State strengths, flag issues with evidence-based reasons, suggest fixes, keep a respectful tone, and follow journal, COPE, and ICMJE guidance.
Why Peer Review Feedback In Medicine Needs Care
Peer review shapes what clinicians read and what patients receive. A short note can shift an editor’s choice, redirect a study, or spark better reporting. Good feedback also keeps trust in the literature. That calls for accuracy, fairness, and clear language that helps both editors and authors act. For ethics and conduct, see the COPE reviewer guidelines and the ICMJE Recommendations.
Section-By-Section Review Checklist
| Manuscript Section | What To Check | Quick Prompts |
|---|---|---|
| Title & Abstract | Match outcomes and design, no overclaiming, key numbers present. | Does the title fit the data? Are effect sizes in the abstract? |
| Keywords | Searchable terms, indexing fit, no marketing phrases. | Would a clinician find this with common terms? |
| Introduction | Clear question, gap, rationale, and prespecified aim. | Can you state the PICO or main aim in one line? |
| Methods: Design | Study type fits question; prospectively set plan; protocol access. | Is the design named and linked to a protocol or registry? |
| Methods: Participants | Eligibility, setting, recruitment dates, consent. | Who was in, who was out, and when? |
| Methods: Interventions/Exposures | Precise description, dose, timing, comparator. | Could a clinician replicate the exposure or care path? |
| Methods: Outcomes | Primary and secondary outcomes defined before data peek. | Is the main outcome clear and clinically meaningful? |
| Methods: Sample Size | Calculation shown, target met or shortfall explained. | Was power planned and achieved? |
| Methods: Randomisation/Blinding | Sequence, concealment, who was masked. | Could allocation be predicted? |
| Methods: Statistics | Plan matches outcome types; checks for assumptions; handling of missing data. | Do analyses align with the protocol and data level? |
| Ethics & Registration | IRB/REC approval, consent, trial or review registration. | Is a registry ID present for trials or reviews? |
| Data & Materials | Availability statement, code sharing when possible. | Can editors or readers reach the data or code? |
| Results: Flow | Numbers at each stage, reasons for loss to follow-up. | Is there a diagram or clear flow? |
| Results: Baseline | Balance across groups, key demographics and illness severity. | Any imbalances that need adjustment? |
| Results: Outcomes | Effect size with precision, clinical relevance, harms captured. | Are CIs and absolute numbers shown? |
| Results: Subgroups | Prespecified only, interaction testing not cherry picked. | Were subgroups set in advance? |
| Results: Figures/Tables | Readable labels, units, no duplication across panels. | Would a busy reader grasp the figure in seconds? |
| Discussion | Answer matches data, limits owned, no new claims. | Do authors link back to the stated aim? |
| Generalisability | Setting and population limits stated. | Where would the findings not apply? |
| Conclusions | Calibrated language and next steps grounded in data. | Do claims match the study design and size? |
| References | Current and fair coverage, correct primary sources. | Any self-citation padding or missing key work? |
Giving Peer Review Feedback In Medicine: The Core Moves
Start fast. Skim the manuscript, figures, and tables. Read the journal’s aims and reviewer notes. Check the registry or protocol if cited. Pull the right reporting checklist from the EQUATOR Network that matches the design. That prep keeps your review aligned with the journal and the study type.
Open with a short, neutral summary in your own words. One or two lines will do. Then list two to three clear strengths. Next, lay out the main issues that block acceptance. Tie each point to a line, table, figure, or registry entry. Offer a practical fix when you can. Use section headings so editors can scan.
Close with a brief overall view for the editor. Say whether the study answers a useful question for the journal’s readers and whether the methods and reporting can be improved within a normal revision. Keep any policy notes or possible ethics flags in the confidential box.
Prepare With The Right Sources
Match the review to standards. For trials and many cohort studies, confirm registration and timing. Use CONSORT, PRISMA, STROBE, or another checklist in the EQUATOR library. If sex and gender reporting matters for the topic, look for clarity on terms and analytic plans. This keeps your notes concrete.
Set The Right Scope
Your task is the science and the reporting. Skip copyediting beyond a few targeted lines. Avoid judging novelty claims; editors handle that. If a method or analysis lies outside your skills, say so in the confidential box. Suggest a statistician or a field expert if needed.
Open With Gratitude And A Summary
A friendly first line sets the tone. Thank the authors for the work and the editor for the invite. Then give a one-line summary of the research question and design. This helps confirm you read the paper closely and keeps your later points grounded.
Use A Strengths-First Structure
List concrete positives first. Common wins include a strong clinical question, a prespecified plan, clear figures, or patient-relevant outcomes. Naming strengths builds trust and makes tougher notes easier to accept.
