Do accredited laboratories perform better in proficiency testing than non-accredited laboratories?

The CALA proficiency testing programme has been in operation since 1991. As CALA is also an accreditation body, every result reported as part of the CALA PT schemes is identified as being from a CALA accredited laboratory or not from a CALA accredited laboratory. For this examination, a data set is a group of analytical results for a single measurand, from a single test item, from a single PT round. Following the identification of data sets that contain a minimum of ten results from an accredited laboratory, and ten results from a non-accredited laboratory, the remaining data comprised 252 analyte/matrix combinations, 14 844 individual data sets and 1 124 630 participant results. Assigned values and standard deviations of proficiency assessment were calculated using Algorithm A from ISO 13528, and z-scores calculated for all reported results, regardless of accreditation status. Comparisons were made of z-scores between accredited and non-accredited laboratories for the data as a whole, as well as for subsets of the data partitioned into groups such as inorganic tests, organic tests and microbiological tests. In all cases examined, the percentage of both Questionable and Unsatisfactory performance was higher for non-accredited laboratories than for accredited laboratories. A comparison was also made between accredited and non-accredited laboratories of the percentage of consecutive Unsatisfactory performance for the same analyte over a ten year period. Finally, comparisons were made for the first, second and third time participation to assess the impact of PT scheme familiarity. This examination shows that accredited laboratories continually outperform non-accredited laboratories as a group.