How It Works
Study Design
A qualitative method comparison study tests the same patient specimens on both the reference method and the test method, then compares the categorical classification assigned by each. CLSI EP12 specifies the following minimum design requirements:
| Study Type | CLSI EP12 Requirements |
|---|---|
| Clinical Agreement | At least 20 specimens with an approximately equal distribution of positive and negative results; samples should include some specimens near the cutoff to adequately challenge the examination's performance |
| Precision Near Cutoff | Three samples at C5, C50, and C95 (below, at, and above the cutoff); at least 20 replicates per specimen spread across at least 5 days (consecutive or nonconsecutive) |
| Days | Not critical for agreement studies; required for precision near-cutoff studies (at least 5 days) |
| Semi-Quantitative | Numeric results are converted to categorical calls using the configured cutoff before computing agreement statistics; the cutoff and direction (e.g., ≥100 = Positive) are verified during experiment setup |
Semi-quantitative note: For tests that report a numeric index (e.g., an absorbance ratio), the cutoff value and direction are configured during the design step. Once applied, all downstream agreement statistics are computed on the resulting categorical calls — not on the raw numeric values. CLSI EP12 addresses semi-quantitative tests (those with an internal continuous response converted to a binary result via a cutoff), but evaluates them as binary output examinations; the cutoff itself is a key performance parameter verified during C5/C95 precision studies.