What are the types of peer review in biomedical journals?

The four main peer review types are: single-blind (authors don't know reviewer identities; most common in biomedicine), double-blind (both parties anonymous; used by Nature Medicine, JAMA, PLOS ONE and others), open review (identities disclosed to both parties; used by BMJ, eLife, EMBO), and post-publication review (formal review after online publication). Within these, some journals use cascading review (sharing reviews with other journals after rejection) and registered reports (pre-registered methods reviewed before data collection).

What do biomedical peer reviewers actually evaluate?

Most biomedical journal reviewer guidelines ask for assessment across the same core dimensions: originality and significance of the research question, methodological rigor (study design, controls, sample sizes), appropriateness of the statistical analysis, accuracy and clarity of data presentation, appropriate interpretation of results without overclaiming, relevance to the journal's readership, and ethical compliance with reporting standards (CONSORT, PRISMA, ARRIVE etc.). Reviewers also check whether limitations are honestly acknowledged.

What does peer review not catch?

Peer review has well-documented limitations. It rarely catches data fabrication or fraud (reviewers don't have access to raw data and can't re-run experiments). It doesn't reliably catch p-hacking, selective reporting, or underpowered studies. Statistical errors often pass review because most reviewers don't re-analyze the data. It also doesn't verify reproducibility. Peer review filters for plausibility, novelty, and relevance - not absolute correctness.

How long does biomedical peer review take?

It varies significantly by journal and submission quality. High-tier journals desk-reject 70-90% of submissions without external review, often within 1-7 days. Manuscripts sent for full peer review typically receive a first decision in 30-90 days. Journals with fast review pledges (PLOS ONE, Nature Communications) aim for 30-45 days. The full timeline from submission to acceptance often takes 3-6 months when revisions are included. See the peer review timelines guide for journal-specific data.

Scope: 4 review types · full editorial process·Data: Publisher editorial policies + peer review research·Last reviewed: February 2026·Source: Manusights editorial team (researchers with publications in Cell, Nature, Science)·Cite this guide ↓

Peer Review in Biomedicine: Types, Process, and What Reviewers Evaluate

Peer review is the system by which scientific manuscripts are evaluated by independent experts before publication. It's imperfect, inconsistent, and sometimes painfully slow, but it's also the primary mechanism for quality assurance in scientific literature.

This guide covers how it works: from the editor's desk to the reviewer's inbox to your decision letter. That context helps you write manuscripts that survive the process and respond to it effectively. Key sources are cited in the references section below.

Types of Peer Review

Single-blind

Reviewers know who the authors are, but authors don't know who the reviewers are. The most common model in biomedical journals.

Used by: Most Cell Press journals, Nature family, NEJM, Lancet, JAMA

✓ Reviewers can consider author expertise and track record

✗ Potential for reviewer bias based on author institution, gender, or nationality

Double-blind

Neither authors nor reviewers know each other's identities. Authors must anonymize their manuscript: removing author names, affiliations, and self-citations.

Used by: Some BMJ journals, certain behavioral and social science journals, increasing adoption in medicine

✓ Reduces bias based on author identity; may benefit early-career researchers

✗ Hard to maintain true anonymity (recognizable writing style, methods, lab); adds preparation burden

Open peer review

Both authors and reviewers know each other's identities; reviews may be published alongside the article.

Used by: BMJ (publishes reviews on request), eLife, F1000Research, PeerJ, EMBO Journal (published reviews)

✓ Accountability; reduces malicious reviews; makes the review process more transparent

✗ Reviewers may be less candid to avoid professional conflict; early-career reviewers may be reluctant to critique senior authors

Post-publication review

Articles are published first, then reviewed by the community. Used by a small number of journals and platforms (PubPeer, F1000Research).

