Precision targeting of teacher burnout using network-informed ecological momentary interventions
1School of Media, Hunan Institute of Science and Engineering, Yongzhou 425199, China.
Acta Psychologica
|August 14, 2026
Summary
Personalized digital mental health support, using affect network dynamics, significantly improved teacher well-being and happiness compared to standard care. This approach offers a scalable solution for enhancing educator mental health.
Area of Science:
- Psychology
- Digital Mental Health
- Network Science
Background:
- Teacher well-being is crucial for classroom functioning and workforce stability.
- Existing digital mental health programs often lack personalization, failing to leverage individual affect dynamics.
- Generic interventions may not effectively address the complex, person-specific nature of teacher burnout and happiness.
Purpose of the Study:
- To evaluate if micro-interventions, selected using high expected influence (EI) nodes from teachers' affect networks, improve burnout-related EI and happiness.
- To compare network-informed intervention allocation against content-matched random allocation.
- To explore network change as a mediator, personality as a moderator, and benchmark alternative allocation rules.
Main Methods:
- A cluster-randomised trial involving 84 public schools.
- Ecological momentary assessment (EMA) collected baseline and 8-week data on happiness, exhaustion, detachment, efficacy, and rumination.
- Person-specific partial correlation networks were estimated, with an optimization engine selecting interventions based on baseline EI; Bayesian multilevel models and mediation analyses were applied.
Main Results:
- EI-based targeting led to greater reductions in the composite EI-change index (mean difference 0.11) and higher week 7 happiness (4.4 points) compared to the active control.
- A positive slope difference of 0.62 points per week indicated sustained improvements in happiness.
- Network change statistically explained approximately half of the observed happiness improvements, with stronger effects noted for teachers high in conscientiousness.
Conclusions:
- Integrating EMA, network modeling, and EI-driven optimization offers measurable mental health gains for teachers beyond standard interventions.
- This study provides a proof of concept for precision mental health at a district scale, requiring further implementation testing.
- Future research should focus on replication, expanded network analyses, and longer follow-up periods to assess durability and generalizability.
