Medical Journal Selection: Indexing, Metrics, and Trust

Indexing badges and impact factors are easy to display and easy to misread. Check what a journal claims before you decide it deserves your manuscript.

Guide · 19 min read · Directive Publications Editorial Team

Four questions that decide where a paper goes

Evaluating a medical journal comes down to four questions, and the order matters.

1. Fit. Does the journal publish this kind of study, in this article type, for readers who need it?
2. Trust. Does it run real peer review and editorial processes, and say so openly?
3. Visibility. Is it indexed in the databases your readers, reviewers and institution actually search?
4. Standing. Where does it sit among comparable journals in your field, on the metrics you have reason to care about?

The first two are pass or fail; the last two rank the journals that survive. Authors who start with metrics tend to build a list of famous journals that were never realistic, then work down it one desk rejection at a time.

Start with fit: scope, article type and readers

A scope page is written to sound welcoming. What the journal actually publishes is a different document, and it is the one that predicts a desk rejection. The general routine for reading scope against recent content is covered in how to choose the right journal without getting scammed; what follows is the part specific to clinical and biomedical work.

Medical journals sort submissions by evidence level as much as by topic. Search the journal's own site for the design words that describe your study — randomised, cluster, cross-sectional, retrospective cohort, case-control, diagnostic accuracy, qualitative, single-centre — and see how many hits come back from the last two years. A journal that has published nothing with your design in that window is either uninterested in it or lacks reviewers for it, and both produce the same letter.

Then read the instructions for authors with four questions in mind:

• Does it accept your article type? Case reports, research letters, protocols and systematic reviews are welcome at some journals and not accepted at all by others. Note the word, table and reference limits for your type.
• Which reporting guidelines does it expect? Many medical journals ask for a completed checklist at submission: CONSORT for randomised trials, STROBE for observational studies, PRISMA for systematic reviews, CARE for case reports.
• What does it require for trials? Read the registration policy and check whether a data sharing statement is required.
• Does it impose a clinical relevance test? Some journals ask explicitly for a statement of what the finding changes for patients or practice, and will decline work that cannot supply one however sound the methods are.

Two tests sharpen the judgement. First, check where your own reference list clusters: if a journal appears three or four times among the papers you are arguing with, your study is already part of its conversation. Second, write the one-sentence claim you expect to make in the abstract and ask whether it would look out of place in the journal's recent table of contents — too preliminary, too narrow, or too far from the readership. If it would, the fit is weak and the metrics do not matter.

Specialty journal or general medical journal?

This choice is less about prestige than about who needs your finding.

Try this test. Explain the finding in two sentences to a clinician outside your specialty. If they would think or act differently afterwards, a general journal is worth an attempt; if it only makes sense with specialist background, go straight to the specialty journal. Decide in advance how many rejection cycles a thesis deadline or grant report can absorb.

What journal indexing means, and what it does not

Indexing means a bibliographic database has decided to include a journal's content. Databases differ sharply in how selective they are: some evaluate peer review, ethics and editorial quality before accepting a journal, while others include anything that meets technical requirements.

Indexing is also a current status, not a permanent achievement. Selective databases re-evaluate journals and stop covering some of them, so a journal that was indexed when a colleague published there may not be indexed now. Check it yourself, close to the date you submit, rather than relying on a recommendation or a badge.

MEDLINE, PubMed and PMC are not the same thing

The three names are used interchangeably on journal websites, and they mean three different things.

MEDLINE is the selective one. A journal gets in by applying and passing NLM review: at least two consultants, usually a scientist and a medical librarian, examine articles from the last two years of the journal's content, looking at scope, editorial policies, scientific rigour, production and impact. Before a journal can apply, it needs a confirmed ISSN registration, at least 12 months of publication, at least 40 primary peer-reviewed articles, and abstracts for its peer-reviewed content. NLM also expects the publisher to have at least a two-year history of quality scholarly publishing in the life sciences. Older guides describe a review committee called the LSTRC; NLM terminated it in 2025, so rely on NLM's current pages rather than secondhand summaries.

PubMed is the search interface. It includes MEDLINE citations, PMC articles and Bookshelf content, and publishers cannot apply to PubMed directly. PMC holds articles from journals selected for it, and also individual articles deposited to comply with funder policies.

So a journal's articles can appear in PubMed because some authors deposited funded manuscripts in PMC, without the journal ever being selected for MEDLINE or PMC. NLM's own FAQ says the presence of an individual article in PubMed does not necessarily mean the journal is included in MEDLINE or PMC. NLM also states that inclusion is not an endorsement of the content.

To check a journal yourself:

1. Search the NLM Catalog by title or ISSN.
2. Open the record whose title and ISSN both match the journal's website.
3. Read the Current Indexing Status line. It says either "Currently indexed for MEDLINE" or "Not currently indexed for MEDLINE."
4. Read the line labelled In:, which lists the databases and the volumes covered, in the form "MEDLINE: v297n6640, July 2, 1988-" followed by the PubMed range. A start date with no end date means coverage is still running; a closed range means it stopped.
5. Note the Electronic Links field, which records the journal's web address as NLM holds it.

