Methods in Authenticating Participants in Fully-online Youth Mobile Application Trials: Lessons from the iREACH Project (Preprint)

BACKGROUND Mobile health (mHealth) applications are important interventions to increase the scale and reach of prevention services, including HIV testing and prevention counseling, pre-exposure prophylaxis (PrEP), condom distribution, and education, all of which are required to decrease incidence rates. The use of these apps online as well as fully online intervention trials can be challenged by the need to remove fraudulent or duplicative entries and authenticate unique trial participants prior to randomization to protect the integrity of the sample and trial results. It is critical to make sure data collected through this modality are valid and reliable. OBJECTIVE This manuscript discusses the electronic and manual authentication strategies for the iREACH randomized control trial (RCT) that were used to monitor and prevent fraudulent enrollment. METHODS iREACH is a RCT focused on same-sex-attracted, cis-gender males aged 13-18 years in the US with a focus on enrolling racial/ethnic minorities and those in rural communities. Evaluation of the data was done by identifying possible duplications in enrollment, identifying potentially fraudulent or ineligible participants through inconsistencies in registration and survey data, and by reviewing baseline completion times to avoid enrolling bots and those who did not complete the baseline survey. Electronic systems flagged questionable enrollment. Additional manual reviews included verification of age, IP addresses, email addresses, social media accounts, and completion times for surveys. RESULTS The strategies used identified and averted 624 cases of potential fraudulent, duplicative or ineligible enrollment using electronic and manual strategies including the integration of social media profiles. Seventy-nine percent of the potentially fraudulent or ineligible cases were identified through electronic strategies, thereby reducing the burden of manual authentication for the majority of cases. A case study is included with scenario, resolution, and authentication strategy response. CONCLUSIONS As online trials are more common, methods to handle suspicious enrollment that compromise data quality are increasingly important for inclusion in protocols. INTERNATIONAL REGISTERED REPORT RR2-10.2196/10174