Q. We are a GMP laboratory and are setting up a pipette calibration program. Do we need “as-found” data? If so, how do we collect it?
A. While not every laboratory needs “as-found” or “as-received” data, these data are crucial components of pipette calibration for those that must comply with GMP regulations or international standards, such as ISO 17025. As-found data are gathered prior to any preventive maintenance or calibration. These data are used to document the performance of the instrument since its last recorded service event. Each pipette is first identified by serial number for record-keeping purposes and then tested for inaccuracy and imprecision against defined tolerances. The as-found test should be conducted using at least three samples per volume at two to three different volume settings (e.g., typically 10%, 50%, and 100% of nominal volume). A statistical analysis of these data produces the as-found pass/fail status of the instrument. Pipettes that pass their as-found analysis can be assumed to have been within tolerance since their last documented calibration event, especially when supported by a performance verification program; pipettes that fail their as-found test should be assumed to have been performing outside of tolerance since their last recorded service event.
GMP regulations require that equipment managers ensure adequate and continuous performance of measurement equipment with respect to accuracy and precision. ISO 17025 requires that out-of-tolerance equipment and the measurements produced with that equipment be investigated to determine what impact, if any, the use of the out-of-tolerance equipment had on customer data and internal processes. These standards, and others, require that organizations conduct studies to determine the consequences of having used out-of-tolerance equipment upon their products, process, data, and results. Any lab that requests and documents as-found, instrument performance data needs to have a plan in place to address as-found failures. As-found failures should generate a series of activities (i.e., impact investigation) to determine the degree to which the instrument was out of tolerance and what impact, if any, that degree of out-of-tolerance had upon the work performed since the pipette was last documented to be within tolerance. The goal of the impact investigation is to determine whether or not the out-of-tolerance pipette could have produced data significantly out of tolerance to the company’s process. All data produced with the out-of-tolerance pipette between calibration events may be suspect; the longer the interval between calibration events, the more data that must be reviewed and analyzed. Many labs also use as-found data statistics for performance trending and adjust their calibration program accordingly to eliminate as-found failures altogether.
If you have a problem with your equipment or process, an industry expert may have the solution. Please send your question to Jennifer Markarian, editor of Equipment and Processing Report, and we may be able to provide an answer in a future issue. All questions will remain anonymous.