Build a reviewer shortlist for a specialised clinical manuscript

1. Translate the manuscript into review questions

Imagine a fictional observational neurology cohort comparing follow-up outcomes between two treatment groups. Before searching for names, list the assessment questions: is the outcome clinically interpretable, is treatment selection adequately described, are missing follow-up observations handled transparently, and does the causal language exceed the design? These are review prompts, not findings about an actual study.

Split the needs into clinical context and methods. Someone may understand the disease well but have limited experience with the analysis used; another person may assess confounding and missingness while needing help with clinical interpretation. Record that division explicitly. A shortlist with several similar specialists can still leave a methods question unanswered.

On small screens, scroll the table horizontally, or focus it and use the arrow keys.

2. Keep the selection decision with the editor

ICMJE describes peer review as expert critical assessment and places responsibility for suitable reviewer-selection systems with journals. It also addresses confidential manuscripts, disclosure of relationships and timely reviewer responses. The editor must assess the declarations and the usefulness of the review; an invitation does not transfer the editorial decision.

The remaining worksheet is a proposed way to operationalise that responsibility, not an ICMJE scoring system. Use the journal’s actual conflict and review policies. Do not invent a universal number of reviewers or a fixed conflict look-back period merely because a spreadsheet needs a value. Where the policy leaves a question open, record the editor’s decision and rationale.

Reference: ICMJE: responsibilities in submission and peer review

3. Establish identity before trusting a suggested address

For this workflow, match a candidate’s name to relevant work and an independently located institutional or professional profile. Check whether the contact route belongs to that person. An author suggestion may be a starting point, but a supplied email address and a matching display name are not independent confirmation of each other.

Keep only the information needed for editorial selection in the approved editorial workspace. If a profile is stale or a same-name author cannot be distinguished, use “identity unresolved” and seek clarification. Avoid turning an unanswered verification step into an accusation. Neither a prominent institution nor a personal email address settles competence or authenticity on its own.

4. Use the matrix to expose uncertainty, not to average it away

The four candidates below are invented roles. Candidate A is clinically relevant but declares a relationship requiring editorial assessment. B can address methods, with limited disease-specific experience. C has unresolved identity. D appears relevant, but availability has not been established. None has accepted an invitation.

Do not add a high expertise score to a low identity score and call the total acceptable. Some questions need resolution before an invitation: who is being contacted, what they can assess and whether a relationship prevents independent review under the journal’s policy. A methods reviewer’s limited topic scope can be stated openly and complemented by another suitable perspective.

On small screens, scroll the table horizontally, or focus it and use the arrow keys.

5. Make the requested contribution and response state explicit

Once the editor selects a candidate, the invitation should explain the assessment sought, the journal’s confidentiality and conflict requirements, and the requested timetable. Use the approved editorial system and disclose manuscript material only as permitted. Do not pass a confidential submission to an outside tool or colleague simply to make candidate selection easier.

Track “shortlisted”, “invited”, “accepted”, “declined” and “report received” independently. If D has been invited but has not answered, clinical coverage remains unconfirmed. If B accepts a methods-only scope, the acceptance does not promise a clinical assessment. Record any agreed change in deadline rather than silently keeping the original date as though it were still confirmed.

6. Reassess coverage when reports arrive

In the fictional continuation, B submits a methods report and D declines. The editor now has methods input but no completed clinical review. The useful action is to revisit the uncovered questions and choose a suitable next reviewer, not to label the process complete because two invitations were sent.

If a report goes outside its reviewer’s stated expertise, the editor can request clarification or additional assessment of the disputed point. Keep the selection rationale, declarations, accepted scope and reports together so the decision can be understood later. The table supports editorial judgement; it neither measures review quality automatically nor guarantees acceptance, speed or a particular outcome.

Questions for a practical final check

• List the manuscript’s clinical and methods questions.
• Match identity, relevant work and a reliable contact route.
• Record declared relationships for editorial assessment.
• State each reviewer’s requested scope.
• Keep invitation, acceptance and report states separate.
• Reassess missing expertise before concluding the review.