Published 22 October 2026

Outdated Clinical Guidelines: Whose Job Is It to Notice?

Presenting an outdated SOP in trainings

Picture an audit that goes well. Every SOP has an owner and a review date, the training records are complete, and the guideline in the shared folder carries January’s approval. No one in the room knows that the society behind it published a revision in March, moving a threshold the local protocol still quotes. Like many internal audits, it checks that the team follows its documents. Whose job is it to notice when those documents fall behind? Nobody present can say.

Which one are you working from?

January

Your copy, opened yesterday

Guideline_v4_approved.pdf

March

The society’s revision

Same title, same layout. One threshold moved in March. Nothing in your copy says so.

An outdated clinical guideline is a patient-safety risk, and a compliance risk once your own procedures cite it. It rarely looks like either, because everyone involved is behaving sensibly. Careful people follow the copy they were given: a practice’s recall protocol that still reflects last year’s recommendation, or the dosing table a clinician saved to their phone years ago. Nothing about either suggests it has been replaced. The PDF opens exactly as it did yesterday, and the colleague everyone asks still remembers the recommendation the way it used to be.

Key takeaways

An audit can confirm that you follow your procedures without checking whether the guidance beneath them still stands.
News of a guideline revision can arrive within days, while the SOP, training slides and leaflet built on it keep the old number for months.
Approved materials often carry a review date, not a trigger that fires when the guideline they cite moves.
Each external source needs a named person, written down. In a small practice that is whoever runs it, unless someone else is named; a four-step check is below.


How guidelines go out of date between reviews

On an annual cycle, a team whose copy was signed off in January may work from the previous version for ten months before anyone formally notices March’s revision. Rolling reviews and change notices shorten that window, but they rarely bring it to zero.

Between reviews

How the gap opens

January

The review signs the guideline off. It goes into the folder, the protocol and the training slides.

March

The society publishes a revision. Nothing in your copy changes, not even the file name.

Since then

Careful people keep following it, because they have no reason not to.

Rolling review, change notices or an annual cycle: the gap is shorter or longer. It is rarely zero.

When the European Resuscitation Council published its 2025 guidelines in October 2025, any local SOP, training slide or pocket card built on the 2021 edition fell a version behind.[1] It stayed there until someone assessed the change. Resuscitation guidance usually has a clear owner; the harder cases are quieter, such as an instrument-reprocessing routine or a specialty protocol that quotes a society statement nobody watches. A training deck passes its approval review and stays in use even if the guideline it cites has moved, and the first person to notice may be a clinician in the audience.

Some documents are frozen on purpose. A trial protocol or an approved patient information sheet stays fixed until it is formally amended, with the approvals that requires, and that is good practice. There, staying fixed is the point, and the risk is a guideline change nobody has assessed.

An audit built on the same documents shares the blind spot. A complete training record shows who finished the module, not whether the module still matches the guideline. A clinical audit that scores practice against January’s copy can only ever confirm January.

Audit result · scored against the folder copy

Training record complete
Practice matches the local protocol
Protocol on file: January’s copy
Confirms January
Not on the checklist: the society’s March revision.


How old is a “current” guideline?

Two measures of the guideline itself

The source · not your copy

Spain · 4 national guidelines

Nearly 1 in 5

(18.7%) of 113 sampled recommendations had new evidence that could trigger an update within three years.

Martínez García et al., CMAJ 2014 · Kaplan–Meier estimate

Australia · national guidelines

9.8 years

the mean time since the evidence behind guidelines for the highest-burden conditions was last searched.

White et al., Med J Aust 2026 · 27 of 31 guidelines gave a search date · measured July 2025

Both figures describe the guidelines themselves, and a local copy can be older still.[2][3] An old search does not make a recommendation wrong, but nothing on the page shows that anyone has looked recently. Developers can make that check quicker with a visible last-search date and a stated update plan, or by moving to a living guideline.

Finding the right guideline is a failure point of its own. Jones and colleagues coded 5,708 medication incident reports from England and Wales (2016–2021) that mentioned a guideline, including every report of moderate harm or worse.[4] Of the 642 in which using a guideline went wrong, almost half (300) failed at the first step of finding the right one. A separate 88 involved guideline content that was itself incorrect or out of date. Incident reports cannot give rates, and an outdated guideline is reported only once someone notices it.

