Quick Review Cards for Medical Laboratory Science

Test Performance Specifications

and Verification

Laboratory Operations Review 56

CLIA requires labs to verify manufacturer’s analytical performance claims for # 1–4 below on new methods introduced
after April 24, 2003.

PARAMETER EXPLANATION VERIFICATION

1. Accuracy


2. Precision


3. Reportable range


4. Reference interval

Lab tests samples of known values (controls, calibrators,

proficiency samples, previously tested patient speci-

mens) to see how close results are to known value.

Lab repeatedly tests same samples (on same day &

different days) to see how close the results are.

Lab tests samples with known values at highest &

lowest levels claimed to be accurate by manufacturer.

If manufacturer’s or published reference ranges are

used, lab must test specimens from normal subjects to

verify ranges. Ranges may need to be adjusted to fit

lab’s patient population.

How close measurement is to true value.

Reproducibility. How close results are when same

sample is tested multiple times. (Note: A procedure

can be precise but not accurate.)

Range of values over which lab can verify accuracy

of test system. Also known as linearity.

Formerly called normal value. Can vary for different

patient populations (age, gender, race). Established

by testing minimum of 120 healthy subjects & de-

termining range in which 95% fall. (Note: 5% of

healthy population falls outside of reference range.)

Labs may use manufacturer’s reference ranges or

published reference ranges, if appropriate for their

patient population.

continued...