Point Out Issues With Evidence
Move next to the main issues. Use numbered bullets. Anchor each note to the text. Quote a line or point to a figure panel. Give a short reason, then a fix. Keep the tone calm. Avoid judging the authors; stick to the work.
Typical high-impact points include outcome switching, unclear eligibility, underpowered analysis, missing harms, model overfit, unvalidated scales, and claims that reach past the design. Policy flags include consent gaps, data reuse without permission, duplicate posting, and undisclosed funding ties.
Rapid Checks By Section
- Title/Abstract: name the design and show the key numbers.
- Methods: show the plan, timing, and sample size logic.
- Results: report counts and denominators, plus precision.
- Discussion: use balanced language and own limits.
How To Provide Peer Review Feedback In Medical Journals Without Bias
Bias creeps in fast. Names, institutions, and country can sway judgment. Read the work first, then the cover page. If blinding is partial, pause before drafting. Ask whether a tie or rivalry might color your view. If so, tell the editor and step aside.
Watch your language. Avoid labels about the authors. Write to the work. Use people-first terms where clinical topics involve patients. If the paper studies sex or gender, ask for clear definitions and analytic plans that match those terms. Keep identity out of the verdict.
Structure Your Report For Editors
Editors read fast. Help them. Split comments into Major and Minor. Major points change decisions or core claims. Minor points tune clarity, labels, or small fixes. Put policy or conduct notes in the confidential box only. Never contact the authors outside the system.
End with a short, steady line on next steps, such as “publish after major revision,” “publish after minor revision,” or “reject.” Many systems capture this as a form tick. Keep the public comments free of that tag; that line is for the editor only.
Ethics, Consent, And Data Checks
Peer review guards patient rights and the record. Confirm ethics approval and consent for human work, or justified waivers. For trials, look for a registry ID and prespecified outcomes. For systematic reviews, look for a registered plan. Ask for data and code access when the journal supports sharing. The ICMJE section on reviewer duties outlines prompt replies, conflicts, and polite, constructive notes.
If a red flag arises, outline the facts and quote the lines. Do not accuse. Ask the editor to advise. Keep the case private. Never use ideas or data seen in review for your own work.
Timing, Confidentiality, And Professional Conduct
Reply to invites fast. If you accept, meet the due date or ask for an extension early. Keep all files secure. Do not share with trainees unless the editor grants permission and you add their names. Delete local copies after the decision, per journal policy.
Tone travels. Short, plain sentences help. So do headings, lists, and numbered points. Skip sarcasm. Praise in public comments; reserve sensitive points for the editor box. End with thanks.
Common Scenarios And Ready-To-Use Lines
Use these short lines when you need a prompt. Tailor as needed. Keep each line tied to the text or data. For more reporting support across designs, use the EQUATOR reporting guidelines search.
Phrase Bank For Frequent Review Goals
| Goal | Say This | Why It Helps |
|---|---|---|
| Ask For Registration Proof | Please add the registry ID and confirm that outcomes were prespecified before enrolment. | Ties claims to a dated plan. |
| Clarify Eligibility | Kindly expand the inclusion and exclusion criteria with dates and settings. | Helps readers judge fit. |
| Power Shortfall | The sample appears underpowered for the main outcome; please add the calculation and discuss impact. | Brings precision to claims. |
| Outcome Switching | The abstract highlights an outcome that does not match the prespecified primary outcome; please align text or explain. | Protects against spin. |
| Missing Harms | Please report adverse events with counts and denominators for each arm. | Balances benefits and risks. |
| Model Overfit | The model includes many predictors for few events; a simpler model or penalised method may be safer. | Reduces false signals. |
| Data Sharing | Please add a data availability statement and, if allowed, a link to a repository. | Supports reuse and checks. |
| Language Tone | A light edit by a fluent speaker would help readers follow the results. | Keeps scope on clarity. |
| Figure Clarity | Please enlarge axis labels and ensure units appear on every panel. | Aids fast reading. |
| Causal Claims | Word choices imply causation in an observational design; please rephrase to reflect association. | Aligns claims and design. |
When To Say No To A Review
Decline when you lack time, when the topic sits outside your skills, or when a tie or rivalry could bias you. Suggest a colleague if the editor requests names, and disclose any links. Fast declines help editors keep papers moving. Many journals adopt the COPE stance on conflicts and respectful conduct.
Final Polish: A Quick Pass Before Submission
Reread your review once end to end. Trim repetition. Check that each major point links to a line or figure. Make sure your tone stays respectful. Verify that you used the journal’s score sheets and forms. Then submit.