The fields a board reads hardest.
Applications differ in wording between institutions and hardly at all in substance. A reviewer is checking a small number of things, and the same handful of fields generate most of the conditions. The last item below catches more candidates than the rest combined, because researchers frequently propose to recruit from the setting they work in without noticing what that does to voluntariness.
- Recruitment: who approaches whom, in what words, and why that is not coercive
- Consent: what a participant is told, when, and how they decline without any cost
- Risk: named honestly, with a mitigation attached to each rather than to the study at large
- Data: where identifiable material lives, who may open it, and when it is destroyed
- Withdrawal: what happens to a participant who stops, and to the data already collected
- Vulnerability: whether any participant is a student, a patient or an employee of the researcher
The revision letter is a list of conditions, in an order.
A first submission rarely clears, and the letter that follows is not a judgment; it is a set of conditions expecting a reply. What decides whether approval arrives in weeks or in months is how that reply is written, and the common mistake is to revise the protocol carefully while answering the board loosely.
Each condition is therefore addressed in the order the board raised it, with the amended wording quoted and its new location named, so a reviewer confirms a point instead of hunting for it. Reordering the list into a more comfortable sequence reads as evasion. Where the study then has to be written up, the long-form column takes it from approval onward.
What this column will not do.
It will not invent data, and it will not compose a study whose author could not describe it under questioning. A protocol exists to be defended by the person who signed it, and there is no honest way to prepare somebody for a question about their own sample size except by involving them in the decision that produced it.
Nor is a date promised. Timing belongs to the board and to its meeting calendar, and a service naming one is describing something it does not control. What can be influenced is how many rounds a submission takes, which is where the months go. The research desk holds the wider study work around it.
Send the study and the form
Research questions, the proposed design, the board's current application and any letter already received. Read at no charge and answered within two hours.
Composed inside the application
Every field is answered in the board's own vocabulary, with consent documents, instruments and permissions assembled as a single submission.
Conditions answered in order
Each item in the letter is addressed in the sequence it was raised, with the change quoted and located, until the approval posts.
Questions to the desk.
Can a whole protocol be drafted here?
Yes, working from your research questions, your setting and your proposed design, and written directly into your board's current application rather than adapted into it afterwards. Consent documents, recruitment materials and instruments are built alongside it so the submission arrives complete. Every methodological decision remains yours to approve, since you are the person who will have to defend it.
Only the revision letter needs answering. Is that a request you take?
It is a frequent one and a sensible one. The letter, the submitted application and the board's published guidance are read together, then each condition is answered in the order it was raised, with amended wording quoted and its location named. Replies built that way are quicker for a reviewer to verify, and quick verification is usually what shortens a second round.
Are consent forms written as well?
They are built with the protocol rather than after it, since a board reads both together and checks that one account matches the other in every particular: procedures described identically, risks listed in the same sequence, withdrawal handled the same way, data held by the same people. Consent is pitched at a level a participant could genuinely follow, which is not the register of a methods chapter.
What about CITI training?
The certification itself is completed by the researcher, since it is a personal training record and cannot honestly be anything else. What is covered is confirming the requirement is met before filing, that every listed investigator holds a current certificate, and that the protocol reflects the standards the training sets out. A lapsed certificate is a common and entirely avoidable delay.
How long does approval take?
That belongs to the board and its meeting schedule, and no honest service names a date. What can be influenced is how many rounds a submission takes, since each round costs weeks rather than days. An application complete at first filing and a reply a reviewer can verify quickly are the two things that move that count, and both are what this column produces.
