Why Science? | FDA Decentralized Trials, Remote Visits and Local Healthcare

FDA defines a decentralized clinical trial as one with trial-related activities at locations other than traditional trial sites. Examples include telehealth visits, in-home visits and visits with local healthcare providers. The regulatory expectations for investigations remain; decentralisation changes how activities are implemented, supervised and recorded.

For a decentralized element changes location, not scientific responsibility, the useful scientific move is to keep the human decision and the evidence trail together. FDA defines a decentralized clinical trial as one with trial-related activities at locations other than traditional trial sites. Examples include telehealth visits, in-home visits and visits with local healthcare providers. The regulatory expectations for investigations remain; decentralisation changes how activities are implemented, supervised and recorded. Keep this boundary beside the conclusion: Convenience does not reduce the need for participant protection or reliable evidence.

A strong science note separates observation from interpretation. List each trial activity, its location, responsible person, source record and escalation route. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.

Contents · Next chapter

FDA identifies the guidance as final Level 1, dated September 2024, under docket FDA-2022-D-2870. CDRH is one of the issuing centres. It covers trials requiring INDs or approved IDEs and notes that its considerations may also be relevant to nonsignificant-risk device studies.

The evidence in the september 2024 guidance is current and final becomes stronger when roles, records and uncertainty remain visible. FDA identifies the guidance as final Level 1, dated September 2024, under docket FDA-2022-D-2870. CDRH is one of the issuing centres. It covers trials requiring INDs or approved IDEs and notes that its considerations may also be relevant to nonsignificant-risk device studies. The conclusion stops here: The guidance is cross-product and each protocol still needs product-specific risk and regulatory analysis.

A strong science note separates observation from interpretation. Record the document status, date, docket, issuing centres and exact study scope. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.

Contents · Previous chapter · Next chapter

A decentralized element may reduce travel, improve access or allow observations in daily life. The protocol should explain why each element belongs and whether the target population can use it. Adding technology or remote visits without a clear evidentiary purpose can introduce burden, missingness and operational variability.

A careful reader can retrace decentralization should answer a real trial need from the planned question to the recorded action. A decentralized element may reduce travel, improve access or allow observations in daily life. The protocol should explain why each element belongs and whether the target population can use it. Adding technology or remote visits without a clear evidentiary purpose can introduce burden, missingness and operational variability. Preserve this guardrail: A remote option is not automatically more inclusive or more efficient.

A strong science note separates observation from interpretation. Write the participant and scientific problem that each decentralized element is meant to solve. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.

Contents · Previous chapter · Next chapter

The investigational plan should describe visits, procedures, assessments, data collection and communication across traditional and remote locations. Activities requiring physical examination, specialised equipment or urgent clinical judgement may still need in-person performance. Contingency routes should cover technical failure or a participant’s changing condition.

For the protocol should specify where and how activities occur, the useful scientific move is to keep the human decision and the evidence trail together. The investigational plan should describe visits, procedures, assessments, data collection and communication across traditional and remote locations. Activities requiring physical examination, specialised equipment or urgent clinical judgement may still need in-person performance. Contingency routes should cover technical failure or a participant’s changing condition. Keep this boundary beside the conclusion: A nominally remote visit cannot substitute for an examination that requires direct physical assessment.

A strong science note separates observation from interpretation. Classify every visit as remote, local or central and justify the setting against the task and risk. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.

Contents · Previous chapter · Next chapter

FDA says sponsors may consider telehealth when no in-person interaction is needed. Suitability depends on the investigational product, medical condition, required observations, participant capability, privacy and available technology. The protocol should clarify identity verification, documentation and when a remote contact becomes an in-person referral.

The evidence in telehealth suitability depends on product and population becomes stronger when roles, records and uncertainty remain visible. FDA says sponsors may consider telehealth when no in-person interaction is needed. Suitability depends on the investigational product, medical condition, required observations, participant capability, privacy and available technology. The protocol should clarify identity verification, documentation and when a remote contact becomes an in-person referral. The conclusion stops here: Video quality and participant self-report may not support every clinical judgement.

A strong science note separates observation from interpretation. Run one assessment through remote and in-person modes and identify differences that could change the endpoint. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.

Contents · Previous chapter · Next chapter

Local providers may conduct routine examinations, collect samples, perform imaging or deliver other protocol-specified care when appropriately qualified and authorised. The investigator should know what was delegated, receive results promptly and retain oversight. Local care records need a clear path into the trial record.

A careful reader can retrace local healthcare providers can perform defined activities from the planned question to the recorded action. Local providers may conduct routine examinations, collect samples, perform imaging or deliver other protocol-specified care when appropriately qualified and authorised. The investigator should know what was delegated, receive results promptly and retain oversight. Local care records need a clear path into the trial record. Preserve this guardrail: Ordinary clinical experience does not automatically include training on the protocol or investigational product.

