2015/10/04 by Rion Brattig Correia, Correia, Rion Brattig, Lang Li +3 · 1 citation
Psychology · Medicine · #Psychedelics and Drug Studies #Schizophrenia research and treatment #Mental Health Research Topics
paper · pdf · doi:10.48550/arxiv.1510.01006
Much recent research aims to identify evidence for Drug-Drug Interactions\n(DDI) and Adverse Drug reactions (ADR) from the biomedical scientific\nliterature. In addition to this "Bibliome", the universe of social media\nprovides a very promising source of large-scale data that can help identify DDI\nand ADR in ways that have not been hitherto possible. Given the large number of\nusers, analysis of social media data may be useful to identify under-reported,\npopulation-level pathology associated with DDI, thus further contributing to\nimprovements in population health. Moreover, tapping into this data allows us\nto infer drug interactions with natural products--including cannabis--which\nconstitute an array of DDI very poorly explored by biomedical research thus\nfar. Our goal is to determine the potential of Instagram for public health\nmonitoring and surveillance for DDI, ADR, and behavioral pathology at large.\nUsing drug, symptom, and natural product dictionaries for identification of the\nvarious types of DDI and ADR evidence, we have collected ~7000 timelines. We\nreport on 1) the development of a monitoring tool to easily observe user-level\ntimelines associated with drug and symptom terms of interest, and 2)\npopulation-level behavior via the analysis of co-occurrence networks computed\nfrom user timelines at three different scales: monthly, weekly, and daily\noccurrences. Analysis of these networks further reveals 3) drug and symptom\ndirect and indirect associations with greater support in user timelines, as\nwell as 4) clusters of symptoms and drugs revealed by the collective behavior\nof the observed population. This demonstrates that Instagram contains much\ndrug- and pathology specific data for public health monitoring of DDI and ADR,\nand that complex network analysis provides an important toolbox to extract\nhealth-related associations and their support from large-scale social media\ndata.\n