Description
Deliver brief interventions on mobile devices in daily life, tailored by in-the-moment data, and test their components by randomising them repeatedly within each person.
Ecological momentary interventions (EMIs) are brief psychological interventions delivered through mobile technology, stand-alone or blended with therapy. Carter's three levels are low interaction (the same prompts and content for all), interactive (content triggered by reported scores) and integrative (learning systems tailored to each person, such as just-in-time adaptive interventions using sensor and response history). Designers decide when to deliver (fixed, random or after events), what data feed tailoring and how complex the system should be. The micro-randomised trial is a sequential factorial design: at each decision point each participant is randomised to receive a component or not, giving estimates of proximal effects and how they vary by context. Evaluation follows the WHO digital health life cycle (prototype, pilot, demonstration, scale-up, integration). Most studies stop at early phases, engagement drops quickly, and high-quality trials are few.
Attributes
- Discipline
- Experimentation & Causal Inference
- Loss type
- Siloed
- Stage
- Design, Evaluation, Post-launch
- Data
- Quantitative
- Effort
- Months
Steps
- Define the target moments and the decision rule for when to intervene.
- Choose the complexity level and the data (self-report, sensors, history) that feed tailoring.
- Involve clinicians or domain experts and end users qualitatively in early design.
- Prototype and test functionality, stability and usability.
- Run a micro-randomised trial: at each decision point randomise each participant to a component or no intervention, typically with fixed probabilities.
- Estimate proximal effects and their moderators (time, context, prior response), then run larger trials with longer follow-up.
Sources
- The Open Handbook of Experience Sampling Methodology (Myin-Germeys & Kuppens, 2022)
Also known as
- EMI
- Just-in-time adaptive interventions (JITAI)
- Micro-randomised trial
- Micro RCT
Categories
- Analytics & Behaviour
- Experimentation & Causal Inference