English 
Italian La stima dell’incertezza delle misure nel laboratorio clinico
AUTORI
1ASL di Vallecamonica-Sebino, Laboratorio di Patologia Clinica, Presidio Ospedaliero di Edolo, Edolo (BS)
2Azienda Ospedaliera Universitaria Senese, UOC Patologia Clinica, Policlinico Santa Maria alle Scotte, Siena
3ASL4 Chiavarese, Laboratorio Analisi Ospedale Civile di Lavagna, Lavagna (GE)
4ASL di Vallecamonica-Sebino, Laboratorio di Patologia Clinica, Presidio Ospedaliero di Esine, Esine (BS)
4Azienda Ospedaliera Universitaria di Parma, UO Diagnostica Ematochimica, Dipartimento di Patologia e Medicina di Laboratorio, Parma
6IRCCS Ospedale San Raffaele, Servizio di Medicina di Laboratorio, Milano
ABSTRACT
Measurement uncertainty calculation in clinical laboratories
The result of a measurement is only an estimate of the value of the measurand and it is complete only when accompanied by a statement of the measurement uncertainty. The ISO 15189 standard requires that “the laboratory shall determine measurement uncertainty for each measurement procedure”. The approach to calculate the measurement uncertainty proposed by the “Guide to the expression of uncertainty in measurement” (GUM) requires the quantification of every source of variability to sum them up for the final calculation (“bottom-up” approach). To overcome inherent difficulties in the systematic application of this approach in a clinical laboratory, a “top-down” approach is proposed, through the calculation of measurement uncertainty from already existing data of IQC. The proposed approach is checked by applying it to the IQC data for serum glucose and creatinine measurements collected from 19 clinical laboratories. Different approaches to cope with the issue of the estimate of systematic error (bias) are proposed, based either on value-assigned trueness control/reference materials, on the mean value of the employed material defined by the laboratory at the start of the IQC program or on the peer group mean of an interlaboratory program material. The availability of a standardized way to estimate the measurement uncertainty provides a tool to evaluate the analytical quality of results and it allows comparison of the quality of results made available by different laboratories.