A rule worth adopting: "indexed in PubMed" on a journal website tells you nothing until you have read that status line yourself.

Web of Science, Scopus, DOAJ and Google Scholar

Web of Science Core Collection. Clarivate evaluates journals against 28 criteria: 24 quality criteria for editorial rigour and best practice, and four impact criteria based on citation activity. Journals meeting the quality criteria enter the Emerging Sources Citation Index (ESCI). Those that also meet the impact criteria enter the Science Citation Index Expanded (SCIE), Social Sciences Citation Index or Arts and Humanities Citation Index. A journal that fails the quality criteria is removed. Clarivate's free Master Journal List shows which collection a journal is in, but it also covers specialty collections such as BIOSIS Previews, so "indexed in Web of Science" does not by itself mean the Core Collection.

Scopus. New titles are reviewed by the Scopus Content Selection and Advisory Board. Minimum requirements include peer-reviewed content with a public description of the review process, an ISSN registered with the ISSN International Centre, English titles and abstracts, and a public publication ethics and malpractice statement. Scopus re-evaluates journals and publishes a list of discontinued titles, updated monthly. Content already indexed stays in the database, so finding a journal's older articles in Scopus does not show that the journal is still covered.

DOAJ. The Directory of Open Access Journals lists open access journals that meet its criteria. These include review of each article by at least two independent reviewers in most fields, a public editorial board, all author fees listed, clearly stated licensing terms, and an ISSN confirmed at issn.org. It is a strong check for an open access journal. Its absence says nothing against a subscription journal.

Google Scholar. Its inclusion guidelines are technical: file formats, metadata and pages its crawler can reach. They cover preprints and drafts as well as journal papers. "Indexed in Google Scholar" tells you the journal has a crawlable website, and nothing about review.

How to verify an indexing claim yourself

Work from the ISSN rather than the journal name, because names are easy to copy.

1. Copy the ISSN from the journal's site. Print and online versions carry separate ISSNs; note which one you have.
2. Look it up in the ISSN Portal and confirm the title matches. The ISSN International Centre is clear that an ISSN does not guarantee the quality or validity of a publication's contents.
3. Check each claimed database at its own source, by ISSN and exact title.
4. Compare the details. Title, ISSN, publisher and web address should agree across the site and the database record.
5. Note the date you checked, in case you later need to show the status at the time you chose.

Watch the verbs too. "Applied to", "submitted for" and "under evaluation" are not indexing.

Journal metrics: what each number measures

Every journal metric is an average or a ranking built from citations to a whole journal. None measures the quality of a single article, including yours.

A hypothetical example makes the JIF concrete. Suppose a journal published 200 articles and reviews over the two previous years, and its content from those years was cited 600 times in the year being measured. Its JIF would be 3.0. Editorials and letters are not counted in the denominator.

Three practical points follow:

• A JIF comes only from Clarivate's Journal Citation Reports, which covers Web of Science Core Collection journals. Any other "impact factor" is a different product with a similar name.
• CiteScore is free to check on Scopus, without institutional access.
• Missing metrics are not a red flag. Google Scholar Metrics leaves out publications with fewer than 100 articles in its five-year window, and a newly indexed journal has no citation history yet.

Reading metrics without being misled

Compare within a field, never across fields. Specialties differ in typical citation levels, which is why Journal Citation Reports groups journals into subject categories; a Q1 journal ranks in the top quarter of its category on that metric.

Know what lifts an average. Clarivate points out that review articles are generally cited more often than typical research articles, so a review-heavy journal can post a higher figure without its original research being cited more.

Treat small differences as noise. A journal at 4.1 and one at 3.8 in the same category are not meaningfully different. A difference worth acting on moves a journal between quartiles.

Trust signals specific to medical journals

The general transparency standard is the Principles of Transparency and Best Practice in Scholarly Publishing, published jointly by COPE, DOAJ, OASPA and WAME. Its current version sets out 16 principles, covering the journal's name, website, peer review, ownership, editorial team, author fees, publication ethics and more, and it is a useful benchmark to read a journal's website against.

Medical journals should also be explicit about research on patients and participants. Work through this list on the journal's own policy pages:

1. Peer review model described, including who makes the final decision
2. Editorial board of named people whose affiliations you can confirm
3. Ethics approval and informed consent statements required for research involving people
4. Clinical trial registration required, in line with ICMJE: registration in a public registry at or before the time of first patient consent for enrolment, in ClinicalTrials.gov or a WHO ICTRP primary registry
5. Data sharing statements required for trial reports
6. Reporting guideline checklists required or encouraged
7. Conflict of interest disclosure required from authors
8. A corrections and retractions policy you can actually find
9. Fees, waivers and licence stated before submission
10. Received, accepted and published dates printed on articles

Two badges deserve particular care. A claim to follow the ICMJE Recommendations is not something ICMJE certifies: in April 2025 it decided to stop maintaining its list of journals claiming to follow them, because it could not verify the list and many listed journals did not actually adhere. Judge the concrete policies above instead. COPE membership, by contrast, can be checked in the member directory on COPE's own website, and a membership claim that does not appear there should stop you.