A strong science note separates observation from interpretation. Create a local-provider task sheet with qualification, training, source record, transfer time and escalation trigger. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.

Contents · Previous chapter · Next chapter

Home visits can make participation easier, but they introduce environmental, privacy, specimen-handling and emergency-response questions. The study should define permitted procedures, supplies, identity checks, infection control, documentation and how staff obtain help. Variability in home settings should be considered in measurement and analysis.

For home visits need the same precision as site visits, the useful scientific move is to keep the human decision and the evidence trail together. Home visits can make participation easier, but they introduce environmental, privacy, specimen-handling and emergency-response questions. The study should define permitted procedures, supplies, identity checks, infection control, documentation and how staff obtain help. Variability in home settings should be considered in measurement and analysis. Keep this boundary beside the conclusion: A familiar setting can still produce protocol deviations or unrecognised risk.

A strong science note separates observation from interpretation. Perform a home-visit failure-mode analysis covering environment, equipment, privacy and emergency response. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.

Contents · Previous chapter · Next chapter

The sponsor should identify decentralized activities in the protocol, choose qualified investigators and service providers, provide training, monitor conduct and maintain appropriate records. Contracts can allocate tasks, but they should not create blind spots between organisations. Up-front and continuing risk assessment helps focus oversight where failure could harm participants or distort results.

The evidence in the sponsor designs the operational system becomes stronger when roles, records and uncertainty remain visible. The sponsor should identify decentralized activities in the protocol, choose qualified investigators and service providers, provide training, monitor conduct and maintain appropriate records. Contracts can allocate tasks, but they should not create blind spots between organisations. Up-front and continuing risk assessment helps focus oversight where failure could harm participants or distort results. The conclusion stops here: Outsourcing an activity does not outsource the sponsor’s regulatory responsibility.

A strong science note separates observation from interpretation. Build a sponsor oversight map linking each vendor deliverable to a verification record and review frequency. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.

Contents · Previous chapter · Next chapter

The investigator remains responsible for trial conduct and participant protection at the site, including appropriately delegated remote activities. Delegation should match qualifications and be documented. The investigator needs timely access to participant data, safety signals and local test results to make medical decisions and maintain source records.

A careful reader can retrace the investigator retains supervision and access to information from the planned question to the recorded action. The investigator remains responsible for trial conduct and participant protection at the site, including appropriately delegated remote activities. Delegation should match qualifications and be documented. The investigator needs timely access to participant data, safety signals and local test results to make medical decisions and maintain source records. Preserve this guardrail: A central platform does not replace investigator judgement or supervision.

A strong science note separates observation from interpretation. Trace one abnormal local result from measurement to investigator review, participant contact and final record. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.

Contents · Previous chapter · Next chapter

Trial personnel, local providers, participants and caregivers may need different training on protocol tasks, technology, reporting and investigational products. Training materials should use understandable language and be updated when systems or procedures change. Competence can be checked through demonstrations or practice rather than attendance alone.

For training must match each role and tool, the useful scientific move is to keep the human decision and the evidence trail together. Trial personnel, local providers, participants and caregivers may need different training on protocol tasks, technology, reporting and investigational products. Training materials should use understandable language and be updated when systems or procedures change. Competence can be checked through demonstrations or practice rather than attendance alone. Keep this boundary beside the conclusion: Training completion records do not prove a person can perform the task correctly.

A strong science note separates observation from interpretation. Define a practical competency check for every role that creates critical data or manages risk. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.

Contents · Previous chapter · Next chapter

Consent may be obtained remotely when applicable requirements are satisfied and the IRB approves the process and materials. Participants still need opportunity for questions, understandable information and a copy of the consent documentation. The IRB should see how decentralized procedures affect privacy, local context, recruitment and risk.

The evidence in informed consent and irb oversight remain central becomes stronger when roles, records and uncertainty remain visible. Consent may be obtained remotely when applicable requirements are satisfied and the IRB approves the process and materials. Participants still need opportunity for questions, understandable information and a copy of the consent documentation. The IRB should see how decentralized procedures affect privacy, local context, recruitment and risk. The conclusion stops here: Electronic or remote consent does not weaken voluntariness and comprehension requirements.

A strong science note separates observation from interpretation. Test the consent process under limited bandwidth, low digital confidence and a need for an interpreter. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.

Contents · Previous chapter · Next chapter

For investigational devices, the plan should address shipping, receipt, storage, use, return or disposal, maintenance and accountability. Devices used at home may require setup checks, training and technical support. The investigator and sponsor should know who has the device and whether its configuration remains correct.

