The elements, in the order a participant meets them.
Boards differ in wording and scarcely at all in substance, because the required elements of informed consent come from the regulation a board applies rather than from local preference. A participant encounters them in a settled order, and a document presenting them in that order reads as competent before anybody has assessed a word of its content.
The final pair of elements is more than administrative. A participant with a complaint about the conduct of a study has to be able to reach somebody who is not the researcher, and a form offering only a doctoral student's own address has quietly removed the safeguard that element exists to provide. Whether a population is reachable at all is also the first thing that decides whether a study can be finished, which the note on topics takes up.
- A statement that this is research, framed as an invitation rather than as an instruction
- The purpose, the expected duration, and what taking part involves, step by step
- Foreseeable risks and discomforts, named individually rather than gathered under one word
- Benefits, to the participant or to others, described without inflation
- Where identifiable information lives, who may open it, and when it is destroyed
- Compensation, if any, and what becomes of it if somebody stops partway through
- That taking part is voluntary, that refusal costs nothing, and that withdrawal is possible
- Whom to contact about the study, and separately, whom to contact about participant rights
The mirror rule, which prevents most of the returns.
Whatever the consent document states must also be what the protocol states, in the same numbers and as nearly as possible the same words. Where the protocol describes a forty-minute interview and the consent form says thirty, a reviewer does not read a typing slip; they read evidence that one of the two documents was written from memory. The same applies to the number of sessions, the recording arrangements, the retention period, the compensation and every named risk.
Reviewers check the pair against each other precisely because inconsistency is cheap to find and tends to predict deeper problems. Before submitting, read the two documents side by side and confirm that every procedural fact appears identically in both. It is a dull hour, and it removes one of the commonest causes of a month's delay.
Readability is a requirement, not a courtesy.
A consent document is written for the population being recruited rather than for the committee reading it. A form dense with methodological vocabulary has failed its purpose even where every required element is present, because a participant who cannot follow the description has not given informed consent in any meaningful sense. Short sentences, ordinary words, direct address, and a definition wherever a technical term genuinely cannot be removed.
Special populations change the document rather than merely lengthening it, and boards examine those changes closely because that is where consent stops being routine.
- Write to the reading level of the people being recruited, not to the committee
- Define any technical term you cannot remove, in the sentence where it first appears
- Pair parental or guardian permission with an assent form written for the child's age
- Ask separately for permission to record, and say what becomes of the recording afterward
- Supply a translated version wherever recruitment reaches people who do not read English
The sentences that get a form returned.
Certain phrases are flagged almost reflexively, because they promise what no researcher controls. Complete confidentiality is one: a transcript can be compelled, and a small sample can identify itself to anybody who knows the setting, so the honest formulation describes the measures taken and admits the limits that remain. There are no risks is another, since there is nearly always a risk of discomfort, of lost time, or of a subject arising that a participant would rather not discuss.
The third pattern is subtler and costs more. Where a researcher recruits from their own workplace, classroom or caseload, the form has to say plainly that taking part has no bearing on employment, grade or care, and the recruitment process should put a step between the researcher and the decision. Voluntariness is the central question a board is asking, and a document that ignores the researcher's own authority over its participants has answered it badly. Where the documents rather than the study are the obstacle, the protocol column drafts them to the board's own template.
Questions to the desk.
Does every study need a signed consent form?
Not always. Boards can waive documentation of consent for certain minimal-risk studies, most commonly anonymous surveys, where signatures would be the only record linking a person to the data. In those cases an information sheet carrying the same elements usually replaces the signature page. The decision belongs to the board rather than to the researcher, so the request is made in the application and justified there.
What is the difference between assent and consent?
Consent is given by a person legally able to authorize their own participation. Assent is the agreement of somebody who cannot, most often a child, and it accompanies rather than replaces permission from a parent or guardian. An assent document is written in language the participant can actually read, and a child who declines is not enrolled even where a parent has agreed.
Can consent be collected on a web page instead of paper?
Commonly, yes, and boards see it constantly. The elements do not change; only the mechanism does. The usual arrangement presents the full information first, requires an affirmative action to proceed, offers a version the participant can keep, and prevents anyone reaching the questions without passing through it. Say in the protocol exactly how the platform records that agreement.
Does CITI training cover writing the consent form?
It covers the principles behind the document and the history that produced them, which is why institutions require it before submission. It does not draft the form or teach a particular board's template, both of which are local. Complete the training early, since a certificate is frequently a precondition for submitting at all, then write the document against your own board's current template.