The warning signs that should end your evaluation outright, such as guaranteed acceptance or pressure to pay before review, are covered in 12 red flags of predatory journals.

Hijacked and cloned journals

A hijacked journal is a fraudulent website that impersonates a real journal. Retraction Watch describes hijacked journals as mimicking legitimate ones by adopting their titles, ISSNs and other metadata, often while offering fast publication and indexing in databases such as Scopus. Clarivate warns that journals with a Journal Impact Factor are a more attractive target for this kind of fraud, so the most credible-looking journals are the ones worth checking hardest.

A clone copies the title and ISSN, so no check built on those two will separate it from the original, and even the web address can mislead. Retraction Watch has documented cases in which a genuine journal let its domain lapse and hijackers registered it, with the Master Journal List and Scopus each carrying the compromised link. A matching domain is therefore necessary but not sufficient. Treat these checks as cumulative:

1. Never start from an email link or a search advertisement. Reach the journal from its NLM Catalog record or a database record you found yourself.
2. Compare the domain character by character. A different domain is decisive evidence of a clone; a matching one is only the absence of that evidence.
3. Check whether the domain changed hands recently. A public WHOIS lookup shows registration dates. A domain registered a year or two ago, under a journal that claims decades of publication, fits the documented pattern.
4. Follow a DOI you trust from a recent article found in PubMed or Scopus, and read the result as one signal among several.
5. Search the Retraction Watch Hijacked Journal Checker.
6. Confirm with the editorial office through contact details you find independently, such as the publisher, the owning society or an editor's institutional page.

If you have already submitted to a site you now suspect, make no further payment, keep every email and receipt, ask in writing for the manuscript to be withdrawn, and tell your co-authors and research office before it goes anywhere else.

Open access, subscription and hybrid journals

The business model tells you who pays and who can read. It does not tell you how carefully papers are reviewed. Reputable and disreputable journals exist in every model, and open access myths still lead some authors and research offices to assume otherwise.

Evaluate an open access journal with every check above, then confirm three more things: the licence for published articles is stated on the site, all charges and the waiver process are visible before submission, and, for a fully open access journal, it has a DOAJ record.

For subscription and hybrid journals, check what you may deposit in a repository and when, and whether that satisfies your funder. Jisc's Open Policy Finder includes a compliance checker that compares a journal's policies with a funder's open access requirements. The checker covers a pre-selected, predominantly UK list of funders, so many medical authors will not find theirs, though the journal policy records remain useful. Either way, confirm against the funder's own current policy.

Build a ranked shortlist

Put the evaluation into a worksheet so the decision is visible and repeatable.

Work through it in this order:

1. Collect six to ten candidates from your own reference list, from where similar recent studies appeared, and from colleagues in your specialty.
2. Remove any that fail fit.
3. Remove any that fail trust.
4. Verify indexing at the source and remove any that do not meet a requirement your institution or funder actually imposes.
5. Rank the rest by strength of fit first, then by metrics within the category, then by practical factors such as fees and open access compliance.
6. Keep three to five, in order, and record why each sits where it does, so a rejection does not send you re-arguing the whole ranking.

Date the worksheet. If months pass before you reach the third journal, re-verify its indexing first.

Metrics, indexing and your career

Early-career researchers hear two contradictory messages: publish in the highest-impact journal possible, and journal metrics do not matter. Both oversimplify. Start by finding the written rules that actually apply to you. Some institutions, funders and national assessment systems name specific databases or quartiles. Others follow principles like those of the San Francisco Declaration on Research Assessment, which recommends against using journal-based metrics such as the Journal Impact Factor as a surrogate for the quality of individual articles, or in hiring, promotion or funding decisions. Clarivate itself says journal-level indicators are designed to support journal evaluation, not to assess individual papers or researchers.

When a rule exists, read it precisely. Which database? Which collection? Does status count at submission, acceptance or publication? Ask your research office for the policy in writing.

Where no rule applies, weigh a higher number against time. A strong paper in a journal that reaches the right readers, published after one round of review, often does more for you than the same paper still circulating after three rejections.

Where to go from here

The checks in this guide take an afternoon the first time and perhaps an hour once you know where each status is issued, which is cheap against a submission cycle measured in months.

Directive Publications sets out its author guidelines and peer review process publicly, so you can run the same checks on its journals that you would run on any other before you decide where to submit.