Evidence generation plan for artificial intelligence (AI) technologies to aid opportunistic detection of vertebral fragility fractures: early value assessment
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 companies can strengthen the evidence base by 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 impact of the technologies.
2.1 Essential evidence for future committee decision making
Impact of the artificial intelligence technologies on health-related quality of life
The committee noted that there was limited evidence about how the artificial intelligence (AI) technologies affect health-related quality of life in the short term of at least 12 months. EQ‑5D‑3L is the preferred tool for measuring this outcome. At committee, a review of the evidence from the technologies showed that the utility gain had the largest impact on the cost-effectiveness results for these technologies.
Resource use
More information is needed on how using the technologies would affect resource use during and after implementation, to help the committee understand their long-term resource use impacts. Key areas that will help to address this evidence gap are:
-
long-term resource use costs, such as number and extent of treatments and number of hospital appointment or visits
-
the downstream impacts of using the technologies on the NHS, such as:
-
the number of people referred for spine X-ray or dual-energy X-ray absorptiometry (DEXA) scan, or
-
the number of people receiving medication for osteoporosis
-
-
the time taken to process diagnostic images by reporting practitioners, including additional reviews by specialists.
Ideally, information about implementation, technology acquisition and maintenance costs and payment models could also be collected.
Impact of using AI technologies on the NHS care pathway
A key part of the committee discussion was around the impact of AI technologies on the NHS care pathway for fragility fractures and osteoporosis. For example, changes in the fracture liaison services diagnosis and treatment routes may be needed to accommodate the AI technology. Collecting evidence on this will help the committee understand how using the technologies will affect care in the NHS.
Failure rates and diagnostic accuracy of the AI technologies ideally compared with NHS standard care
The committee noted that the failure rates and diagnostic accuracy outcomes for NHS standard care were not reported adequately. More evidence is needed on the failure rates and diagnostic accuracy of the AI technologies compared with current NHS care.
2.2 Evidence that further supports committee decision making
Diagnostic accuracy of the AI technologies in people under 50 years
The failure rate and diagnostic accuracy outcomes for people younger than 50 years and at risk of VFF, for example people with long-term corticosteroid use or malignancy in the vertebrae, were not reported adequately. More evidence is needed on the failure rates and diagnostic accuracy of the AI technologies when used in these groups.
This page was last updated: