A rapid, Web-based method for obtaining patient views on effects and side-effects of antidepressants

This project was undertaken to develop a rapid method for obtaining a widespread sample of patient views on the efficacy and side-effects of antidepressants.

A Web-based method is described for rapidly and objectively obtaining patient views on the effects and side-effects of treatment with antidepressants. The method entails a systematized search of many URLs (Uniform Resource Locators, or Web page addresses), using keywords and phrases to extract the named drug and symptom that are reliably relevant to the medication being taken by the individual reporting the experience online. Unwanted medical conditions (e.g., cancer) are excluded.

Three successive searches of thousands of Web pages revealed a cumulative total of 835 "mentions" of patient experience on duloxetine, 756 for venlafaxine, 637 for citalopram, 636 for sertraline, 559 for paroxetine, 457 for fluoxetine, 318 for desvenlafaxine, 289 for fluvoxamine, and 210 for mirtazapine, in association with various symptoms. A comparison of the associated symptoms for each of the antidepressants found that the prevalence of the combined factor of fatigue, drowsiness, tiredness or lethargy ranged from 6.4±0.8% down to 2.9±0.15% of the mentions, where the S.E. was derived from three repeats of the Web-based analysis. The prevalence of dizziness or vertigo ranged from 7.6±0.8% down to 1.9±0.3% of the mentions.

Given the increasing number of patient narratives about drug experiences on open-access Web forums, this rapid novel method will have increasing utility in post-marketing surveillance and in comparing the effects of psychiatric medications.