Used by: eLife's new model (reviewed preprints); F1000Research; PubPeer (community)

✓ Faster publication; ongoing quality assessment

✗ Less rigorous gate-keeping; unreviewed material is publicly attributed to authors

The Peer Review Process: What Actually Happens

Most authors imagine peer review as one continuous black box. In reality it is a chain of separate bottlenecks, and each one can stall for a different reason. That is why a paper can sit for three weeks without anyone actually reviewing it yet.

1Stage

Days 1–5

Submission and desk check

After you submit, a managing editor checks that the manuscript meets basic formatting requirements and isn't obviously out of scope. This takes 1–3 days at most journals. Then the submission goes to an editor who handles your field.

2Stage

Days 5–30, varies widely

Editorial assessment (desk decision)

The handling editor reads the manuscript, or at least the abstract, introduction, and figures, and decides whether to send it for review. At top journals (Nature, NEJM, Cell), over 85% of manuscripts are desk rejected here. This step takes a few days to a few weeks depending on the journal. A desk rejection is not a comment on scientific quality; it usually means scope mismatch or insufficient novelty for that specific journal.

3Stage

1–2 weeks

Reviewer recruitment

If the editor decides to send the paper for review, they identify 2–3 independent experts and invite them. This is harder than it sounds: busy scientists decline frequently. Recruiting reviewers can take 1–2 weeks on its own, which is part of why total review time is longer than people expect.

4Stage

2–6 weeks

Peer review

Reviewers typically have 2–4 weeks to return comments. They evaluate scientific rigor, novelty, significance, and whether conclusions are supported by data. Most reviewers spend 3–8 hours per manuscript. Review quality varies enormously.

5Stage

1–2 weeks after reviews

Editorial decision

The editor reads the reviews and makes a decision: Accept (rare on first submission), Major revision, Minor revision, Reject with invitation to resubmit, or Reject. The editor weighs reviewer comments but isn't bound by them: a reviewer who hates a paper doesn't automatically mean rejection if the editor disagrees.

6Stage

Author-controlled, usually 1–3 months

Revision (if applicable)

If you receive a revision request, you typically have 1–3 months to respond. You revise the manuscript and write a point-by-point response letter addressing every reviewer comment. The revised manuscript usually goes back to the same reviewers.

7Stage

2–12 weeks post-acceptance

Acceptance and publication

After minor revisions or a satisfactory major revision, the editor accepts the paper. Accepted manuscripts then go through production (copyediting, typesetting, proofing) before appearing online. This takes weeks to months depending on the journal.

What Biomedical Peer Reviewers Actually Evaluate

Reviewer evaluation criteria vary by journal, but most biomedical journals ask reviewers to assess the same core dimensions. Understanding these helps you anticipate reviewer concerns before they arrive in your decision letter.

Novelty and significance

Does this advance understanding in a meaningful way? Is the finding new, or confirmatory of something already known? At top journals, this is weighted heavily: technically perfect but incremental work still gets rejected.

Scientific rigor and methodology

Are the methods appropriate for the claims? Are controls adequate? Are statistical analyses correct and reported fully? Sample sizes justified? This is where most rejection-driving critiques originate.

Support of conclusions

Do the data actually support what the paper claims? Reviewers routinely flag overreaching conclusions, causal language where only correlation exists, or claims that outstrip the presented evidence.

Clarity and presentation

Is the paper well-written and logically structured? Can the figures be understood? Is the abstract accurate? Poor writing doesn't cause outright rejection but it slows review and creates reviewer frustration.

Reproducibility

Are methods described in sufficient detail for others to replicate the work? Are key reagents, strains, or datasets available or deposited? This has become more explicitly evaluated in recent years.

Fit for this journal

Reviewers sometimes flag that a paper is too specialized or too broad for the journal: even after the editor passed it to review. This is feedback you can use for future submissions even if the current one is rejected.

What Peer Review Doesn't Catch

Peer review has documented limitations that are worth understanding, not to dismiss it, but to understand what it can and can't do.

