Informed Consent, Confidentiality and Honesty in Reporting Data

Concept

Informed consent is a participant’s voluntary agreement to take part after being told what the research involves. It is a process, not just a signature on a form. Its elements are disclosure (purpose, procedures, duration, risks, benefits, alternatives, confidentiality arrangements and whom to contact), comprehension (language and wording the person understands, in their own language), voluntariness (no coercion or undue influence), competence (capacity to decide) and authorisation (signature, thumbprint or witnessed agreement). The information is given in a participant information sheet and recorded in a consent form. Participants may withdraw at any time without penalty or loss of benefits, and the form must not ask them to give up legal rights. In India the consent process in regulatory clinical trials is audio-visually recorded, and participants who suffer trial-related injury are entitled to free medical care and compensation under the New Drugs and Clinical Trials Rules, 2019.

Concept

Where a person cannot legally consent, a legally authorised representative (parent, guardian or other) gives permission, and the participant’s own assent is sought as far as they can understand. In the ICMR 2017 guidelines, children aged about 7 to 12 years give oral assent and those aged about 12 to 18 years give written assent, in addition to parental consent. For people who cannot read, an impartial witness is present during the consent process. Participants with mental illness, pregnant women, prisoners, students and employees need extra protection because of limited autonomy or power relationships. When new information emerges that may affect willingness to continue, participants are informed and consent is renewed.

Concept

An ethics committee may waive informed consent for minimal-risk studies when obtaining consent is impracticable, such as research on anonymised records. Reasonable compensation for time, travel and inconvenience is acceptable, but a payment so large that it clouds judgement is undue inducement. Coercion is the threat of harm or penalty for refusing, as when a teacher requires students to take part or a doctor implies that care depends on participation. Exploitation means taking unfair advantage of vulnerable people. The therapeutic misconception occurs when participants wrongly believe that the research is designed for their personal benefit. In community studies, permission from community leaders does not replace individual consent.

Concept

Anonymity means that nobody, including the researcher, can link data to a participant. Confidentiality means that the researcher knows the identity but promises not to disclose it. Privacy is a person’s right to control access to personal information and to be free from intrusion. Techniques include removing direct identifiers, assigning codes (pseudonymisation) with the key stored separately, encryption, password protection, locked storage and limited access. Small samples or unusual details can make a person re-identifiable even without a name, so reports should aggregate data and alter identifying details. Confidentiality has limits: researchers may have to disclose when someone is at risk of serious harm or when the law requires it, and participants should be told so in advance. In focus groups, confidentiality cannot be guaranteed because other participants hear the discussion, so ground rules are agreed.

Concept

In India the Digital Personal Data Protection Act, 2023 governs the processing of digital personal data. It speaks of the data principal (the individual) and the data fiduciary (the entity deciding the purpose and means of processing), requires notice and consent for stated purposes, limits use to those purposes, and gives principals rights to access, correction and erasure. It provides conditional exemptions for research, archiving and statistical use when safeguards are in place, and it requires verifiable parental consent for children; check the status of its implementing rules. Earlier, the Information Technology Act, 2000 and its rules covered sensitive data. Abroad, the EU GDPR (2018) and US HIPAA (1996, for health information) are leading models. Secondary use of data should stay within the scope of the original consent or be approved by an ethics committee, and publicly posted online material still raises privacy and terms-of-service questions.

Concept

Honest reporting means describing the methods, the results and the limitations accurately, including negative and non-significant findings. It excludes fabrication and falsification, and it also excludes subtler distortions: selective reporting of outcomes, cherry-picking of data, p-hacking (trying analyses until one is significant), HARKing and unplanned post-hoc subgroup analysis presented as planned. Publication bias (the file drawer problem) arises when only positive results are published, and outcome-reporting bias when only some measured outcomes are reported. Images must be adjusted uniformly and any adjustment disclosed. Reporting guidelines improve completeness: CONSORT for randomised trials, PRISMA for systematic reviews, STROBE for observational studies, ARRIVE for animal studies and COREQ or SRQR for qualitative work, catalogued by the EQUATOR Network.

Concept

Pre-registration means recording the hypotheses, design and analysis plan in a time-stamped public registry before collecting data, so that later choices cannot be disguised as plans. Registries include OSF Registries, AsPredicted, ClinicalTrials.gov, the Clinical Trials Registry of India (CTRI) and PROSPERO for systematic reviews. In registered reports a journal reviews the protocol first and commits to publishing the paper whatever the result. Transparency also covers sharing data, code and materials, following the FAIR principles and the Transparency and Openness Promotion (TOP) guidelines, and including a data-availability statement. The reproducibility crisis, highlighted when a large replication project in psychology (2015) reproduced fewer than half of the studies, drives these reforms. Sharing must still respect consent and confidentiality, so the aim is data “as open as possible, as closed as necessary”.

Concept

When an error is found after publication, the proper response is a published correction. An erratum corrects an error made by the publisher, and a corrigendum corrects an error made by the author; both leave the paper standing when the conclusions are unaffected. An expression of concern is a notice from editors that questions about a paper are under investigation. A retraction withdraws a paper because of serious error or misconduct (fabrication, plagiarism, duplicate publication or unethical research); the paper remains accessible but is marked as retracted, following COPE guidelines. Retraction is not always a finding of fraud, since honest error can lead to it, and researchers who discover their own honest error are expected to report it promptly to the journal.