A careful reader can retrace medical devices require controlled use and accountability from the planned question to the recorded action. For investigational devices, the plan should address shipping, receipt, storage, use, return or disposal, maintenance and accountability. Devices used at home may require setup checks, training and technical support. The investigator and sponsor should know who has the device and whether its configuration remains correct. Preserve this guardrail: Successful delivery does not demonstrate correct use or intact device performance.

A strong science note separates observation from interpretation. Create a device chain-of-custody record from release through return, including version and condition checks. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.

Contents · Previous chapter · Next chapter

Direct distribution may reduce visits, yet temperature, security, timing, labelling and recipient verification still matter. The sponsor should use appropriate packaging and shipping methods, document disposition and plan for delays or damaged shipments. A failed shipment can also affect treatment continuity and study endpoints.

For packaging and shipping can affect product integrity, the useful scientific move is to keep the human decision and the evidence trail together. Direct distribution may reduce visits, yet temperature, security, timing, labelling and recipient verification still matter. The sponsor should use appropriate packaging and shipping methods, document disposition and plan for delays or damaged shipments. A failed shipment can also affect treatment continuity and study endpoints. Keep this boundary beside the conclusion: A courier tracking number is not a complete integrity record.

A strong science note separates observation from interpretation. Map environmental controls, custody events, acceptance criteria and contingency actions for one shipment. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.

Contents · Previous chapter · Next chapter

FDA recommends a safety monitoring plan that reflects decentralized conduct, captures adverse events and medication or use errors, instructs local providers how to report concerns and tells participants where to seek local help. DHT alerts may also feed monitoring. Roles, timing and escalation authority should be explicit.

The evidence in safety monitoring must work across locations becomes stronger when roles, records and uncertainty remain visible. FDA recommends a safety monitoring plan that reflects decentralized conduct, captures adverse events and medication or use errors, instructs local providers how to report concerns and tells participants where to seek local help. DHT alerts may also feed monitoring. Roles, timing and escalation authority should be explicit. The conclusion stops here: Collecting alerts without a staffed decision route can create false reassurance.

A strong science note separates observation from interpretation. Simulate a serious event outside business hours and time every handoff until care and reporting occur. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.

Contents · Previous chapter · Next chapter

Platforms may schedule visits, communicate, capture consent, receive data, manage products or hold trial records. Systems should be fit for purpose, secure and compliant with applicable electronic-record requirements. Access, training, audit trails, backup, data transfer and inspection readiness should be planned.

A careful reader can retrace electronic systems should support traceable operations from the planned question to the recorded action. Platforms may schedule visits, communicate, capture consent, receive data, manage products or hold trial records. Systems should be fit for purpose, secure and compliant with applicable electronic-record requirements. Access, training, audit trails, backup, data transfer and inspection readiness should be planned. Preserve this guardrail: A convenient app may be unsuitable for regulated records or sensitive communication.

A strong science note separates observation from interpretation. Create a system inventory with function, data, owner, validation evidence, access controls and retention. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.

Contents · Previous chapter · Next chapter

Measurements made by different people, devices or settings may not be directly interchangeable. The protocol and analysis plan should address mode effects, calibration, standardisation, clustering, missingness and deviations. Noninferiority or other designs need special care if decentralisation affects assessment sensitivity.

For mode and location can add variability, the useful scientific move is to keep the human decision and the evidence trail together. Measurements made by different people, devices or settings may not be directly interchangeable. The protocol and analysis plan should address mode effects, calibration, standardisation, clustering, missingness and deviations. Noninferiority or other designs need special care if decentralisation affects assessment sensitivity. Keep this boundary beside the conclusion: More data points cannot compensate for systematically different measurement methods.

A strong science note separates observation from interpretation. Compare distributions and agreement for central, local and home measurements before pooling them. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.

Contents · Previous chapter · Next chapter

The receipt records each decentralized element, rationale, location, responsible party, qualification, training, source record, data transfer, product custody, consent method, safety route, electronic system, deviations, analysis handling and source guidance. It becomes the map that lets a reviewer reconstruct a trial spread across many places.

The evidence in a complete dct receipt keeps the distributed trial coherent becomes stronger when roles, records and uncertainty remain visible. The receipt records each decentralized element, rationale, location, responsible party, qualification, training, source record, data transfer, product custody, consent method, safety route, electronic system, deviations, analysis handling and source guidance. It becomes the map that lets a reviewer reconstruct a trial spread across many places. The conclusion stops here: A map exposes gaps; it does not make fragmented oversight adequate.

A strong science note separates observation from interpretation. Give the receipt to an independent reviewer and ask them to follow one participant through every location and decision. Record the source, version or date, population or device context, method, result and one uncertainty. If the official record does not show a field, write “not shown” instead of filling the gap from memory.

Contents · Previous chapter