⚠Data fabrication and fraud (reviewers can't verify raw data they don't see)

⚠Selective reporting bias (reviewers only see what authors choose to show)

⚠Reviewer agreement is moderate at best; the same paper often gets opposite verdicts from different reviewers [see references]

⚠Publication bias: negative results are harder to publish regardless of quality

⚠Slow error detection: high-profile retractions often happen years after publication

⚠Image manipulation (routinely missed in traditional review; specialized tools now exist for journals that use them)

AI in Peer Review: Where Things Stand (2025)

AI has entered peer review from both directions: journals are exploring AI-assisted screening, and reviewers are using AI tools to help write reviews. Both raise real questions about what peer review is.

AI-assisted journal screening

Publishers including Springer Nature and Wiley use AI tools to flag potential image manipulation, data fabrication, and plagiarism before peer review begins. Statcheck and similar tools scan statistics in submitted papers. These are pre-review quality checks, not replacements for expert review.

Some journals also use AI to match manuscripts with suitable reviewers — a real problem at high-volume journals where editors may not know the right specialists for niche topics.

Reviewers using AI to write reviews

A 2024 study by Liang et al. in Nature detected that 6.5–17% of peer review reports showed signs of AI writing (based on word frequency analysis). Major journals including Science, NEJM, and Nature explicitly prohibit reviewers from submitting AI-generated reviews or sharing manuscript content with AI systems.

The concern: sharing a manuscript with ChatGPT or Claude breaches confidentiality. The manuscript is unpublished, proprietary, and submitted under trust that it will be reviewed by the person invited to review it.

For authors: Most journals don't yet use AI as a decision-maker in peer review — human editors still make accept/reject decisions. AI is screening for misconduct, not evaluating scientific merit. The peer review report you receive is still written by a human (or should be, per most journal policies).

References

Committee on Publication Ethics (COPE). (2017). Ethical Guidelines for Peer Reviewers. Retrieved February 2026. [publicationethics.org ↗]
Ioannidis JPA. Why most published research findings are false. PLoS Med. 2005;2(8):e124. [doi.org/10.1371/journal.pmed.0020124 ↗]
Lee CJ, Sugimoto CR, Zhang G, Cronin B. Bias in peer review. J Am Soc Inf Sci Technol. 2013;64(1):2-17. [doi.org/10.1002/asi.22784 ↗]
Tennant JP, Dugan JM, Graziotin D, et al. A multi-disciplinary perspective on emergent and future innovations in peer review. F1000Research. 2017;6:1151. [doi.org/10.12688/f1000research.12037.3 ↗]
NIH. (2014). Principles and guidelines for reporting preclinical research. National Institutes of Health. [nih.gov ↗]
PLOS ONE. Reviewer guidelines. Public Library of Science. Retrieved February 2026. [journals.plos.org ↗]

Suggested Citation

APA

Manusights. (2026). Peer review in biomedicine: Types, process, and what reviewers evaluate. Retrieved from https://manusights.com/resources/peer-review-explained

MLA

Manusights. "Peer Review in Biomedicine: Types, Process, and What Reviewers Evaluate." Manusights, 2026, manusights.com/resources/peer-review-explained.

VANCOUVER

Manusights. Peer review in biomedicine: types, process, and what reviewers evaluate [Internet]. 2026. Available from: https://manusights.com/resources/peer-review-explained

CC BY 4.0 - share and adapt freely with attribution to Manusights (manusights.com/resources).

About these resources: Manusights is a pre-submission manuscript review service staffed by researchers with publications in Cell, Nature, Science, and related journals. These reference guides are produced as free, independent resources for the research community. No sign-up required. Data sources and methodology are cited on each page. Browse all 25 resource guides or learn about Manusights.

Related guides in this collection

Peer Review Timelines →Responding to Reviewers →How to Write a Peer Review →Reporting Guidelines →AI Tools for Researchers →

← Predatory Journals All Resources Cover Letter Guide →