How Rhazes calculates compliance

The exact compliance formula a Rhazes audit uses, why exclusions and exceptions leave the denominator, and a worked example.

Updated

The formula

Compliance for a standard is:

met ÷ (met + not met)

Rounded to a whole percentage.

What is not in it

Excluded and exception records are not in the numerator or the denominator. They leave the calculation entirely.

That is standard clinical audit methodology, not a Rhazes shortcut. Compliance is a statement about the records where compliance was possible, so a record that was never in scope, or that had a documented reason not to comply, cannot sensibly count either way.

A worked example

An audit of 100 records against one standard, target 90%:

Records submitted100
Excluded, outside inclusion criteria20
Exception, valid documented reason5
Met60
Not met15

The denominator is 60 + 15 = 75. Compliance is 60 ÷ 75 = 80%. Against a 90% target, the standard is not met.

Look at what the other two answers would have been. Dividing by all 100 records gives 60%, which understates your service by counting paediatric records in an adult audit. Counting the exceptions as passes gives 65 ÷ 80, or 81%, which flatters it by treating a refusal as a success. Neither is defensible, and both are what happens when this is tallied by hand.

Why your criteria matter so much

The formula is fixed. The only things you control are which records reach the denominator and which are removed from it, and those are set entirely by your inclusion criteria and exceptions.

Which means a compliance figure you disagree with is almost never a calculation problem. It is a scoping problem, and the fix is in the standard rather than in the maths. See Inclusion criteria and exceptions.

Meeting the target

A standard meets its target when the achieved percentage is greater than or equal to it. Exactly on target is a pass.

When nothing is in the denominator

If every record was excluded or an exception, compliance shows as 0%. That is not a failing service, it is an audit that judged nothing, and it almost always means the inclusion criteria are wrong.

Related articles