How to interpret EP15-A3 precision verification reports for CLIA compliance: SD, CV, total imprecision, and what surveyors look for in your documentation.
Frequently Asked Questions
Why is precision verification required?
42 CFR §493.1253 requires laboratories to verify performance specifications, including precision, before reporting patient results and after changes that could affect performance.
What do mean, SD, and CV mean on the report?
Mean is the average of replicates; standard deviation is how widely replicates spread around it; coefficient of variation is the SD as a percent of the mean, CV = (SD / mean) x 100.
How many replicates does it need?
CLSI EP15-A3 recommends at least 20 replicates for within-run precision and a 5-day by 5-replicate design for full within-laboratory precision. Fewer than 20 widens the confidence interval and weakens the estimate.
What do Pass, Fail, and Uncertain mean?
Against a vendor SD goal, Pass means the upper bound of the 95 percent confidence interval for the observed SD is below the goal; Fail means the lower bound exceeds it; Uncertain means the goal falls inside the interval and more replicates may be warranted.
What acceptance criterion should I use?
A fraction of the CLIA TEa is common, often the ADLM half-TEa convention, or the manufacturer's published within-run SD. Final acceptance rests with the medical director or designee.