A guided chatbot-based psychological intervention for psychologically distressed older adolescents and young adults: a randomised clinical trial in Jordan - Report - MDSpire
Advertisement
A guided chatbot-based psychological intervention for psychologically distressed older adolescents and young adults: a randomised clinical trial in Jordan
Chatbot-Facilitated Intervention Reduces Psychological Distress in Jordanian Youth
Overview
A randomized clinical trial in Jordan demonstrated that the STARS chatbot intervention significantly reduced anxiety and depression symptoms in older adolescents and young adults experiencing psychological distress compared to enhanced usual care. The intervention showed moderate to large effect sizes and clinically meaningful improvements at 3-month follow-up.
Background
Anxiety and depression affect approximately 4–5% of young people globally, with many cases developing by age 20. In low- and middle-income countries (LMICs), access to mental health care is limited due to resource constraints and stigma. Digital mental health interventions, especially chatbot-based programs, offer a promising approach to bridge this treatment gap. The World Health Organization developed the STARS chatbot, a rule-based digital intervention designed to deliver stress coping strategies to youth in LMICs, with this trial representing its first controlled evaluation in Jordan.
Data Highlights
Outcome
Mean Difference (STARS vs EUC)
95% CI
Effect Size (Cohen's d)
p-value
Good Outcome % (STARS)
Good Outcome % (EUC)
Odds Ratio (OR)
Number Needed to Treat (NNT)
HSCL Total Score
10.21
6.04–14.39
0.68
<0.001
68.6%
41.6%
3.1
2.7
HSCL Anxiety
4.25
2.53–5.97
0.70
<0.001
65.9%
41.9%
2.7
2.7
HSCL Depression
6.09
3.16–9.02
0.61
<0.001
59.1%
39.8%
2.2
3.1
Key Findings
The STARS chatbot intervention led to significantly greater reductions in overall psychological distress (HSCL total score) compared to enhanced usual care at 3 months.
Participants receiving STARS showed large effect size reductions in anxiety symptoms and moderate effect size reductions in depression symptoms.
A higher proportion of STARS participants achieved clinically meaningful improvements in anxiety (65.9% vs 41.9%) and depression (59.1% vs 39.8%) compared to controls.
The number needed to treat to achieve one good outcome ranged from 2.7 to 3.1 across anxiety, depression, and total distress outcomes.
Engagement with the intervention was good, with two-thirds completing at least seven chatbot lessons and over 60% attending four or more e-helper support sessions.
Fidelity of e-helper support calls was high, with 89% of call content delivered adequately.
Clinical Implications
The STARS chatbot intervention offers an effective, scalable digital approach to reduce psychological distress among youth in LMIC settings where traditional mental health resources are scarce. Integration of brief human support enhances engagement and fidelity. Clinicians and policymakers should consider incorporating such chatbot-facilitated programs to expand access to evidence-based mental health care for young people facing barriers to conventional services.
Conclusion
This trial provides robust evidence that a rule-based chatbot intervention, supported by non-specialist helpers, can significantly reduce anxiety and depression symptoms in psychologically distressed young adults in Jordan. Digital mental health tools like STARS hold promise for addressing treatment gaps in LMICs.
References
World Health Organization 2024 -- STARS Chatbot Intervention Trial in Jordan
by Richard A. Bryant, Anne M. de Graaff, Rand Habashneh, Sarah Fanatseh, Dharani Keyan, Aemal Akhtar, Adnan Abualhaija, Muhannad Faroun, Ibrahim Said Aqel, Latefa Dardas, Hadeel Afar, Chiara Servili, Dusan Hadzi-Pavlovic, Mark van Ommeren, Kenneth Carswell
