Responsible data selection method for algorithmic personalization of health apps: a case study on promoting mental health
-
By
-
Esra Cemre Su de Groot
-
Ujwal Gadiraju
-
Olya Kudina
-
Loes Keijsers
-
Manon H. J. Hillegers
-
Willem-Paul Brinkman
-
June 24, 2026
-
Clinical Scorecard: Ethical Data Selection Strategies for Personalized Health Applications: A Case Study Focused on Mental Health Promotion
At a Glance
| Category | Detail |
|---|
| Condition | Mental Health Promotion |
| Key Mechanisms | Algorithmic personalization using user input data for health apps. |
| Target Population | Adolescents |
| Care Setting | Digital health applications |
Key Highlights
- Proposes a stepwise method for Responsible Data Selection (ReDS) for algorithmic personalization.
- Demonstrates the ReDS method through a case study involving 1181 adolescents.
- Identifies emotional state and prior completion rates as input data with ethical implications.
- Suggests tiredness data as a less sensitive alternative to emotion data.
- Highlights the importance of ethical considerations in algorithm development.
Guideline-Based Recommendations
Diagnosis
Management
Monitoring & Follow-up
Risks
- Privacy risks associated with sensitive user data.
- Potential biases in emotion recognition systems.
Patient & Prescribing Data
Adolescents using mental health promotion apps.
Utilization of coping strategy challenges based on cognitive behavioral therapy.
Clinical Best Practices
- Incorporate ethical considerations in the development of personalization algorithms.
- Assess the utility-risk trade-off when selecting input data.
Related Resources & Content
