Clinical audit: checking practice against the standard

A clinical audit measures actual practice against an agreed standard and then changes something as a result. The second half is what distinguishes it from a survey, and it is the half most often missing: a great many audits are completed, presented and filed, and nothing about the care delivered is different afterwards.

What a clinical audit requires

• A standard that exists before you start — a guideline, a protocol, a stated target — rather than one inferred from the results.
• A topic where practice can realistically change, because auditing something nobody can influence wastes everybody's time.
• A defined sample: which records, over what period, chosen by a rule rather than by convenience.
• Criteria that are measurable from the records as they are actually kept.
• A re-audit date set at the start, since the second cycle is where the method earns its name.
• Named ownership of the change, not just of the audit.

Choosing a topic worth auditing

The useful ones come from signals rather than from a list: a complaint, a near miss, a guideline that changed recently, a treatment where the team suspects variation. Topics chosen because data is easy to extract produce audits that confirm good performance in areas nobody was worried about — reassuring, well presented and worthless. If the likely finding is that everything is fine, the topic is probably wrong.

Decide the sample before looking at any data. Choosing records after a first glance — even with good intentions — makes the result unusable, and in a contested case it is the first thing an outside reviewer will examine.

Closing the loop

1. State the finding plainly, including the proportion rather than only the direction.
2. Agree one or two changes rather than a list of ten, since a long action plan is a way of not deciding.
3. Give each change an owner and a date, and record them where somebody will look again.
4. Make the change easy to follow — a checklist, a template, a default — because relying on everybody remembering is what produced the gap.
5. Re-audit on the date set at the start, with the same method, or the comparison means nothing.
6. Report the second cycle whatever it shows, including no improvement, which is itself a finding about the change rather than about the team.

Why audits stall at the first cycle

Because the first cycle produces a document and the second produces accountability. Re-auditing asks whether the change you agreed actually happened, which is uncomfortable in a way that describing current practice is not — and it is the only part that tells you whether anything improved. An organisation with twenty first-cycle audits and no second cycles has an audit process that measures effort rather than care.

Ettex Records suits the audit record itself: topic, standard, sample rule, findings, agreed changes with owners, and the re-audit date as a field you can filter on. That date is the one that matters, because it is the one that gets quietly missed.