ASPPH logo


Member Research & Reports

Member Research & Reports

Kentucky: Machine Learning Provides Faster Drug Overdose Mortality Surveillance

Timely data is key to effective public health responses to epidemics. Drug overdose deaths are identified in surveillance systems through ICD-10 codes present on death certificates. ICD-10 coding takes time, but free-text information is available on death certificates prior to ICD-10 coding. Investigators at the University of Kentucky College of Public Health are working to develop a machine learning method to classify free-text death certificates as drug overdoses to provide faster drug overdose mortality surveillance. A new paper in PLoS One details their work.

Dr. Patrick J. Ward, student in epidemiology and biostatistics and an epidemiologist at the Kentucky Injury Prevention and Research Center, is first author of “ Enhancing timeliness of drug overdose mortality surveillance: A machine learning approach”. Co-authors are Mr. Peter J. Rock, doctoral student in epidemiology and biostatistics, associate professor of biostatistics Dr. Svetla Slavova, associate professor of epidemiology Dr. April M. Young,  professor of preventive medicine and environmental health Dr. Terry L. Bunn, and associate professor of bioinformatics Dr. Ramakanth Kavuluru.

Machine learning using natural language processing is a relatively new approach in the context of surveillance of health conditions. This method presents an accessible application of machine learning that improves the timeliness of drug overdose mortality surveillance. As such, it can be employed to inform public health responses to the drug overdose epidemic in near-real time as opposed to several weeks following events.

Read more.

Tags: ,