England & Wales · 2016–2021

Incident reports · not rates

642 reports where using a guideline went wrong

Step 1 · finding the right guideline300
Step 2 · finding the relevant information in it166
Step 3 · reading and using it correctly176
+88 further reports where the guideline’s content was itself incorrect or out of date

Source: Jones et al., Drug Safety 2024, Tables 1–2. 5,708 coded reports, including every report of moderate harm or worse. An outdated guideline only reaches a report if someone notices it.


Whose job is it to notice?

Who should notice depends on the setting. In a small practice it is whoever runs it, unless they name someone else in writing, such as the practice manager or quality lead. In a GP practice, guideline updates, drug-safety alerts and formulary changes may all land in that one inbox; in England, for example, from NICE and the MHRA. Larger organisations often have a route for national guidance, such as a clinical effectiveness or quality team, or a medical affairs team following its therapy area, though that route does not always reach the people who build the training.

The gap opens around the society statement, the specialty protocol and the local adaptation that sit outside those routes. Ownership tends to attach to the local document rather than to the source it was built on: the SOP or e-learning module has an owner who reviews it on schedule, while the guideline behind it often has nobody. Everyone else reasonably assumes that whatever is in the folder is current. Approved materials often carry a review date rather than a trigger that fires when that source moves, so the SOP, the teaching slides, the leaflet and, in industry, the approved field or MSL deck can keep the old threshold for months.

Who owns what

The local document

SOP · e-learning module

Teaching slides · leaflet

✓ Owner · review date

The source it was built on

Society statement

Specialty guideline

Owner: often nobody

A review date does not fire when the source moves.


Three questions worth asking this week

01

Would you know when a guideline you rely on changes?

If the answer is “at the next review”, that is how long you could be working from the old version.

02

Would the people affected hear about it quickly?

A change can reach whoever handles quality long before it reaches the people who use the guideline every day.

03

Could you show which version informed a decision?

It tends to come up in an inspection, a review or after an incident.

A four-step check

  1. Pick

    Pick the external sources your protocols, training or patient materials rely on most; five is enough to start. They may be guidelines and, for some teams, manufacturers’ instructions for use, product information or field safety notices. If you use guidelines rather than own them, start with the one you would least want to be wrong about.

  2. Check

    Check each one on the issuing body’s own website or register. Is it current, revised or withdrawn? Look at the last-search or review date, not only the publication date, and if there is none, write that down.

  3. Trace

    If it has moved, list the documents that quote it (SOPs, training modules, decks, letters, leaflets). Tell the owner of any that are not yours; for your own, decide what happens to each: update it through your normal approval route, withdraw it, or keep it on purpose and record why. Then tell the people who use it.

  4. Name

    Either way, write down the version you checked, a name and a next-check date beside each source. A few extra columns in the document list or register you already keep will do.

Your turn

How do you find out that a guideline you rely on has changed?

A society alert A colleague A group chat Not until someone asks


References

[1] Greif R, Lauridsen KG, Djärv T, et al.; European Resuscitation Council Guidelines 2025 Collaborator Group. European Resuscitation Council Guidelines 2025: executive summary. Resuscitation. 2025;215(Suppl 1):110770. https://doi.org/10.1016/j.resuscitation.2025.110770

[2] Martínez García L, Sanabria AJ, García Álvarez E, et al. The validity of recommendations from clinical guidelines: a survival analysis. CMAJ. 2014;186(16):1211–1219. https://doi.org/10.1503/cmaj.140547

[3] White H, Fernandes E, Peduru-Arachchige K, Turner T, McGloughlin S, McDonald S. Time for an update: the currency of clinical guidelines in Australia. A cross-sectional analysis. Medical Journal of Australia. 2026;224(9):e70288. https://doi.org/10.5694/mja2.70288

[4] Jones MD, Liu S, Powell F, et al. Exploring the role of guidelines in contributing to medication errors: a descriptive analysis of national patient safety incident data. Drug Safety. 2024;47(4):389–400. The 642, 300, 166, 176 and 88 figures are from Tables 1 and 2. https://doi.org/10.1007/s40264-024-01396-7

Arthur Boritzka, Chief Product Officer @ Elumity