A Randomized Trial of Social Media Versus Search Engine Advertising to Increase Awareness of Treatments for Female Stress Urinary Incontinence.

Importance: Outreach based on user characteristics through social media advertising is significantly more effective than outreach based on user interests with search engine advertising for female stress urinary incontinence.

Objective: The aim of the study was to compare the effectiveness of 2 online outreach platforms to engage women at risk of stress urinary incontinence with a high-quality patient education website.

Study Design: A geographic randomized trial was performed comparing online advertising on Facebook, a social media site, versus Google Ads, a search engine, to direct viewers to VoicesForPFD.org. Advertisements were developed using health communication theory and pretested before the randomized trial. Counties in the Pacific Northwest were grouped into approximately equal populations, called "supercounties," and randomized to advertisement platforms. Educational website views resulting from each campaign, called sessions, were measured using Google Analytics; the primary outcome was the number of sessions per supercounty. County characteristics were obtained from national survey data. Effectiveness of the advertising platforms was compared between randomization arms using Welch's unequal variances t test, and crude and adjusted linear regression models.

Results: Mean user sessions were significantly higher in areas randomized to Facebook advertising (470 sessions) compared with areas randomized to Google Ads advertising [44 sessions; mean, 426 session difference (95% confidence interval, 199-653 sessions); P = 0.001]. After accounting for adult female population and rurality, randomization to Facebook instead of Google Ads still resulted in 409 additional sessions (95% confidence interval, 317-501 sessions; P < 0.001).

Conclusions: Social media was 10 times more effective (470 versus 44 mean sessions per supercounty) than search engine outreach advertising and should be prioritized in outreach efforts. Future work is needed on engagement and behavior change with online outreach.