Using Chatbot Technology to Improve Brazilian Adolescents’ Body Image and Mental Health at Scale: Randomized Controlled Trial
Bibliographic record
Abstract
BACKGROUND: Accessible, cost-effective, and scalable mental health interventions are limited, particularly in low- and middle-income countries, where disparities between mental health needs and services are greatest. Microinterventions (ie, brief, stand-alone, or digital approaches) aim to provide immediate reprieve and enhancements in mental health states and offer a novel and scalable framework for embedding evidence-based mental health promotion techniques into digital environments. Body image is a global public health issue that increases young peoples' risk of developing more severe mental and physical health issues. Embedding body image microinterventions into digital environments is one avenue for providing young people with immediate and short-term reprieve and protection from the negative exposure effects associated with social media. OBJECTIVE: This 2-armed, fully remote, and preregistered randomized controlled trial assessed the impact of a body image chatbot containing microinterventions on Brazilian adolescents' state and trait body image and associated well-being outcomes. METHODS: Geographically diverse Brazilian adolescents aged 13-18 years (901/1715, 52.54% girls) were randomized into the chatbot or an assessment-only control condition and completed web-based self-assessments at baseline, immediately after the intervention time frame, and at 1-week and 1-month follow-ups. The primary outcomes were mean change in state (at chatbot entry and at the completion of a microintervention technique) and trait body image (before and after the intervention), with the secondary outcomes being mean change in affect (state and trait) and body image self-efficacy between the assessment time points. RESULTS: Most participants who entered the chatbot (258/327, 78.9%) completed ≥1 microintervention technique, with participants completing an average of 5 techniques over the 72-hour intervention period. Chatbot users experienced small significant improvements in primary (state: P<.001, Cohen d=0.30, 95% CI 0.25-0.34; and trait body image: P=.02, Cohen d range=0.10, 95% CI 0.01-0.18, to 0.26, 95% CI 0.13-0.32) and secondary outcomes across various time points (state: P<.001, Cohen d=0.28, 95% CI 0.22-0.33; trait positive affect: P=.02, Cohen d range=0.15, 95% CI 0.03-0.27, to 0.23, 95% CI 0.08-0.37; negative affect: P=.03, Cohen d range=-0.16, 95% CI -0.30 to -0.02, to -0.18, 95% CI -0.33 to -0.03; and self-efficacy: P=.02, Cohen d range=0.14, 95% CI 0.03-0.25, to 0.19, 95% CI 0.08-0.32) relative to the control condition. Intervention benefits were moderated by baseline levels of concerns but not by gender. CONCLUSIONS: This is the first large-scale randomized controlled trial assessing a body image chatbot among Brazilian adolescents. Intervention attrition was high (531/858, 61.9%) and reflected the broader digital intervention literature; barriers to engagement were discussed. Meanwhile, the findings support the emerging literature that indicates microinterventions and chatbot technology are acceptable and effective web-based service provisions. This study also offers a blueprint for accessible, cost-effective, and scalable digital approaches that address disparities between health care needs and provisions in low- and middle-income countries. TRIAL REGISTRATION: Clinicaltrials.gov NCT04825184; http://clinicaltrials.gov/ct2/show/NCT04825184. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): RR2-10.1186/s12889-021-12129-1.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.002 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 teacher head, 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".