Shannon Hughes*, Florida International University, Robert Stempel College of Public Health and Social Work, Miami, United States Track: Research Presentation Topic: Health information on the web: Supply and Demand Presentation Type: Oral presentation Submission Type: Single Presentation Building: MECC Room: Auditorium 2 Date: 2010-11-29 04:45 PM – 06:15 PM Last modified: 2010-09-22

If you are the presenter of this abstract (or if you cite this abstract in a talk or on a poster), please show the QR code in your slide or poster (QR code contains this URL).

Background: A growing minority of online health seekers report using consumer-generated content, such as discussion forum posts and consumer reviews, to find or share information about a disease or treatment. However, the consistency and quality of consumer commentary for a given treatment across various health sites is unknown. Implications arise for informed treatment decision-making, and for increasing the “usability” of inherently dispersed online content.

Objective: This study sought to examine differences in consumer reporting of drugs’ effects across selected websites, and to assess the comparability of online consumer reports in this sample to the published drug literature.

Methods: A grounded theory textual analysis was completed on 960 randomly selected online consumer reviews from 4 websites (2 expert-run: Webmd, RevolutionHealth, and 2 consumer-run: AskAPatient and CrazyMeds) for a popularly prescribed antipsychotic (quetiapine, n=480) and antidepressant (escitalopram, n=480) medication. Consumer-reported effects for each drug were coded using QDA Miner 3.2 software by Provalis Research. Chi-square was used to test for association between reported effects and the website the report originated.

Results: The most frequently mentioned effects by consumers taking either drug were related to symptom improvement or worsening, and changes in sleep. Significantly more consumers on consumer-run websites reported quetiapine worsened their symptoms (17.2% vs 5.0%, p=.000), while more consumers posting reviews on expert-run health sites reported it improved their symptoms (32.7% vs 21.4%, p=.008). This trend was repeated for escitalopram. Consumer reviews on expert-run health sites also reported lower rates for most side effects. For example, 9.5% of consumers posting on expert-run sites compared to 18.6% of those on consumer-run sites reported weight gain on escitalopram (p=0.006). Additionally, for escitalopram, the primary side effects reported by all consumers in this study (effects on sleep, weight, and sexual function) appear congruent with findings from other consumer surveys, including data from iGuard.org and Consumer Reports. Over 60% of consumers in this study reported changes in sleep with quetiapine, which is congruent with studies suggesting its real-world use as a sleep-aid more often than an “antipsychotic.”

Conclusions: Consumer reviews on expert-run health sites were different from those on consumer-run sites in important ways, and several hypotheses may explain these differences. Unless the online health searcher actively seeks a variety of sources to retrieve consumer reviews, these differences in consumers’ reporting of drug effects across websites could potentially and unknowingly hinder informed decision-making. Patient communities, such as patientslikeme.com and iGaurd.org, collect and report back in summary form the data from its users, however, these sites may represent a more pro-active and possibly informed group of healthcare consumers. Achieving a balanced and representative snapshot of consumer drug experiences may require development of a “meta” website, similar to certain sites for travel and consumer electronics, which integrates and organizes the mass of dispersed online consumer reviews. Such a “meta” website would prove useful for individual drug decision-making, as well as hold potential for active post-marketing safety surveillance.