Target product profiles for digital health technologies including those with artificial intelligence: a systematic review
Macdonald, Trystan B Hogg, H D Jeffry Dinnes, Jacqueline Verrinder, Lucy Maniatopoulos, Gregory Taylor-Phillips, Sian Shinkins, Bethany Dunbar, J Kevin Solebo, Ameenat Lola Sutton, Hannah
Macdonald, Trystan B
Hogg, H D Jeffry
Dinnes, Jacqueline
Verrinder, Lucy
Maniatopoulos, Gregory
Taylor-Phillips, Sian
Shinkins, Bethany
Dunbar, J Kevin
Solebo, Ameenat Lola
Sutton, Hannah
Affiliation
University Hospitals Birmingham NHS Foundation Trust; University of Birmingham; NIHR Birmingham Biomedical Research Centre; Lion Health; University of Leicester; University of Warwick; NHS England; Moorfields Eye Hospital NHS Foundation Trust; University College London Great Ormond Street Institute of Child Health; Alder Hey Children's NHS Foundation Trust; Hardian Health; St. George's University Hospitals NHS Foundation Trust; Department of Health and Social Care London; Imperial College London; University of Nottingham; Care Quality Commission; Medicines and Healthcare Products Regulatory Agency; University College London; Birmingham Health Partners Centre for Regulatory Science and Innovation
Other Contributors
Publication date
2025-05-20
Collections
Research Projects
Organizational Units
Journal Issue
Abstract
Digital health technologies (DHTs), including those incorporating artificial intelligence (AI), have the potential to improve healthcare access, efficiency, and quality, reducing gaps between healthcare capacity and demand. Despite prioritisation in health policy, the adoption of DHTs remains limited, especially for AI, in part due to complex system requirements. Target product profiles (TPPs) are documents outlining the characteristics necessary for medical technologies to be utilised in practice and offer a way to align DHTs' research and development with health systems' needs. This systematic review examines current DHT TPPs' methodologies, stakeholders, and contents. A total of 14 TPPs were identified, most targeted at low- and middle-income settings and communicable diseases. Only one TPP outlined the requirements for an AI device specifically. In total, 248 different characteristics were reported across the TPPs identified and were consolidated down to 33 key characteristics. Some considerations for DHTs' successful adoption, such as regulatory requirements or environmental sustainability, were reported inconsistently or not at all. There was little standardisation in TPP development or contents, and limited transparency in reporting. Our findings emphasise the need for guidelines for TPP development, could help inform these, and could be used as a basis to develop future DHT TPPs.
Citation
Macdonald TB, Hogg HDJ, Dinnes J, Verrinder L, Maniatopoulos G, Taylor-Phillips S, Shinkins B, Dunbar JK, Solebo AL, Sutton H, Attwood J, Pogose M, Given-Wilson R, Greaves F, Macrae C, Pearson R, Tufail A, Liu X, Denniston AK. Target product profiles for digital health technologies including those with artificial intelligence: a systematic review. Front Health Serv. 2025 May 20;5:1537016. doi: 10.3389/frhs.2025.1537016.
Type
Article