Analysis of Neonatal Sepsis in Kumasi, Ghana Through Paper-Based Medical Records
Reducing infant mortality is Millennium Development Goal 4 in Ghana, though it has remained high and unchanged. Improving care and reducing health complications related to infections and respiratory distress has shown to decrease the prevalence of neonatal deaths and preterm births. The objective of this study is to present data originating from an electronic medical records (EMR) pilot project and communicate the results of an analysis of neonatal sepsis using health information from paper-based medical records to identify characteristics of the neonatal patients at the Komfo Anokye Teaching Hospital in Kumasi, Ghana. Medical records from the Mother Baby Ward (n=198) from 2009-2014 were processed using scanners connected to laptop computers. Health information was manually extracted and verified by two researchers for quality assurance purposes and included sepsis, respiratory distress, cough, difficulty feeding, lethargy, seizures, jaundice, birth history, birth maturity and birth location. Regression analysis revealed a significant association between sepsis and birth location (p=0.0180, 95% CI) as well as sepsis and jaundice (p=0.0446, 95% CI). A descriptive profile of the population revealed that 63.6% of infants comprised of 97 males and 101 females were differentially diagnosed as having sepsis. There were 21 (10%) twins observed. Of the 198 cases included in the analysis, there were 127 (64%) full-term births, 105 (53%) cases reporting respiratory distress and 93 (46%) reporting jaundice. The process of manually scanning and converting paper-based medical records to later use for manual data extraction provides a safe and secure way to evaluate health information related to infant and neonate health.
American Society of Tropical Medicine and Hygiene Annual Meeting (ASTMH)
Sexton, Jessica K., Evans Afriyie-Gyawu, Isaac Chun-Hai Fung, Zachary McGalliard, Daniel F. Linder.
"Analysis of Neonatal Sepsis in Kumasi, Ghana Through Paper-Based Medical Records."
Biostatistics, Epidemiology, and Environmental Health Sciences Faculty Presentations.