Leveraging social networks for toxicovigilance

J Med Toxicol. 2013 Jun;9(2):184-91. doi: 10.1007/s13181-013-0299-6.


The landscape of drug abuse is shifting. Traditional means of characterizing these changes, such as national surveys or voluntary reporting by frontline clinicians, can miss changes in usage the emergence of novel drugs. Delays in detecting novel drug usage patterns make it difficult to evaluate public policy aimed at altering drug abuse. Increasingly, newer methods to inform frontline providers to recognize symptoms associated with novel drugs or methods of administration are needed. The growth of social networks may address this need. The objective of this manuscript is to introduce tools for using data from social networks to characterize drug abuse. We outline a structured approach to analyze social media in order to capture emerging trends in drug abuse by applying powerful methods from artificial intelligence, computational linguistics, graph theory, and agent-based modeling. First, we describe how to obtain data from social networks such as Twitter using publicly available automated programmatic interfaces. Then, we discuss how to use artificial intelligence techniques to extract content useful for purposes of toxicovigilance. This filtered content can be employed to generate real-time maps of drug usage across geographical regions. Beyond describing the real-time epidemiology of drug abuse, techniques from computational linguistics can uncover ways that drug discussions differ from other online conversations. Next, graph theory can elucidate the structure of networks discussing drug abuse, helping us learn what online interactions promote drug abuse and whether these interactions differ among drugs. Finally, agent-based modeling relates online interactions to psychological archetypes, providing a link between epidemiology and behavior. An analysis of social media discussions about drug abuse patterns with computational linguistics, graph theory, and agent-based modeling permits the real-time monitoring and characterization of trends of drugs of abuse. These tools provide a powerful complement to existing methods of toxicovigilance.

MeSH terms

  • Artificial Intelligence
  • Bayes Theorem
  • Drug Users
  • Drug-Seeking Behavior
  • Humans
  • Illicit Drugs / poisoning
  • Illicit Drugs / toxicity*
  • Linguistics
  • Models, Psychological
  • Pharmacovigilance*
  • Social Media
  • Social Networking*
  • Social Support
  • Substance Abuse Detection / methods
  • Substance-Related Disorders / epidemiology*
  • United States / epidemiology


  • Illicit Drugs