Methods to Analyse High Frequency Patient-Generated Health Data to Assess Treatment Response: Scoping Review
This review identified 190 studies that used high frequency ePGHD to assess treatment response, presenting preliminary findings here for 50. Most were observational (44%) and collected different ePGHD types e.g. symptoms and diet (52%) using mobile apps (80%). Some studies excluded ePGHD before using descriptive statistics or ignored correlations inherent in repeated measures when clustering ePGHD using K-means. However, some used robust methods such as mixed effects models that account for these correlations. The findings highlight the need for broader adoption of robust methods to improve the accuracy of treatment response estimates.
Authors
- Sabine N van der Veer (ORCID: https://orcid.org/0000-0003-0929-436X)
- David Jenkins (ORCID: https://orcid.org/0000-0001-6687-3507)
- Michelo Banda (ORCID: https://orcid.org/0000-0002-7366-716X)
- William G Dixon
- Roberto Cahuantzi (ORCID: https://orcid.org/0000-0002-0212-6825)
- Mariam Al-Attar (ORCID: https://orcid.org/0000-0002-3450-5803)
- Siân Bladon (ORCID: https://orcid.org/0000-0001-9087-6505)
Institutions
- University of Manchester (GB)
Publication Details
- Journal
- Studies in health technology and informatics
- Published
- 2025-05-15
- DOI
- https://doi.org/10.3233/shti250530
- Primary Topic
- Mental Health Research Topics
- Type
- article
- Field-Weighted Citation Impact
- 0.00