Currently submitted to: Journal of Medical Internet Research
Date Submitted: Nov 20, 2025
(closed for review but you can still tweet)
Warning: This is an author submission that is not peer-reviewed or edited. Preprints - unless they show as "accepted" - should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.
Methods for Participant Verification in Social Media Recruitment for a Pilot Study of an mHealth App: Lessons Learned
ABSTRACT
Background:
Web-based advertisements, specifically social media advertisements, are a popular recruitment avenue among research projects involving human participants. Social media recruitment has advantages over other methods (e.g., in-person recruitment), such as aiding teams in reaching the population of interest and increasing enrollment pace at a relatively low cost. Nonetheless, social media recruitment comes with the challenge of fraudulent responses, and therefore effective identity verification procedures must be put in place in order to maintain the integrity of the final sample and data.
Objective:
In this paper, we outline the identity verification methods (herein referred to as “checks”) used in the recruitment process for a pilot study featuring a mobile health (mHealth) intervention app for emerging adults (EAs; aged 18-25) who regularly use cannabis. Each identity verification check is examined for its rate of passing.
Methods:
Participants were recruited via social media advertisements that linked directly to a study eligibility screening survey. Advertisements were posted on Meta (Facebook and Instagram), Snapchat, and TikTok. Participants were enrolled if they met study inclusion criteria (e.g., aged 18-25, reported regular cannabis use), completed the baseline consent and survey, downloaded the app, and passed all identity verification checks. Identity verification checks happened at two checkpoints: directly following screening survey completion (e.g., geolocation check, duplicative IP address check, social media check) and directly following app download and login (duplicative device ID and/or push token check). Failing an identity verification check resulted in exclusion from the study.
Results:
Identity checks were non-exclusive such that a single eligible screening response could undergo multiple checks. Of the 573 eligible screening responses that went through the identity verification process, a total of 3,031 identity verification checks were completed. Of these 3,031 aggregate checks, 396 failed the verification criteria (13.1%), and therefore 396 of the 573 eligible respondents were excluded from continuation in the enrollment process (69.1%). Social media checks, wherein study staff ensured the individual’s public-facing account had personally relevant information, had the highest failure rate (61.5%). The second most common failed check was due to a duplicate device ID upon logging into the app (10.0%), followed by the geolocation check (4.9%), the duplicate IP address check (4.2%), the combination check (time zone; 4.1%), and duplicate push token check (3.2%).
Conclusions:
This paper describes a participant identity verification process for app-based mHealth studies using social media as a recruitment source. A combination of identify verification safeguards is suggested to maintain integrity of the study sample and data. Clinical Trial: ClinicalTrials.gov NCT05824754; University of Michigan IRB: HUM00222194
Citation
