Background. Even though the delta check is a valuable tool for detecting individual random errors, it gives too many false positive results, creating a heavy burden to find the causes and correct the errors. To reduce this burden, the authors propose new criteria and compare their efficiency with previously reported methods. Methods. According to tile new criteria of the 'modified combined absolute delta percent change method (MCAD)', the result is considered to be positive when more than 4 among 14 items are abnormal in the absolute delta percent change method (ADPC). We compared this with well known methods including ADPC, delta percent change (DPC), delta difference (DD), rate percent change (RPC), rate difference (RD), allowable limit delta check (AL) and differential application method (DA). We used a model to compare the efficiencies in detecting specimen mix-up and used 54,779 clinical chemistry results to compare the delta detection rates. Results. RPC and MCAD showed the highest efficiencies of 76.8% and 75.7% in detecting specimen mix-up. In evaluation of workload with actual clinical chemistry data, the delta detection rates were 3.6% in ADPC, 1.7% in DD, DPC and RPC, 1.6% in AL, and 1.3% in RD. Conclusions. Because RPC and MCAD showed the highest efficiencies in detecting specimen mix-up with acceptable sensitivity and lowered workload compared to ADPC and DA, we conclude that RPC and MCAD are the most effective methods.
|Number of pages||5|
|Journal||European Journal of Laboratory Medicine|
|Publication status||Published - 1998|
ASJC Scopus subject areas