Using Social Media to Recruit a Diverse Sample of Participants for a Mobile Health (mHealth) Intervention to Increase Physical Activity: Exploratory Study
Bibliographic record
Abstract
Background: Recruitment of demographically diverse samples in clinical research is often challenging and even more so during the COVID-19 pandemic when traditional in-person recruitment methods could not be implemented. Social media platforms offer an alternative approach for recruiting diverse samples of participants for clinical trials, including those testing digital health interventions. This approach allowed for a quicker recruitment process without the physical constraints associated with traditional in-person methods. Objective: This study aimed to detail the online and social media campaigns used to recruit participants for "Diabetes and Mental Health Adaptive Notification Tracking and Evaluation" (DIAMANTE), a randomized controlled trial testing a smartphone-based intervention (a text messaging system that uses machine learning to personalize content) to increase physical activity for patients with diabetes and depression. In describing the recruitment process, we seek to offer insights to the research community on recruitment through online and social media advertisements for diverse communities. Methods: This study sought to recruit demographically diverse individuals in the United States through social media, including paid advertisements on Craigslist and Facebook (Meta). For the DIAMANTE project recruitment, we created 18 personas that mapped into the population's target demographics using a user-centered design methodology. We deployed targeted English and Spanish ads on Craigslist and Facebook in 78 cities based on county-level demographics and diabetes prevalence data to target diverse individuals aged 18-75 years old, who had been diagnosed with diabetes and had documented depressive symptoms. Results: A total of 1379 individuals completed the study's initial screening survey. Of those, 71 respondents on Facebook and 508 on Craigslist were interested in enrolling in our study. In total, 26 out of 58 (45%) eligible respondents from Facebook and 50 out of 235 (21.3%) eligible respondents from Craigslist were eventually recruited in the randomized controlled trial. In all, both platforms showed poor performance in recruiting Spanish speakers, with Facebook advertisements accounting for 0 and Craigslist for 4 (5.3%) of such participants. When it came to English speakers, Craigslist proved to be the better performing platform compared to Facebook, both in terms of reach (579 vs 71) and cost-effectiveness (US $67.61 in average cost per recruited participant vs US $80.16). While Craigslist ads reached more people, resulting in more completed screening surveys than Facebook ads, there was a higher number of ineligible and incomplete enrollment from Craigslist compared with Facebook, leading to a relatively lower conversion rate (9.4% vs 36.6%). Importantly, participants recruited through Craigslist were more ethnically and racially diverse than those recruited from Facebook. Conclusions: Results from this study revealed that it is possible to recruit diverse sample sets using social media and online advertisements. However, despite targeted recruitment efforts, social media recruitment of Spanish speakers proved especially challenging and costly. Further research is needed to determine systematic, online methods for recruiting marginalized communities.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.019 | 0.020 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.004 | 0.002 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.002 | 0.001 |
| Insufficient payload (model declined to judge) | 0.004 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".