SQUIRE 2.0 Reporting Guideline Explained

SQUIRE 2.0 is the reporting guideline for healthcare improvement work: the Standards for Quality Improvement Reporting Excellence, revised in 2015. It exists because improvement studies were being reported in a format built for trials, which systematically hid the things that decide whether an improvement can be repeated elsewhere.

The problem it was built to fix

A trial report is designed around a fixed intervention. Fidelity to protocol is a virtue, and deviation is a limitation to be declared.

Improvement work is the opposite. The intervention is expected to change as it meets reality, because that adaptation is how it comes to fit the setting. Reported in trial format, all of that disappears: the reader sees a starting state, an ending state, and an intervention described as though it were stable throughout. Anyone attempting to reproduce it in another unit has no idea what actually happened.

SQUIRE responds by asking for things a trial report never requests, most importantly a description of how the intervention evolved and why.

The items people most often miss

Most of the checklist maps onto sections authors already write. Four items do not, and they are where submissions fail.

The rationale. Not a literature review, but an explicit statement of why anyone expected this intervention to produce this outcome, naming the reasoning or framework connecting them. Many projects have never articulated it.
Context. A characterisation of the setting detailed enough for a reader elsewhere to judge whether their situation resembles it. Without this the report cannot transfer.
The evolution of the intervention. What changed during the project, when, and on what evidence. This is the item most often omitted entirely, and the one that most damages reproducibility.
The distinction between the intervention and the study of it. The change you made and the method by which you measured it are separate things and are reported separately. Conflating them is endemic.

Why it has to be adopted before you start

Reporting guidelines are usually applied at write-up. SQUIRE cannot entirely be, because two of its requirements are records rather than prose.

The evolution of the intervention has to be logged as it happens. A team reconstructing at month nine what they changed in month three, and why, will produce something between a guess and a rationalisation. Contextual detail is similarly perishable, because staffing, workload, and competing initiatives at the time of implementation are rarely recoverable later.

The practical instruction is to open the checklist at the design stage and set up a change log alongside the outcome data. This is the same discipline that makes a protocol written in advance more defensible than one assembled afterwards.

Where SQUIRE sits among the other guidelines

SQUIRE is one of a family, and choosing correctly matters because journals check. CONSORT covers randomised trials, STROBE covers observational studies, and PRISMA 2020 covers systematic reviews. SQUIRE covers improvement work specifically, and the boundary case is a project that used a rigorous design to evaluate a local change: it may need SQUIRE for the improvement account and a design-specific guideline for the evaluation.

None of these is an appraisal tool. A reporting guideline governs what appears in the paper, not whether the study was well conducted, and the distinction between reporting quality and study quality is one reviewers apply strictly.

Using it as a reader

The checklist also works in reverse. When an improvement report is being considered for inclusion in a review, the SQUIRE items make useful probes: is the context described well enough to judge transferability, is the rationale stated, is there any account of how the intervention changed. A report failing all three may be accurate and still be unusable as evidence for anywhere else.

Frequently asked questions

What is SQUIRE 2.0?

It is the Standards for Quality Improvement Reporting Excellence, revised in 2015, a reporting guideline for healthcare improvement work. It exists because improvement studies reported in a trial format hide the adaptation that makes an improvement fit its setting, leaving readers elsewhere unable to judge whether it would transfer to theirs.

How does SQUIRE differ from CONSORT and STROBE?

They cover different study types. CONSORT governs randomised trials, STROBE governs observational studies, and SQUIRE governs improvement work. The key structural difference is that SQUIRE expects the intervention to have changed during the project and asks you to report how, whereas trial guidelines treat deviation from protocol as a limitation.

Which SQUIRE items are most often missed?

The rationale connecting intervention to expected outcome, a description of context detailed enough to judge transferability, an account of how the intervention evolved during the project, and a clear separation between the intervention itself and the method used to study it. The evolution item is the one omitted most often and the one that most damages reproducibility.

Can SQUIRE be applied at write-up?

Only partially. Two requirements are records rather than prose: the log of what changed during the project and when, and the contextual detail about staffing, workload, and competing initiatives at the time. Both are perishable, and a team reconstructing them months later produces something closer to rationalisation than record.

Is SQUIRE a quality appraisal tool?

No. A reporting guideline governs what must appear in the published account, not whether the work itself was well conducted. A methodologically weak project can be reported in full compliance with SQUIRE, and a strong one reported badly. Reviewers apply the distinction between reporting quality and study quality strictly.

SQUIRE 2.0: Reporting Improvement Work, and the Two Items You Cannot Add Later