Quality Systems Management
Calibration and calibration verification may seem difficult and confusing to some laboratory personnel, but they don't have to be. Following the manufacturer's instructions, creating written procedures and ensuring that personnel are trained in these procedures will help clarify these important processes.
Calibration is a process of adjusting the instrument or test system to establish a correlation between the instrument's measurement of an analyte and the actual concentration of that analyte. In other words, calibration "sets" the system to give an accurate result for each analyte.
Calibration verification is just as it sounds: It is a process that verifies that the instrument calibration is still valid. A successful calibration verification confirms the instrument is providing accurate results throughout the reportable range of the test.
Another related process is verification of performance specifications. The manufacturer will make certain claims about how the test system should work, but you have to verify that it actually works that way in your lab. This involves verifying several performance characteristics and has to be done before you can begin patient testing. The reportable range is one of the performance specifications that has to be verified.
Calibration is required for non-waived, quantitative test methods. It is not required for waived testing, qualitative methods or manual and microscopic procedures. There are also instruments calibrated by the factory that do not allow user calibration; however, these instruments still require the user to perform calibration verification.
Samples used during the calibration procedure are known as standards, calibrators or calibration materials. They contain a known amount of the analyte(s) to be tested and are run in the instrument's calibration mode, not in the normal test mode like patient samples. The manufacturer determines how calibration should be performed and states the procedure in the instrument's operator's manual. The procedure will also include the number, type and concentration of the calibration materials to be used.
The manufacturer also determines the frequency at which calibration is to be performed, but it may be done more often if your laboratory determines this is needed. Calibration may also be performed if calibration verification fails.
Simply, calibration verification is testing materials with known analyte concentrations in the same manner as patient samples.
• Test at least three samples-low, mid-range and high-that sup the reportable range. These are run in the instrument's normal test mode, not in the calibration mode. (Note that this is different than running calibration samples.) Running the specimens once is sufficient, but running them repeatedly will provide more useful data.
• Compare the results you obtain with the known values of the specimens.
• If the results fall within the acceptable ranges, calibration is verified.
• If one or more of the values do not fall within the acceptable ranges, calibration verification failed and corrective action must be taken. This may include repeating calibration.
• Document all actions taken throughout the verification procedure and maintain documentation for at least two years.
Calibration and calibration verification have to be performed for every analyte on every applicable instrument, even if you have multiple instruments of the same type.
Calibration verification is performed at the frequency established by the manufacturer, but according to CLIA regulations, it must be done at least every six months. Also like calibration, calibration verification can be performed more frequently if you determine that it is needed more often in your laboratory.
If your laboratory's calibration procedure uses three or more levels of calibration materials (including low, mid-range and high levels) and is performed every six months, the periodic calibration verification requirement is met. However, there are other times when calibration verification has to be performed:
• if there is a change in the reagent lot number, formulation or manufacturer (Note: You do not have to perform calibration verification after a reagent lot number change if you can show that the calibration is not affected by these changes. This is demonstrated if you have already documented several consecutive instances where calibration did not have to be adjusted following changes in reagent lot numbers.);
• if major preventive maintenance has been done;
• if critical instrument parts have been replaced; and
• to troubleshoot unacceptable QC results.
A number of specimens can be used for calibration verification. Whatever you decide to use, you must know the analyte concentration and acceptability limits for each analyte. Specimens may include:
• commercially available calibration materials or linearity sets, which have known concentrations (probably the most expensive but the easiest to use);
• previously tested Proficiency Testing (PT) samples, with known results;
• control materials, with known concentrations (cannot be the same lot number in use for your routine QC); and/or
• previously tested patient samples, with known results.
When using commercially prepared samples, whether they are calibration materials, control materials or PT samples, the acceptability limits are already established. If you use previously tested patient samples, you will have to establish your own reasonable acceptability limits before you begin the verification.
Several problems could cause calibration verification to fail:
• Quality Control-check for shifts, trends and other patterns that could be due to calibration problems;
• Verification materials-ensure they are in-date and that the acceptability ranges are appropriate;
• Reagents-determine whether they are appropriate, in-date and/or contaminated, and look for changes in lot number, manufacturer and/or formulation;
• Instrument maintenance/service-ensure that all required maintenance has been performed at the proper time intervals and that appropriate measures were taken following major maintenance and/or service procedures;
• Instrument operators-confirm that all testing personnel are following established testing procedures;
• Environmental conditions-ensure that proper temperature, lighting, humidity, etc. have been maintained, and that there have been no fluctuations in electricity.
You might be able to take simple steps and repeat the verification procedure. No further testing is required if this second verification is acceptable. If troubleshooting and repeat testing do not result in acceptable calibration verification, the instrument will have to be recalibrated.
Teresa A. Scott is COLA's Education Division Medical Technologist.
CLIA Regulations, subpart K-Quality Systems for Nonwaived Tests, section 493.1255-Standard: Calibration and Calibration Verification.