Josie J Sivaraman1,2, Scott K Proescholdbell3, David Ezzell4, Meghan E Shanahan2,5. 1. 2331 Department of Epidemiology, University of North Carolina at Chapel Hill, NC, USA. 2. Injury Prevention Research Center, University of North Carolina at Chapel Hill, NC, USA. 3. Epidemiology, Surveillance and Informatics Unit, Injury and Violence Prevention Branch, Chronic Disease and Injury Section, Division of Public Health, North Carolina Department of Health and Human Services, Raleigh, NC, USA. 4. 6800 Division of Health Service Regulation, Office of Emergency Medical Services, North Carolina Department of Health and Human Services, Raleigh, NC, USA. 5. Department of Maternal and Child Health, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA.
Abstract
OBJECTIVES: Tracking nonfatal overdoses in the escalating opioid overdose epidemic is important but challenging. The objective of this study was to create an innovative case definition of opioid overdose in North Carolina emergency medical services (EMS) data, with flexible methodology for application to other states' data. METHODS: This study used de-identified North Carolina EMS encounter data from 2010-2015 for patients aged >12 years to develop a case definition of opioid overdose using an expert knowledge, rule-based algorithm reflecting whether key variables identified drug use/poisoning or overdose or whether the patient received naloxone. We text mined EMS narratives and applied a machine-learning classification tree model to the text to predict cases of opioid overdose. We trained models on the basis of whether the chief concern identified opioid overdose. RESULTS: Using a random sample from the data, we found the positive predictive value of this case definition to be 90.0%, as compared with 82.7% using a previously published case definition. Using our case definition, the number of unresponsive opioid overdoses increased from 3412 in 2010 to 7194 in 2015. The corresponding monthly rate increased by a factor of 1.7 from January 2010 (3.0 per 1000 encounters; n = 261 encounters) to December 2015 (5.1 per 1000 encounters; n = 622 encounters). Among EMS responses for unresponsive opioid overdose, the prevalence of naloxone use was 83%. CONCLUSIONS: This study demonstrates the potential for using machine learning in combination with a more traditional substantive knowledge algorithm-based approach to create a case definition for opioid overdose in EMS data.
OBJECTIVES: Tracking nonfatal overdoses in the escalating opioid overdose epidemic is important but challenging. The objective of this study was to create an innovative case definition of opioid overdose in North Carolina emergency medical services (EMS) data, with flexible methodology for application to other states' data. METHODS: This study used de-identified North Carolina EMS encounter data from 2010-2015 for patients aged >12 years to develop a case definition of opioid overdose using an expert knowledge, rule-based algorithm reflecting whether key variables identified drug use/poisoning or overdose or whether the patient received naloxone. We text mined EMS narratives and applied a machine-learning classification tree model to the text to predict cases of opioid overdose. We trained models on the basis of whether the chief concern identified opioid overdose. RESULTS: Using a random sample from the data, we found the positive predictive value of this case definition to be 90.0%, as compared with 82.7% using a previously published case definition. Using our case definition, the number of unresponsive opioid overdoses increased from 3412 in 2010 to 7194 in 2015. The corresponding monthly rate increased by a factor of 1.7 from January 2010 (3.0 per 1000 encounters; n = 261 encounters) to December 2015 (5.1 per 1000 encounters; n = 622 encounters). Among EMS responses for unresponsive opioid overdose, the prevalence of naloxone use was 83%. CONCLUSIONS: This study demonstrates the potential for using machine learning in combination with a more traditional substantive knowledge algorithm-based approach to create a case definition for opioid overdose in EMS data.
Entities:
Keywords:
case definition; machine learning; natural language processing; opioid overdose; surveillance
Authors: Eric J Lavonas; Ian R Drennan; Andrea Gabrielli; Alan C Heffner; Christopher O Hoyte; Aaron M Orkin; Kelly N Sawyer; Michael W Donnino Journal: Circulation Date: 2015-11-03 Impact factor: 29.690
Authors: Avik Chatterjee; Marc R Larochelle; Ziming Xuan; Na Wang; Dana Bernson; Michael Silverstein; Scott E Hadland; Thomas Land; Jeffrey H Samet; Alexander Y Walley; Sarah M Bagley Journal: Drug Alcohol Depend Date: 2018-10-25 Impact factor: 4.492
Authors: Carl A Latkin; Rachel E Gicquelais; Catie Clyde; Lauren Dayton; Melissa Davey-Rothwell; Danielle German; Seun Falade-Nwulia; Haneefa Saleem; Michael Fingerhood; Karin Tobin Journal: Int J Drug Policy Date: 2019-04-23
Authors: Jennifer M Boggs; Arne Beck; Debra P Ritzwoller; Catherine Battaglia; Heather D Anderson; Richard C Lindrooth Journal: J Gen Intern Med Date: 2020-02-10 Impact factor: 5.128
Authors: Rebecca E Cash; Jeremiah Kinsman; Remle P Crowe; Madison K Rivard; Mark Faul; Ashish R Panchal Journal: MMWR Morb Mortal Wkly Rep Date: 2018-08-10 Impact factor: 17.586
Authors: Brian Hazlehurst; Carla A Green; Nancy A Perrin; John Brandes; David S Carrell; Andrew Baer; Angela DeVeaugh-Geiss; Paul M Coplan Journal: Pharmacoepidemiol Drug Saf Date: 2019-06-19 Impact factor: 2.890