Scientists at the University of Vermont have designed a new computer algorithm that detects previously unknown drug interactions by searching posts and hashtags on the social media website Twitter. A description of the preliminary research appears in the Journal of Biomedical Informatics.

The algorithm is set up to search Tweets for the names of medications, show how they’re connected, and compare them to research indexed in PubMed and Twitter hashtags associated with the terms and studies. The goal of the algorithm is to find drug interactions that are not yet described in published studies or easily searchable in the literature.

RELATED: Twitter Posts, Online Media Highlight Patients’ Ignorance of Opioid-Induced GI Side Effects

Ahmed Abdeen Hamed, PhD, and colleagues first tested the algorithm using two drugs with a known interaction, aspirin and diphenhydramine HCl. The algorithm did detect this interaction, but it also located a link between ibuprofen and medical marijuana with the hashtag “Alzheimer’s.” The team did not find any support for this in research databases, except for one study from 1989 which found no interaction between ibuprofen and marijuana. As it turns out, the discussion was surrounding a brand-new peer-reviewed study suggesting ibuprofen may block or reduce some of the memory-damaging effects of regular marijuana use that have been associated with Alzheimer’s disease development.

This data could be used to confirm these newly discovered interactions in larger clinical trials or other research, Dr. Hamed concluded.

For more information visit