How we choose journals and write summaries

The journals we monitor

Each specialty has a fixed panel of five journals. We monitor those five for every new article, every week. We do not sample, rank, or pick out the interesting ones — if a paper appears in your panel and has an abstract, it reaches your digest.

That means the number of articles varies week to week. A quiet week in Ophthalmology might bring four papers. A busy week in Cardiology might bring fifty. It also means some weeks a journal in your panel publishes nothing at all, so fewer than five journals appear in that Monday's digest. The panel has not changed; that journal simply had nothing new.

Dental panels

Dentistry is sold separately from the 31 medical specialties. The same method applies.

How the panels were chosen

Panels were selected by a practising NHS consultant, not by an algorithm. Each panel is built from journals specific to that specialty — the titles a clinician in that field would name if you asked them what they ought to be reading.

We favoured journals that publish original clinical research over those that are mainly review or educational, and we favoured the titles more likely to change UK practice — which is why ten of the journals we monitor are British.

We review the panels annually, and sooner if a subscriber tells us we are missing something obvious in their field. If you think your panel is wrong, tell us — that feedback has changed panels before.

Which articles are included

We query PubMed for every article published in your panel's five journals, matched by ISSN. We look back over a rolling ten-day window each week, which overlaps deliberately so nothing falls through the gap between two Mondays. Articles we have already sent you are filtered out by DOI, so an overlap never means a repeat.

One filter applies: the article must have an abstract. Beyond that we do not exclude by article type. Original research, systematic reviews, guidelines, editorials, commentary and correspondence all reach you if they carry an abstract and appeared in your panel that week. The summary names the study type at the top, so you can see in a second whether you are reading an RCT or an editorial.

Articles without an abstract — most letters, many news items, some errata — are not included, because there is nothing substantive to summarise.

How each summary is written

Summaries are generated by a large language model (Anthropic's Claude) working from the PubMed record: title, authors, journal, and abstract. Every summary follows the same fixed structure:

• Study type — what kind of study this is
• Key question — the clinical question in one sentence
• Key findings — two or three points, with the actual numbers, effect sizes and confidence intervals where the abstract gives them
• Clinical relevance — why it matters in UK practice
• Limitations — the most important caveat

Every article carries a direct DOI link to the source. We expect you to follow it before anything reaches a patient.

Quality checks

Three checks run on every digest. The model is instructed to summarise only what the abstract states and not to infer beyond it. Summaries that come back empty or malformed are held rather than sent, so a failed generation never reaches you as a blank entry. And the digest that lands in your inbox is the same one published on our samples pages — the summaries on this site are real sends, not marketing copy written separately.

We do not clinically peer-review each summary before it sends. With this volume, across 35 panels, no one could and still deliver on Monday morning. We would rather say that than imply a review process we do not run.

Corrections

If a summary misrepresents a paper, email info@mondayclinicalbrief.co.uk and tell us which article. We will check it against the source.

If we got it wrong, we correct it in the next digest to that specialty, naming the article and what was wrong, and we fix the published sample page if that week is on the site. We do not quietly edit and move on. If the error came from the abstract rather than from us, we will say that too.

What this is not

The Monday Clinical Brief is an awareness service. It tells you what was published in your field last week and gives you enough structure to decide what to read. It is not a guideline, not a systematic review, and not clinical advice. Read the paper before you act on it, and apply your own judgement — the same as you would with any journal alert.