Step 2 CK · Ethics, Biostatistics & Patient Safety

Research Ethics & Human Subjects Protection

Name: Research Ethics & Human Subjects Protection
Creator: Veli Bakalov, MD

Consent, equipoise, and oversight: the guardrails that turn patients into protected participants

Ethics, Biostatistics & Patient Safety·Ethics·5 min read·9 high-yield points·Reviewed by Veli Bakalov, MD @HemeOncBuddy

Study materialsNo account needed

Card deckActive-recall cards · quick checks · quiz

High-yield

Belmont principles: respect for persons (informed consent), beneficence (favorable risk-benefit balance), and justice (fair selection of subjects).
No participant may be enrolled before IRB approval, and protocol changes need IRB approval before implementation unless they remove an immediate hazard.
Research participants may withdraw at any time without penalty or any effect on their usual care.
Correct the therapeutic misconception by explaining randomization and placebo; payment is for time and inconvenience and must not be large enough to be undue inducement.
Children need parental permission plus the child's assent, and a child's dissent to non-beneficial research is respected; adults lacking capacity enroll through a legally authorized representative.
The emergency exception from informed consent requires a life-threatening condition, infeasible consent, prior community consultation, a DSMB, and informing the family as soon as possible.
Randomization requires clinical equipoise; placebo is acceptable only when no proven therapy exists or withholding it causes no serious or irreversible harm.
Research misconduct is fabrication, falsification, or plagiarism and is reported through institutional channels such as the research integrity officer.
Use identifiable health data only with participant authorization or an IRB/privacy board waiver; a Certificate of Confidentiality protects research data from subpoenas.

Foundations: the Belmont principles and the IRB

Where the rules came from. The Nuremberg Code (1947) made voluntary consent essential, and the Declaration of Helsinki added international standards such as limits on placebo use. In the United States, exposure of the Tuskegee syphilis study (1932–1972), which withheld penicillin from Black men with syphilis for decades after it became standard therapy, led to the Belmont Report (1979) and the federal Common Rule.

Belmont principles

Principle	Meaning	Application
Respect for persons	Treat people as autonomous agents; protect those with diminished autonomy	Informed consent
Beneficence	Maximize possible benefits, minimize harms	Favorable risk-benefit assessment
Justice	Share the burdens and benefits of research fairly	Fair subject selection; no exploiting convenient or disadvantaged groups

The institutional review board. An IRB has at least 5 members with varied backgrounds, including at least one scientist, one nonscientist, and one member unaffiliated with the institution. It must approve the protocol, consent documents, recruitment materials, and payment plan before any participant is enrolled, and it continues to review amendments, adverse events, and progress. Changes to an approved study require IRB approval before they are implemented, except when needed to eliminate an immediate hazard to participants.

Informed consent for research

Required content. Research consent explains the purpose, procedures, and duration of the study; foreseeable risks and discomforts; possible benefits; alternatives, including standard care outside the study; how confidentiality will be protected; whom to contact with questions; and that participation is voluntary. A participant may refuse or withdraw at any time without penalty and without any effect on usual care.

Therapeutic misconception. Participants often assume that research procedures are chosen for their individual benefit. Investigators must explain that assignment is random, that the participant may receive placebo, and that the study's purpose is generalizable knowledge. When a treating physician recruits his or her own patient, having a neutral team member obtain consent limits undue influence. A phrase such as "I'm enrolling you because this is the best treatment for you" is a red flag.

Payment. Reasonable payment for time, travel, and inconvenience is acceptable. Coercion is an overt or implied threat of harm for refusing; undue inducement (undue influence) is an offer so large that it distorts judgment about risk. Payments are usually prorated across visits rather than withheld until completion, and payment is never counted as a benefit when an IRB weighs risks.

Vulnerable participants and emergency research

Added protections

Group	Safeguards
Children	Parental permission plus the child's assent once developmentally able, often from about age 7; a child's refusal of non-beneficial research is respected
Adults lacking capacity	Consent from a legally authorized representative; the participant's assent is sought and any dissent honored
Prisoners	Research limited to minimal-risk or prisoner-relevant questions; a prisoner representative sits on the IRB; participation cannot affect parole
Pregnant women	The woman consents for herself; fetal risk must be minimal unless the research offers direct benefit to the woman or fetus

Protection is not exclusion: keeping pregnant women out of research without good reason is itself unjust, because it leaves pregnancy without evidence.

Emergency research. The FDA's exception from informed consent (EFIC) allows enrollment without prospective consent in narrowly defined circumstances, such as trials in cardiac arrest or major trauma:

Life-threatening conditionAvailable treatments are unproven or unsatisfactory.
Consent not feasibleThe patient is incapacitated, and no representative can be reached within the therapeutic window.
Prospect of direct benefitThe intervention could help the enrolled patient.
Community safeguardsCommunity consultation and public disclosure before the trial starts, with an independent DSMB.
Prompt notificationThe patient, representative, or family is informed as soon as possible and may withdraw at any time.

Ethical trial conduct

Clinical equipoise. Randomization is ethical only when the expert community is genuinely uncertain which arm is better. If one treatment is already known to be superior, randomizing patients knowingly assigns some of them to inferior care.

Placebo controls. Placebo is acceptable when no proven therapy exists or when withholding proven therapy risks no serious or irreversible harm (for example, short trials of symptomatic treatment for allergic rhinitis). When effective therapy exists for a serious condition, such as HIV infection, diabetic ketoacidosis, or anticoagulation for a mechanical valve, the new drug is compared with standard care or added on top of it (add-on design).

Monitoring. An independent data and safety monitoring board reviews unblinded interim results and, following prespecified stopping rules, recommends ending the trial for clear benefit (continuing would deny the control arm a proven therapy), unacceptable harm, or futility. Investigators must report serious unanticipated problems to the IRB promptly.

Research integrity and data protection

Conflicts of interest. Financial interests such as equity, consulting fees, or royalties must be disclosed to the institution, the IRB, journals, and, when relevant, participants. The institution then manages them (for example, independent monitoring or divestment).

Research misconduct. Misconduct means fabrication (inventing data), falsification (manipulating materials, processes, images, or data, including selectively dropping data points, so the record misrepresents the research), or plagiarism (using others' ideas or words without credit). Honest error and differences of opinion are not misconduct. Suspected misconduct is reported through institutional channels, usually the research integrity officer, rather than by confronting the colleague publicly, quietly fixing the data, or going to the press; institutions protect good-faith whistleblowers.

Protecting identifiable data. Collect only the identifiers the study needs, code or de-identify data (the HIPAA Safe Harbor method removes 18 identifiers), encrypt files, and restrict access. Protected health information may be used for research with the participant's written authorization or with a waiver from an IRB or privacy board when the research poses minimal privacy risk and authorization is impracticable. A Certificate of Confidentiality shields identifiable research information from compelled legal disclosure, such as subpoenas.