Scientific job: CLAIMED. The existing Evidence Web owns how studies become evidence. This article owns the human-participant movement from protocol → regulatory/ethics authority → eligibility → research-specific informed consent → enrolment → allocation/randomisation where applicable → investigational intervention/procedure → safety and endpoint monitoring → new-information/re-consent → withdrawal/continuation → study completion → results/publication handoff.

Wait, what? Joining a clinical trial is not the same as receiving ordinary treatment with extra paperwork.

Research has a different purpose from ordinary clinical care. Treatment is primarily intended to benefit the individual patient; research is designed to answer a question that can produce generalisable knowledge. A participant may benefit, receive no benefit, or experience harms that are still being characterised.

That is why clinical-trial participation needs a separate public node. The scientific question, participant rights, protocol rules and safety monitoring must remain visible at the same time.

The research-participation tube

Research question → protocol → regulator/ethics approval → site activation → candidate identification → eligibility → informed consent → enrolment → randomisation/allocation where applicable → investigational intervention → scheduled assessments → adverse-event monitoring → new information/re-consent → protocol completion or withdrawal → end-of-study follow-up → results and evidence handoff.

1. A protocol is a contract for how the question will be tested

The protocol defines objectives, eligibility, interventions, comparators, outcomes, visits, monitoring, statistical methods and safety procedures. It constrains what the research team may do and helps prevent the study from changing its rules after seeing the results.

For eduKateAI, the protocol should travel with study identifier, version, sponsor, principal investigator/site, jurisdiction, ethics/regulatory approvals and amendment history.

2. Regulatory and ethics approval are different gates

Clinical trials may require review by a regulatory authority and an institutional ethics or review board, depending on the type of study and jurisdiction. The regulator evaluates applicable product/trial requirements; the ethics review focuses strongly on participant protection, consent and acceptable risk.

For eduKateAI: ethics approval ≠ regulatory approval, and neither should be silently assumed from the existence of a trial listing.

3. Eligibility is not merely a diagnosis match

Inclusion and exclusion criteria can involve age, disease stage, biomarkers, previous treatments, organ function, medicines, pregnancy status, performance status and many other variables. A person can have the right disease and still be ineligible for a particular protocol.

The useful rule is: disease matches trial title ≠ participant eligible.

4. Research consent is an ongoing process

Singapore HSA’s clinical-trial guidance treats informed consent as an ongoing process rather than a one-time signature. Participants should receive relevant information about purpose, procedures, foreseeable risks, alternatives, confidentiality, compensation/treatment for research injury where applicable, and the voluntary nature of participation.

For eduKateAI: signed form ≠ consent complete forever.

5. Therapeutic misconception is a specific risk

Participants may assume that every research decision is individually optimised for them. But randomisation, protocol-defined dosing or fixed visit schedules can exist because the study is trying to answer a scientific question.

The consent route should therefore make explicit which parts are standard care, which are research procedures, and what remains uncertain.

6. Randomisation protects the comparison, not the participant from uncertainty

In randomised studies, allocation by chance helps reduce systematic differences between treatment groups. Blinding can reduce bias further. But randomisation also means the participant may not receive the intervention they hoped for.

For eduKateAI: enrolled ≠ assigned to experimental treatment.

7. Investigational treatment must remain labelled as investigational

A medicine, device, cell/gene therapy or procedure may be tested because its benefits and harms are not fully established for that use. Trial participation should therefore preserve product/intervention identity, trial arm, dosing/procedure history and regulatory status.

The Advanced Cell & Gene Therapy Web, Pharmacy Web or Device Safety Web may own the product itself. This node owns the fact that the product is being used under a research protocol.

8. Safety monitoring can change the study

Adverse events, serious adverse events and important new safety information can trigger review, protocol changes, treatment interruption, study suspension or new consent discussions. HSA’s trial framework requires safety information to be managed throughout the trial.

The return loop is therefore participant event → site assessment → sponsor/regulator/ethics reporting where required → updated risk state → participant-level and study-level action.

9. New information can require re-consent

If new findings materially affect willingness to continue—such as new risks, new alternatives or important changes to procedures—the participant may need updated information and an opportunity to decide again.

For eduKateAI: consented at enrolment ≠ automatically consented to a materially changed trial.

10. Withdrawal is a protected participant state

HSA guidance protects voluntary withdrawal. A participant may decide to stop the investigational intervention or withdraw from the study. What happens to already-collected data or whether safety follow-up is still recommended depends on protocol, law, ethics and the participant’s choices.

The system should distinguish withdraw treatment, withdraw study participation, lost to follow-up and investigator/sponsor discontinuation.

11. Protocol deviation and participant non-adherence are different objects

A research site can deviate from the approved protocol, and a participant can also miss visits or doses. These have different causes and governance implications. Blaming the participant for a system deviation destroys useful accountability.

For eduKateAI, preserve who/what deviated, why, whether safety was affected and what corrective action occurred.

12. Trial completion is not evidence completion

After the last participant visit, data still need cleaning, analysis, interpretation and reporting. Registration and publication can help prevent selective disappearance of inconvenient results.

This is where the Clinical Trials Web hands off to the Evidence Web: participant experience becomes study data; study data becomes an evidence object; evidence still requires appraisal and synthesis.

13. The participant needs a post-trial receiver

When investigational treatment stops, ordinary clinical care still continues. Patients may need transition back to standard therapy, ongoing safety follow-up, access to an approved treatment, or another trial depending on the circumstances.

The study should therefore not end with a research database event while the human patient has no next clinical owner.

Characteristic failure modes

• Trial = treatment error: research purpose disappears behind clinical language.
• Listed = approved error: registry presence treated as proof of regulatory/ethics approval.
• Disease match = eligible error: protocol criteria disappear.
• Signature = permanent consent error: ongoing information and re-consent are ignored.
• Enrolled = experimental arm error: randomisation is forgotten.
• Adverse event = individual-only error: study-level safety implications are missed.
• Withdrawal = failure error: participant autonomy is treated as protocol non-compliance.
• Trial complete = evidence settled error: analysis, publication and synthesis are skipped.

The eduKateAI routing contract

• Canonical public owner: Clinical Trials & Research Participation Web.
• Input state: person being considered for, enrolled in or followed within a clinical study.
• Primary job: preserve protocol authority, eligibility, voluntary consent, allocation, investigational exposure, safety monitoring and participant rights through the study.
• Do not collapse: study listing ≠ approval; consent form ≠ ongoing consent; enrolment ≠ treatment assignment; withdrawal ≠ wrongdoing; trial completion ≠ evidence certainty.
• Handoffs: Clinical Ethics/Consent, Evidence Web, Pharmacy, Medical Devices, Advanced Cell & Gene Therapy, Laboratory Medicine, Oncology and all relevant disease specialties.
• Return receipt: eligible/not eligible, consented/declined, enrolled, arm assigned where applicable, treatment received, adverse events captured, re-consent completed where needed, withdrawal/completion documented, clinical follow-up owner confirmed.

Authoritative routes

• Singapore HSA — Conducting clinical trials
• Singapore HSA — Informed consent in clinical trials
• ClinicalTrials.gov
• WHO ICTRP — International Clinical Trials Registry Platform

Educational boundary: this article explains clinical-trial participation and governance architecture. It does not determine eligibility, advise whether someone should join or leave a study, interpret trial consent documents or replace the investigator, ethics board, regulator or treating clinical team.