition
• Early health seeking
• Incentives
• New bed net
• ORS
• Soap
The Pocket Doc
Building a mHealth Platform Designing for Health Workers
Predicting admission to hospital (A)
Oxygen saturation
Respiratory rate
Temperature
Cough
Diarrhea
Difficulty breathing
Chest indrawing
Lethargic
Irritable PLoS One. 2015; 11(1): e0147560 https://doi.org/10.1371/journal.pone.0143213
IPSI Model
Age
Temperature
RR
SpO2
Cough
Difficulty breathing
Chest indrawing
Lethargy
Irritability
Diarrhea
Risk Icon Meaning
≥50% Emergency cases
25-49% Priority cases
<25% Non-urgent cases
PLoS One. 2015; 11(1): e0147560 https://doi.org/10.1371/journal.pone.0143213
The Pocket Doc
Summer 2017
User design
workshops and
building
partnerships
Spring 2018 2019Fall 2018
Usability and
phase II data
collection
Summer 2018Winter 2017
Study protocol
and ethics
applications
Fall 2017
First prototype
complete
Data collection
phase I begins
Partner with
additional
hospital sites
App ready for
usability study
Google Impact Challenge
Runners Up
Google Impact Challenge Timeline
http://www.thesensorproject.org/
Infection is the leading cause of child mortality worldwide
WHO-MCEE estimates for child causes of death, 2000-2015. (https://www.who.int/healthinfo/global_burden_disease/estimates_child_cod_2000_2015/en/)
•More than 5 million children
• Largely due to infection
• Malaria
• Measles
• Pneumonia
• Sepsis