Predicting Participant Engagement in a Social Media–Delivered Lifestyle Intervention Using Microlevel Conversational Data: A Pilot Feasibility Randomized Trial
Bibliographic record
Abstract
Background Social media–delivered lifestyle interventions have shown promising outcomes, often generating modest but significant weight loss. Participant engagement appears to be an important predictor of weight loss outcomes in these interventions, but engagement generally declines over time and is highly variable within and across studies. Research on factors that influence participant engagement remains scant in the context of social media–delivered lifestyle interventions. Objective The objective of this study was to identify predictors of participant engagement in a social media–delivered lifestyle intervention, including the characteristics of the participants, the posts, and the conversations that followed the posts. Methods We performed secondary analyses using data from a pilot randomized trial that delivered 2 lifestyle interventions via Facebook. We analyzed 80 participants’ engagement data (defined as replies/comments) for each of the 761 posts generated over the 16-week intervention period and linked them to predictors, including characteristics of the posts they engaged on, conversation sentiment, and participant characteristics, using a mixed-effects model. We also performed machine learning–based classification to explore how well the aforementioned measures can predict whether participants will engage with a specific post. Results We found that the probability of participants’ engagement with each post decreased by 0.28% (95% CI 0.16%-0.4%; P<.001) each week, and the probability of participants engaging with posts generated by interventionists was 6.3% (95% CI 5.1%-7.5%; P<.001) higher than their probability of engaging with posts generated by other participants. Participants also had a 6.5% (95% CI 4.9%-8.1%; P<.001) and 6.1% (95% CI 4.1%-8.1%; P<.001) higher probability of engaging with posts that directly mentioned weight and goals, respectively, rather than other types of posts. Participants were 44.8% (95% CI 42.8%-46.9%; P<.001) and 46% (95% CI 44.1%-48%; P<.001) more likely to engage with posts when they were replied to by other participants and by interventionists, respectively. A 1 SD decrease in the sentiment of the conversation on a specific post was associated with a 5.4% (95% CI 4.9%-5.9%; P<.001) increase in the probability of participants’ subsequent engagement with the post. Participants’ engagement on previous posts was also a predictor of engagement in subsequent posts (95% CI 0.74%-0.79%; P<.001). In addition, an ensemble of gradient boosting machine and deep learning–based classification algorithms confirmed the importance of the predictors previously identified and achieved an accuracy of 90.9% in terms of predicting participants’ engagement, using a balanced testing sample with 1600 observations. Conclusions Our findings revealed several predictors of engagement derived from the content generated by interventionists and other participants. Our results have implications for increasing engagement in asynchronous, remotely delivered lifestyle interventions, which could improve outcomes. Our results also point to the potential of data science and natural language processing to analyze microlevel conversational data and identify factors influencing participant engagement. Future studies should validate these results in larger trials. Conflicts of Interest None declared.
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 | 0.016 | 0.021 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.003 | 0.003 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.003 | 0.004 |
| Insufficient payload (model declined to judge) | 0.009 | 0.002 |
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".