Clearing the review board on the first submission.

The mismatches that send work back

Almost every returned protocol contains at least one of these. The consent form promises anonymity while the method collects identifiable data. The recruitment script implies that participation is expected. The instrument in the appendix is not the one described in the methods section. The data plan says a personal laptop without saying encrypted. The risk section says none, where a board expects a candid paragraph about foreseeable discomfort.

None of these are ethical failings. They are documents that disagree with each other, and the cure is reading your own packet as a single set rather than as separate forms written on different evenings.

Get the review level right before anything else

This decision costs or saves more time than any other. Boards typically operate exempt, expedited and full-board routes, and they differ enormously in speed: a determination can return in a fortnight while a full-board review waits for a scheduled meeting, which can put six weeks between submissions.

The route is decided by what you are actually doing — the population, the sensitivity of the data, whether participants are identifiable, and whether anyone involved is in a vulnerable category. Answer the determination questions carefully and honestly. Requesting a lower level than your study warrants does not speed anything up; it produces a return and then the higher review anyway.

• ✦Read your board's own guidance before drafting, not after
• ✦Answer determination questions honestly — the wrong level costs weeks
• ✦Make the consent form promise exactly what the method delivers
• ✦Put the real instrument in the appendix, matching the methods text
• ✦Describe data handling concretely: where, encrypted how, destroyed when

Write consent as though a participant will read it

Because one will. Consent documents fail readability expectations constantly, arriving in the register of a research proposal rather than of a letter to a person with no background in your field.

Say what will happen to them, how long it takes, what could go wrong, what happens to their data, and that they may stop at any point without consequence. Short sentences. No jargon that a participant would have to look up. Boards read this document more carefully than any other part of your submission.

The response matrix that ends the cycle

When a board writes back, most students rewrite the protocol and resubmit it. Reviewers find this maddening, because they must now search a changed document to discover whether their points were addressed.

Do this instead: reproduce every comment in the board's own order, put your response directly under each one, and cite the page and line where the change appears. It converts a re-review from an investigation into a checklist, and it is where most of the saved cycles come from.

Quality improvement or research, and why it decides everything

For DNP students and applied doctoral candidates this is the first question, not a detail. If a project is quality improvement intended to change practice locally, many institutions treat it differently from research intended to produce generalisable knowledge, and the route through the board is not the same.

The distinction turns on intent and dissemination, and institutions define it differently, so do not settle it from a textbook or from what a classmate did. Ask your board directly, in writing, early. It occasionally changes the design, which is fine in month one and very expensive in month six.

Timing, and the mistake that cannot be repaired

Build board review into your project timeline as a real phase with real weeks, not as a formality between finishing the proposal and starting collection. A single returned protocol can move your data collection by two months and your defense by a term.

And the one genuinely unrepairable error: do not collect anything before you have your determination in writing. Data gathered before approval usually cannot be used, no matter how good the study is, and there is rarely a way back. Where you want the packet built properly the first time, that is the whole of this art.

Tell the guild your tale.

Name the course, the program and the date it is due. A mentor credentialed in that subject reads it and replies with counsel and a figure within two hours. The reading costs nothing, and sometimes the answer is an honest no.