Factors Associated With Perceived Trust of False Abortion Websites: Cross-sectional Online Survey

BACKGROUND: Most patients use the internet to search for health information. While there is a vast repository of searchable information online, much of the content is unregulated and therefore potentially incorrect, conflicting, or confusing. Abortion information online is particularly prone to being inaccurate as antichoice websites publish purposefully misleading information in formats that appear as neutral resources. To understand how antichoice websites appear neutral, we need to understand the specific website features of antichoice websites that impart an impression of trustworthiness. OBJECTIVE: We sought to identify the characteristics of false or misleading abortion websites that make these websites appear trustworthy to the public. METHODS: We conducted a cross-sectional study using Amazon’s Mechanical Turk platform. We used validated questionnaires to ask participants to rate 11 antichoice websites and one neutral website identified by experts, focusing on website content, creators, and design. We collected sociodemographic data and participant views on abortion. We used a composite measure of “mean overall trust” as our primary outcome. Using correlation matrices, we determined which website characteristics were most associated with mean overall trust. Finally, we used linear regression to identify participant characteristics associated with overall trust. RESULTS: Our analytic sample included 498 participants aged from 22 to 70 years, and 50.1% (247/493) identified as female. Across 11 antichoice websites, creator confidence (“I believe that the creators of this website are honest and trustworthy”) had the highest correlation coefficient (strongest relationship) with mean overall trust (coefficient=0.70). Professional appearance (coefficient=0.59), look and feel (coefficient=0.59), perception that the information is created by experts (coefficient=0.59), association with a trustworthy organization (coefficient=0.58), valued features and functionalities (coefficient=0.54), and interactive capabilities (coefficient=0.52) all demonstrated strong relationships with mean overall trust. At the individual level, prochoice leaning was associated with higher overall trust of the neutral website (B=−0.43, 95% CI −0.87 to 0.01) and lower mean overall trust of the antichoice websites (B=0.52, 95% CI 0.05 to 0.99). CONCLUSIONS: The mean overall trust of antichoice websites is most associated with design characteristics and perceived trustworthiness of website creators. Those who believe that access to abortion should be limited are more likely to have higher mean overall trust for antichoice websites.