Evidence generation plan for artificial intelligence technologies to help detect fractures on X-rays in urgent care

2 Evidence gaps

This section describes the evidence gaps, why they need to be addressed and their relative importance for future committee decision making.

The committee will not be able to make a positive recommendation without the essential evidence gaps (see section 2.1) being addressed. The company can strengthen the evidence base by also addressing as many other evidence gaps (see section 2.2) as possible. This will help the committee to make a recommendation by ensuring it has a better understanding of the patient or healthcare system benefits of the technology.

2.1 Essential evidence for future committee decision making

Diagnostic accuracy

To evaluate the efficacy of these technologies, it is essential to have further understanding about their diagnostic accuracy in urgent care settings that reflects the technology users in the NHS. Higher diagnostic accuracy for the presence or absence of fractures can minimise risks and costs associated with incorrect or delayed treatment.

Clinical and service outcomes

To assess the impact of these technologies on healthcare delivery and patient health, it is essential to measure outcomes related to both clinical effectiveness and efficiency of urgent care services. To evaluate clinical effectiveness, data collection should focus on changes in patient outcomes associated with reduced misdiagnosis rates and the impact of missed fractures. To provide further understanding on service efficiency, evidence should be collected on whether the technology can improve the service, influence decisions, and reduce the need for additional imaging and onward referral to fracture clinics.

2.2 Evidence that further supports committee decision making

Effectiveness in different subgroups

There was limited evidence for the use of the technologies in certain subgroups. Analysing the data collected to consider these groups will help the committee to understand the benefits of the technologies to broader populations. Subgroups to consider include:

  • age (for example, children and young people)

  • sex

  • ethnicity

  • socioeconomic status

  • fracture types

  • conditions that affect bone health (for example, myeloma, osteoarthritis, osteoporosis, osteogenesis imperfecta, Paget's disease, rickets, osteomalacia and metastatic bone disease).

People with conditions that affect bone health, and some people with joint replacements, may have different healthcare needs or present additional diagnostic challenges to healthcare professionals.

Costs associated with implementing the AI technologies

Collecting data on the costs associated with establishing the infrastructure for AI technologies is important for understanding the financial investment that is needed. It will also provide understanding of the feasibility and sustainability of integrating AI technologies into routine healthcare. This information can contribute to estimates of cost effectiveness.