Overcoming technical barriers to enable real-time risk prediction in primary care: recommendations from expert workshops

It is increasingly acknowledged that there are long lags in implementing individualised risk prediction models in healthcare. This contrasts sharply with the accelerating pace of development of risk prediction models. We aimed to develop recommendations to overcome the technical barriers for real-time risk prediction in primary care through review of the academic and policy literature and workshops with key stakeholders. Recommendations include the need for: greater partnerships between academia and industry; the introduction of common standards and interfaces; system-wide investment in the data infrastructure; streamlined, clear and adaptive regulatory pathways for risk assessment models as medical devices; support for on-going monitoring in clinical practice; and the development of sustainable business models for adoption. Addressing these will require system-wide change from researchers, policy makers, funders and regulators. Without these, the benefits of risk prediction models will never be realised.