PHG Foundation examines structural barriers to risk prediction tools in primary care
A new commentary from the PHG Foundation at the University of Cambridge argues that deploying disease risk prediction models in primary care requires overcoming organisational and workflow challenges that go well beyond model accuracy.
A commentary published on 2 September 2026 by the PHG Foundation, the health policy think tank based at the University of Cambridge, addresses what it describes as the non-technical obstacles to embedding disease risk prediction tools in primary care settings.
The piece argues that even technically sound models can fail to achieve real-world impact if the structural conditions in which they would be used — including clinical workflow integration, practitioner trust, and system-level incentives — are not adequately considered during design and deployment planning. The authors position this as a wider challenge for translational genomics and precision public health, drawing on examples from cardiovascular and cancer risk stratification research.
The commentary does not evaluate a specific tool or model but instead frames implementation science as a prerequisite for any risk prediction technology that aims to reach patients through primary care. It situates the discussion within ongoing debates about polygenic risk scores and multi-factorial risk tools entering health systems without sufficient attention to how frontline clinicians will use them.
The piece is relevant to policymakers, health systems researchers, and primary care educators thinking about how genomic and epidemiological evidence is translated into practice. It does not provide clinical recommendations, and its observations are framed at the level of health systems research rather than individual clinical guidance.
Plain-language version
For patients, families, and general readers. Educational only — not medical advice.
Researchers at the PHG Foundation, a health policy group connected to the University of Cambridge, have published a commentary about why it can be difficult to introduce tools that predict the risk of diseases into GP surgeries and other primary care settings.
The authors argue that even when a risk prediction tool works well technically, it may still not be used effectively unless the way GP practices are organised, and the way doctors and nurses work day-to-day, are also taken into account when the tool is being planned.
The piece does not describe a new test or tool available to the public. It is an academic commentary aimed at researchers and health system planners who design and evaluate these kinds of tools.
This is an educational summary, not medical advice. If anything here raises questions for you, please speak with your GP or a clinical professional.
Sources
Read the original reporting — these are the public sources this summary draws from.
-
Primary source PHG Foundation · 2026-09-02Risk prediction tools in primary care: it's not just about the model