Can an unregistered digital health technology be used in a clinical trial?

The short answer

Yes, potentially. A digital health technology does not necessarily need ordinary commercial marketing authorisation before it can be used to acquire data in a clinical investigation. That does not make its use unregulated, nor does it mean that any unapproved device can simply be placed into a trial.

The governing route depends on the technology's actual function, its role in the protocol, the importance of its data, the risks it introduces and the rules of each jurisdiction. Investigational-device requirements, ethics review, informed consent, import controls, data protection and evidence of fitness for purpose may all apply even when full commercial registration does not.

Registration is the wrong first question

Teams often begin by asking whether the product is registered in the target country. That is understandable, but it collapses several different questions into one. Commercial authorisation permits a product to be placed on a market for a defined intended use. A clinical investigation operates under a different framework designed to control experimental use, protect participants and preserve the credibility of the evidence.

The better opening question is: what role does this technology perform in this particular study? A sensor collecting exploratory data is not equivalent to a system that determines eligibility, supplies a primary endpoint, changes a dose or triggers immediate clinical action.

The protocol helps determine the regulatory position

A product's physical design and manufacturer's intended use matter, but the study protocol can materially change its regulatory significance. The same technology may present a very different case when used for exploratory observations than when its output drives treatment or a safety decision.

• Is the technology itself under investigation?
• Does it diagnose, treat, monitor or control therapy?
• Does it influence dosing, eligibility or immediate patient management?
• Is its output a primary endpoint, secondary endpoint or exploratory measure?
• Would false, missing, delayed or corrupted data expose a participant to serious harm?
• Could the technology interfere with another device or impair an established safety function?

An ancillary role can support a lighter route

Consider a generic reusable connected accessory used alongside a drug-delivery device. It detects events, records information and transfers trial data, but does not control delivery, alter the dose, interfere with the injector's safety functions or give clinicians information used for immediate treatment decisions.

That fact pattern may support treatment as ancillary data-acquisition equipment or, where it is an investigational device, a nonsignificant-risk route rather than full premarket authorisation. The conclusion must arise from the final configuration, protocol, data flows and risk analysis—not from calling the product 'ancillary' in a project presentation.

The United States illustrates the distinction

Under the US investigational-device framework, significant-risk and nonsignificant-risk device studies follow different routes. A significant-risk study generally needs FDA approval of an Investigational Device Exemption as well as Institutional Review Board approval. A nonsignificant-risk study does not ordinarily require a separate IDE application to FDA, but it remains subject to the abbreviated IDE requirements and IRB approval.

Some investigations are exempt from most of 21 CFR Part 812, but exemption should not be assumed merely because a DHT is not being marketed. For a trial-only data-acquisition technology, a protocol-specific nonsignificant-risk justification may be the more defensible starting position when its operation and data cannot create a plausible pathway to serious harm. The IRB—and ultimately FDA—may reach a different determination.

A lighter pathway is not a lighter evidence standard

Even where commercial clearance is unnecessary, the sponsor must be able to show that the technology is fit for its intended purpose in the investigation. FDA describes fitness for purpose in terms of whether the validation is sufficient to support the proposed use and the interpretability of the resulting data.

• Define the exact hardware, firmware, software, mobile platform, cloud service and connectivity configuration.
• Verify measurement performance, reliability, time handling, transmission and data completeness.
• Validate the technology for the population, environment and study task in which it will be used.
• Evaluate usability, participant training and foreseeable use error.
• Analyse safety, cybersecurity, privacy and data-integrity risks.
• Explain missing, delayed, duplicated or implausible data and how they will be handled.
• Control updates and configuration changes throughout the investigation.

Scientific importance changes the burden of proof

The more important the data are to the trial conclusion, the stronger the evidence and controls should be. Exploratory data may tolerate uncertainty that would be unacceptable for a primary endpoint. A measurement used to determine eligibility, treatment or safety escalation demands still greater assurance.

This is not merely a regulatory distinction. Poorly controlled technology can create missing data, systematic bias, participant burden, protocol deviations or an endpoint that cannot be interpreted. A technically safe failure can still damage the clinical investigation.

There is rarely only one approval pathway

In a multinational study, resolving the medical-device question does not resolve the complete route. The sponsor may still need clinical-trial authority and ethics approvals, trial-specific import permission, investigational labelling, radio or wireless approval, privacy arrangements and a suitable local legal role.

The practical assessment should therefore separate these workstreams. A country should not be marked simply 'approved' or 'not approved' when its device, ethics, import, wireless and data-protection positions remain at different levels of certainty.

The practical decision framework

Begin with the study, not the product brochure. Freeze the proposed configuration and map exactly how the technology interacts with the participant, other devices, investigators, endpoints and clinical decisions. Then determine the applicable investigational route and the evidence needed to support it.

• Define the technology's role in the protocol.
• Determine whether it is a medical device in that use.
• Identify whether it is under investigation or supporting another investigation.
• Assess the potential harm arising from operation, failure and misleading data.
• Determine the investigational-device, ethics, import, wireless and privacy routes separately.
• Establish fit-for-purpose evidence proportionate to the importance of the data.
• Confirm the conclusion with the relevant authority, ethics body or qualified local adviser.

The conclusion

An unregistered DHT can sometimes be used in a clinical trial, but 'unregistered' is not a regulatory strategy. The defensible route comes from a coherent account of what the technology does, what the protocol asks of it, what evidence supports it and how participants and trial data are protected.

The most useful change in thinking is simple: stop asking only whether the device is registered. Ask what role it plays in the study—and what must be true for that role to be safe, reliable and regulatorily acceptable.