Required section · Section 3 of 6

Matching each performance characteristic to a study

Select a study from three questions: the intended use, the comparator available, and the performance characteristic under review. First classify the method as verification or validation; then document which characteristics apply and why.

For quantitative methods, precision and bias may need evidence; measuring interval, detection capability, interference, reference interval, and comparison evidence are selected only when the intended use and claimed performance make them relevant. A comparison is not automatically required for every modified or noncleared method.

Use the design name only after that selection. EP15 can support a short precision-and-bias verification; EP05 supports broader precision establishment; EP06 addresses measuring-interval work; EP09 addresses comparison and bias against a suitable comparator. These names do not replace the decision about what evidence is needed.

For qualitative or semiquantitative methods, select cutoff-focused precision and agreement evidence appropriate to the intended use rather than applying quantitative CV language.

Write the characteristic, comparator, and acceptance criterion before collecting data, including why a characteristic does not apply.

Study-selection matrix with intended use, comparator, and performance characteristic as three inputs, followed by examples of EP15, EP05, EP06, and EP09 only when the selected question requires them. — Figure 1Study-selection matrix routing evidence by intended use, comparator, and performance characteristic rather than a universal requirement list.

Three method scenarios and the pathway each one triggers
Scenario	Method status	Pathway	Key studies owed
A	FDA-cleared method, used exactly per package insert	Verification	Precision, bias, reportable-range spot check, reference interval fit
B	FDA-cleared method with a laboratory-changed dilution step	Validation	Precision, bias, reportable range, specificity/interference, reference interval, all established locally
C	Laboratory-developed qualitative assay, no FDA clearance	Validation	Precision near cutoff (C5/C50/C95), PPA/NPA/OPA against a comparator, interference, stability

Knowledge checks

Reading and checks are open. Sign in only to save.

Knowledge check 1

A laboratory is validating a modified version of an FDA-cleared quantitative method. Which statement about the required performance characteristics is accurate?

Choose one option.

The laboratory establishes accuracy, precision, reportable range, specificity/interference as applicable, and reference interval from its own data

The laboratory only needs to verify precision, since the modification did not change the chemistry

Reference interval work can be skipped if the original manufacturer interval is kept unchanged

Knowledge check 2

A method-comparison study shows a correlation coefficient of 0.99 between the candidate and comparator methods using ordinary least-squares regression. What does CLSI EP09 say about relying on this result alone?

Choose one option.

A high correlation coefficient is not sufficient evidence of agreement; Deming or weighted Deming regression and a bias estimate against a predefined criterion are needed

A correlation coefficient above 0.95 is generally accepted as proof the two methods agree

Ordinary least-squares regression is preferred here because it is simpler to calculate

Section status

Finish this section

Reading and checks are open. Sign in only to save.

The module finishes after every required section is marked done and every check in those sections is correct.