LINEARITY TESTING (REPORTABLE RANGE) CALIBRATION VERIFICATION

Effective Date: 
Tue, 05/01/2012
Reviewed: 
Tue, 01/10/2017
Policy: 

PURPOSE

Linearity studies are performed to determine the linear reportable range for an analyte. The linearity for each analyte is assessed by checking the performance of recovery throughout the manufacturer’s stated range of the testing system. This is done using a set of standards containing varying levels of an analyte in high enough and low enough concentrations so as to span the entire range of the test system.  Linearities are performed whenever a new analyzer, analyte or method is introduced into the laboratory, or an analyzer is replaced. Linearities may also be performed for troubleshooting purposes when quality control is unacceptable and deviations from acceptable data cannot be explained, or major analyzer repair or replacement of components has taken place.

Procedure: 

PROCEDURE

1. Linearity studies will be performed as part of the procedure "Evaluation of Automated Test Methods" in order to determine linear reportable range. For each analyte, a set of linearity standards will be tested in the same manner as patient samples.

2. Testing should be performed in triplicate, and at a minimum, in duplicate, when performed within a single run. If one value deviates greatly from the others due to random error, it may be removed from the data analysis and repeated.

3. The test results will be graphed and statistically analyzed as described below under "Evaluation of Linearity Study Data."

4. Once a linearity study has been performed to determine the linear reportable range for a test method, it may be repeated as recommended by the manufacturer (i.e.: following relocation of the instrument or after major maintenance) or calibration verification may be performed in accordance with CLIA guidelines, to verify continued acceptable performance of calibration and stated reportable range of the analyzer or analyte.

EVALUATION OF LINEARITY STUDY DATA

The data from the linearity study will be recorded on a linearity study sheet or programmed or downloaded into an available software program. Some manufacturers of linearity standards provide online data entry with real time comparison with peer group data and the capability to download linear regression graphs.

Values are plotted as observed values (Y axis) vs. expected values (X axis). Examine the raw data for obvious errors. If an analytical or technical problem is found, repeat the testing. Assessment will be made by evaluating the data and statistics using the following guidelines.

ACCURACY AND PRECISION

Review the linearity data for acceptable accuracy and precision. Ideally, endpoint assays should be within 10% of the standard’s stated value or peer group comparison value, but at a minimum, manufacturer’s stated tolerance limits should be met. Coefficient of Variation, which is a measure of precision, and is the standard deviation expressed as a percentage of the mean, ideally should also be less than 10%, or at a minimum, remain within the threshold of the manufacturer’s stated acceptable performance. It is ultimately the responsibility of the laboratory director to determine acceptability of this data and the validity of analyzer results with respect to accuracy and precision.

SLOPE AND Y-INTERCEPT

Two key statistical values in determining linearity are:

Slope:

Ideally, the slope is equal to 1.0

Acceptable Range: 0.9-1.1

lf the slope is outside the acceptable range; examine the results of the highest standard first. It is possible that the test is nonlinear at its highest value.

Y-intercept:

Ideally, the Y-intercept is equal to zero. For enzyme determinations and other assays with results in high numerical values, the Y—intercept may be much higher with no clinical significance. The Y—intercept for assays with low numerical values should be 0.0 + /— 1.0.

REPORTABLE RANGE

A reportable range will be established for each analyte tested. The upper limit of the reportable range will be set at the concentration of the highest standard tested which exhibited acceptable results for linearity, accuracy and precision. This concentration, however, may not exceed the manufacturer’s stated linear range.

For analytes which have a lower limit of linearity, the lower limit of the reportable range will be set at the lowest standard tested which exhibits acceptable results, however, this concentration may not exceed the manufacturer’s lower limit. Patient samples with · concentrations which exceed the reportable range will be diluted with the appropriate diluents and retested, when the analyzer provides this capability. Samples with concentrations which are lower than the reportable range will be reported as “Less than (the lower limit)”.

CALIBRATION VERIFICATION

Calibration verification is necessary to verify that an analyte’s calibration is still valid, and confirms that testing provides continued accurate results throughout the previously established reportable range.      

If calibration of an analyte or test system is performed every six months, utilizing three or more calibrators across a majority of the reportable range, then calibration verification is automatically met, and the laboratory does not need to perform further verification.

For analyzers and analytes that are not calibrated with a minimum of three calibrators verifying the low, midpoint and high end of the reportable range, calibration verification must be performed to substantiate the continued accuracy of the monitors throughout the reportable range, after initial validation studies are performed with the setup of the analyzer.

Calibration verification is performed every six months, as stated in current CLIA regulations. Calibration verification should also be performed under the following conditions:

1. Whenever major maintenance is performed or a critical component part of an analyzer has been replaced

2. Whenever reagent lots are completely changed (unless it has been stated and shown that these lot changes do not affect test results, as with manufacturer’s instructions and guidelines in package inserts and analyzer specific manuals)

3. When control values are found to be continually unacceptable, as with shifts and trends in Levy-Jennings graphs over a period of time

To perform calibration verification, low, midpoint and high level standards are tested in the same manner as patient samples. Evaluation of this analysis is achieved through use of slope, intercept, correlation coefficient or manufacturer established guidelines for acceptability criteria.

Each laboratory and its director should establish its own acceptance criteria for calibration verification. When acceptable performance is met, the calibration has been verified. If calibration verification is found to be unacceptable, the instrument must be recalibrated and all corrective action must be documented.