Building the case for reimbursement after approval — an AI registered as a device
A trial evaluating the clinical effectiveness and health economic impact of an AI medical device that has already cleared regulatory review, in order to seek reimbursement under national health insurance. The registered intervention is a device.
Trial overview (primary data)
- StatusRecruiting
- ConditionsOsteoporosis
- InterventionsDEVICE: VeriOsteo OP
- SponsorTaichung Veterans General Hospital
- Target enrollment1,180 participants
- Period2026-03-13 〜 2026-11-12
Key points
- The project evaluates clinical effectiveness and health economic impact for smart medical devices that have cleared regulatory review and seek reimbursement.
- It is led by Taichung Veterans General Hospital with two collaborating hospitals in a multi-site network.
- Designs used include randomized controlled trials, before-and-after studies, pragmatic and cluster randomized trials and stepped-wedge designs.
- A health economist analyses cost-effectiveness, quantifying reduced costs, diagnostic efficiency and treatment outcomes.
- The intervention is VeriOsteo OP for osteoporosis, target enrolment 1,180, with completion expected 12 November 2026.
1Building the reimbursement case for VeriOsteo OP, an approved AI device
This trial (NCT07620834) evaluates the clinical effectiveness and health economic impact of an AI medical device that has already cleared regulatory review and now seeks reimbursement under national health insurance. The intervention is VeriOsteo OP, for osteoporosis; Taichung Veterans General Hospital leads, with Kaohsiung Medical University Chung-Ho Memorial Hospital and Changhua Show Chwan Memorial Hospital collaborating.
The target is 1,180, it is recruiting, and it runs from 13 March to 12 November 2026. Why a trial is needed after approval follows.
2Approval is the entrance, not the finish
A medical-AI trial usually brings to mind an AI still under validation. Passing review, though, only opens the door to the clinic; beyond it waits a separate hurdle, showing the case for paying for it.
3Two reviews, asking different things
Permission to use and payment from public funds are separate judgments. The registration states that integrating new technology into a reimbursement system while securing clinical value and economic benefit is a challenge worldwide.
4How the study is built
- 1A network of sitesTaichung Veterans General Hospital leads, with Kaohsiung Medical University Chung-Ho Memorial Hospital and Changhua Show Chwan Memorial Hospital collaborating
- 2A range of designsRandomized controlled trials, before-and-after studies, pragmatic and cluster randomized trials, and stepped-wedge designs
- 3Economic evaluationA health economist analyses cost-effectiveness, quantifying reduced costs, diagnostic efficiency and treatment outcomes
- 4Standardizing the procedureA framework covering design, data collection, statistical analysis, economic evaluation and ethical review
What marks this registration is that standardizing the assessment procedure itself is among its aims. It is not only evaluating a device but building the way such evaluation is done.
5The device box
Only 14.6% of these AI trials are registered as a device. Approval does not settle which box a registration falls into. The next article takes up an AI registered as a behavioural intervention.
Why it matters
That a device may be used and that public funds will pay for it are separate judgments requiring different evidence: the first concerns safety and performance, the second whether the cost is warranted. That is why a trial is mounted after approval, and that this registration also aims to standardize the assessment procedure reflects an expectation that the same judgment will keep recurring.
FAQ
Why run a trial on an approved device?
How does regulatory review differ from a reimbursement decision?
What is the standardized impact assessment framework?
Sources (primary)
Source: ClinicalTrials.gov (U.S. NIH/NLM, public domain). This site does not provide medical advice. Verify the latest and exact details with the official source. This site is not endorsed or certified by the NIH/NLM.
- ClinicalTrials.gov (study record, original)
- NCT ID: NCT07620834