It's a common misreading of "risk-proportionate review" to assume it means less scrutiny across the board — a lighter-touch process that trades rigour for speed. The actual principle is narrower and more useful than that: the depth and nature of scrutiny should match the actual risk a study poses, which sometimes means a faster, simpler process for genuinely low-risk research, and sometimes means a more thorough one for higher-risk research that a generic process would otherwise under-examine.
The problem
Submissions get built to a single template regardless of the study behind them, and that habit fails in both directions. A short, fully anonymous survey with no sensitive data and no vulnerable population doesn't need documentation designed for a fundamentally different risk profile — over-engineering a low-risk submission doesn't make it safer, it just makes it slower to review, without adding any genuine protection for participants who were never meaningfully at risk.
The opposite mistake is more consequential. Treating a study as low-risk based on its category — "it's just a survey," "it's just an interview study" — without examining whether the specific population, topic, or data involved carries real risk is a common, avoidable failure. An interview study asking about deeply sensitive personal experiences, or a survey collecting data from a genuinely vulnerable population, can carry real risk despite superficially resembling a routine low-risk design.
The challenge
A one-size template can't reflect wildly different risk profiles, so the practical approach is to assess risk honestly and specifically before deciding how to structure the submission, rather than inferring the appropriate level of documentation from the study's general type. What data is being collected, from whom, using what method, and what could realistically go wrong — answering these questions concretely produces a much more reliable basis for the submission than a category label ever will.
A useful sanity check is to ask, for any given safeguard or piece of documentation, what specific risk it's actually addressing. If the honest answer is "none in particular, but it seemed safer to include," that's a sign of over-engineering. If the honest answer reveals a risk that isn't addressed anywhere else in the submission, that's a sign something is missing. A reviewer is looking for evidence that the risk assessment itself was done thoughtfully, not for a submission that simply matches a template — and getting the balance wrong in either direction tends to generate the kind of follow-up questions that slow a submission down rather than speed it up.
The solution
Matching depth to risk, not habit, comes down to three practical commitments:
- Match the depth of your submission to actual, assessed risk — not to the study's category or a generic template.
- Don't under-document just because a study looks low-risk on the surface. A routine-sounding design can still involve a sensitive topic or a vulnerable population.
- Don't over-engineer safeguards a low-risk study doesn't need. Padding a submission with unnecessary documentation slows review without protecting anyone.
It's worth stating this reasoning explicitly in the submission itself, rather than leaving a reviewer to infer it — a short section explaining why the documentation provided is proportionate to the specific risks identified turns an implicit judgement call into an explicit, checkable claim. And it's worth revisiting: a risk assessment made at the design stage isn't the last word on a study's risk profile. New information can emerge once recruitment starts, or a protocol amendment can shift the balance in either direction — so treat proportionality as a living judgement, not a box ticked once at first review.
