Troubleshooting / Common Issues
One or more concentration levels fail linearity (residual exceeds SEa)
Possible causes
- –Sample preparation error at that level — pipetting inaccuracy or incorrect dilution ratio
- –Carryover effect — high-concentration sample contaminating adjacent low-concentration wells
- –Imprecision at that level is high relative to other levels, inflating the mean residual
- –Genuine nonlinear behavior at the high or low extreme of the measurement range
Recommended action
First, review the raw replicates at the failing level for outliers. If a single replicate is responsible for the residual, investigate whether a documented technical error justifies its exclusion. If multiple replicates are consistently off, check sample preparation records and repeat testing at that level before concluding true nonlinearity.
Residuals show a systematic curved pattern (U-shape or S-shape)
Possible causes
- –The measurement procedure has genuine nonlinear calibration behavior
- –A hook effect is present near the upper limit of quantitation
- –Samples at the extremes of the range were prepared with a different matrix than intermediate levels
- –Commutability issues between the linearity panel samples and patient specimens
Recommended action
A consistent curved residual pattern is a methodological finding, not a random error. Report the actual linearity interval as the concentration range within which all residuals pass the SEa criterion. Consult the manufacturer if the failing concentrations fall within the published AMR.
Regression slope is significantly different from 1.0
Possible causes
- –Systematic proportional bias between the measurement procedure and the expected values
- –For Design A1: the high sample concentration is inaccurate or the diluent contains measurand
- –For Design A3: discrepancy between assigned values and the measurement procedure's calibration traceability
- –Carryover at the highest concentration level skewing the regression
Recommended action
A slope significantly above or below 1.0 indicates proportional bias. For Design A1, verify that the diluent is truly measurand-free and that the high sample was processed correctly. For Design A3, compare the measurement procedure's calibration traceability with that of the reference material. Investigate whether a method comparison (CLSI EP09) study is warranted.
AMR verification fails proximity check
Possible causes
- –The highest or lowest concentration levels in the study are too far from the claimed AMR bounds
- –Available samples did not span the full manufacturer-claimed range
- –Proximity limits were set too stringently relative to available sample concentrations
Recommended action
Obtain samples with concentrations closer to the claimed AMR upper and lower limits. If samples at the AMR extremes cannot be obtained, document the actual interval verified and note the limitation in the study report. Consider adjusting proximity limits if the gap between the study extremes and the claimed bounds is clinically acceptable.
Accuracy analysis fails at one or more levels (Design A3)
Possible causes
- –Calibration bias between the measurement procedure and the assigned value traceability chain
- –Matrix effects — assigned concentrations were established in a different sample matrix
- –Reagent lot or calibrator lot drift during the study period
- –Incorrect expected values entered for the failing levels
Recommended action
Confirm that the assigned values were entered correctly. If so, check calibration records and reagent lot information for the study period. A systematic accuracy failure across all levels suggests a calibration issue; a failure isolated to a single level suggests a sample-specific problem. Consider a method comparison study to further characterize the bias.