91828 Chap.11 Research Ethics and Governance
Research Ethics and Governance
Ethical clinical research requires social value, scientific validity, fair selection, proportionate risk and meaningful consent. This chapter treats consent as an ongoing communication process and governance as the system that implements approved protections.
Cases on optional biospecimen storage, language exclusion and identifiable data turn principles into procedures for privacy, monitoring, amendments, safety reporting and accountable oversight.
What this chapter covers
- 01
Social value
- 02
Scientific validity
- 03
Informed consent
- 04
Fair participant selection
- 05
Privacy and data governance
- 06
Safety and institutional oversight
Worked example · free
Govern optional future sample use
- +1Separate samples required for the trial from optional future storage.
- +1Obtain clear choices that explain future scope, withdrawal limits and lack of guaranteed benefit.
- +1Define coding, access, linkage, retention, sharing and destruction controls.
- +1Require future uses to fit consent and receive appropriate ethics and governance review.
Key terms
- Informed Consent
- Informed consent is voluntary agreement based on understandable information, capacity and genuine choice.
- Equipoise
- Equipoise is genuine uncertainty within the relevant expert community about which trial strategy is preferable.
- Data Governance
- Data governance assigns controls for collection, access, linkage, retention, sharing and destruction of research data.
- Adverse Event
- An adverse event is an unfavourable medical occurrence during research that requires defined assessment and reporting.
Research Ethics and Governance FAQ
Is ethics approval enough for the whole study?
No. Approval confirms a reviewed plan at a point in time. The team must implement consent, privacy, safety and data procedures, report deviations or unexpected risks, control amendments and maintain site governance throughout conduct and closeout.
Why is a weak design an ethical problem?
Participants accept burden and risk to generate valuable knowledge. An underpowered, biased or incoherent study may be unable to answer its question, wasting that contribution. Scientific validity and feasible delivery are therefore part of ethical justification.
Can vulnerable groups simply be excluded?
Automatic exclusion can deny relevant evidence and produce unfairly narrow findings. Inclusion needs proportionate safeguards for capacity, language, power and risk. Any exclusion should follow a clear scientific or safety reason rather than convenience or administrative cost alone.
What is therapeutic misconception?
It is confusion between research designed to produce generalisable knowledge and personalised clinical care. Consent should explain uncertainty, randomisation, alternatives and which procedures occur only for research. Clinician involvement and hopeful language need care because they can strengthen mistaken expectations of individual benefit.
How should privacy risk be managed?
Minimise identifiers, separate contact details, use approved secure storage, restrict access by role and document linkage, retention, sharing and destruction. Explain residual limits honestly. Low physical risk does not imply low informational harm when health data could affect social or professional life.
What happens when a protocol changes?
Assess why the amendment is needed, its effect on risk and validity, and which approvals or site authorisations are required before implementation. Update training, documents and participants when relevant. Urgent safety actions may proceed under specific procedures but still need prompt documentation and reporting.
Assessment move
Use a protocol rather than a list of principles as the centre of ethics revision. Write the social value in one sentence and connect it to the unresolved clinical decision. Test scientific validity by asking whether design, sample, outcome and analysis can answer that question; exposing participants to a study that cannot produce useful knowledge is not ethically neutral.
Review eligibility criterion by criterion and mark whether it protects safety, preserves the estimand or merely simplifies recruitment. For excluded groups, consider both unfair burden and unfair denial of relevant evidence. Turn consent into a process map containing approach, information, questions, capacity, voluntary decision, documentation and communication of new information.
Adapt language, time and support to the participants rather than relying on signature alone. Identify power relationships and possible therapeutic misconception, then plan wording that distinguishes research from personalised care. Build a risk register with physical, psychological, privacy, social and group harms, their likelihood and severity, mitigation, monitoring and response owner.
Data planning should specify identifiers, coding, storage, role-based access, linkage, sharing, retention, destruction and breach response. For biospecimens or secondary data, separate use required for the primary study from optional future use and define withdrawal limits honestly. Map governance responsibilities across investigator, site, sponsor, ethics body and any independent monitor.
Rehearse one protocol deviation, unexpected safety signal and amendment: who records it, who assesses seriousness, who is notified and what changes for participants? Use proportionality so safeguards match actual risk while maintaining valuable low-risk research. Include community engagement where collective experience, trust or group identity affects design, without treating engagement as a substitute for individual consent.
For AT2, place the most important protection beside the method it governs rather than adding a generic ethics sentence at the end. Finish by explaining how the protocol would change if a privacy risk, recruitment inequity or new harm appeared. Ethical reasoning becomes convincing when it produces an operational decision and a traceable responsibility.
