Social Media Mining for Toxicovigilance: Automatic Monitoring of Prescription Medication Abuse from Twitter

• DOI: 10.1007/s40264-015-0379-4
• Publication Date: 2016-01-09T14:56:55Z
• Authors: Abeed Sarker; Karen O’Connor; Rachel Ginn; Matthew Scotch; Karen Smith; Dan Malone; Graciela Gonzalez
• Keywords: Prescription Drug Misuse

Prescription medication overdose is the fastest growing drug-related problem in USA. The nature of this necessitates implementation improved monitoring strategies for investigating prevalence and patterns abuse specific medications.Our primary aims were to assess possibility utilizing social media as a resource automatic prescription devise an classification technique that can identify potentially abuse-indicating user posts.We collected Twitter posts (tweets) associated with three commonly abused medications (Adderall(®), oxycodone, quetiapine). We manually annotated 6400 tweets mentioning these control (metformin) not subject due its mechanism action. performed quantitative qualitative analyses data determine whether on contain signals abuse. Finally, we designed supervised distinguish containing from those do assessed utility over time.Our show clear be drawn percentage are significantly higher case (Adderall(®): 23 %, quetiapine: 5.0 oxycodone: 12 %) than proportion (metformin: 0.3 %). Our approach achieves 82 % accuracy overall (medication class recall: 0.51, precision: 0.41, F measure: 0.46). To illustrate classification, how used analyze study indicates crucial obtaining abuse-related information medications, approaches involving natural language processing hold promises essential future intervention tasks.