Evaluating sample stability in the clinical laboratory with the help of linear and non-linear regression analysis

Abstract

As it is common practice for laboratories to store patient samples for a predefined period, allowing clinicians to request additional tests on previously collected samples, knowledge about sample stability is indispensable for the laboratorian. A common approach to estimating the maximum storage time is to use a discrete study design, measuring the analyte of interest at various time-points and then checking for significant differences with the help of a statistical test, such as Student’s t-test, Wilcoxon’s test or an analysis of variance (ANOVA) test. Because only discrete time intervals are considered, stability data can just be approximated. Alternatively, a continuous study design, as described by the Clinical and Laboratory Standards Institute (CLSI) for performing stability experiments for in vitro diagnostic reagents, can also be adopted by the clinical laboratory to evaluate the stability of biological samples. The major advantage of this approach is that it allows laboratories to define individual stability limits for different medical situations and offers more flexibility when choosing time-points for measurements. The intent of this paper is to demonstrate the evaluation of sample stability in the clinical laboratory with a continuous study design implemented with linear or non-linear regression analysis. Appropriate statistical modeling and acceptance criteria are presented, stability functions are described briefly, and checking the overall validity of the results is discussed.

The CRESS checklist for reporting stability studies: on behalf of the European Federation of Clinical Chemistry and Laboratory Medicine (EFLM) Working Group for the Preanalytical Phase (WG-PRE)

“…PD% a × time + b × time 2 + c × time 3 + ⋯ + n × time n This complex equation will fit to non-linear behaviour observed for some analytes [4,22], but is difficult to implement. A first-order equation could fit properly for many equations.…”

Impact of Individualized Hemolysis Management Based on Biological Variation Cut-offs in a Clinical Laboratory

“…We used the interferogram dates to calculate the straight-line trend value and R 2 regression coefficient. In all cases, the selected regression model showed R 2 >0.98 [ 16 ]. Interferograms for the five selected analytes were generated using the data described in Fig.…”

“…The RCV was calculated according to the following formula [16]: RCV = square root(2) × 1.96 × (square root((CVI) 2 +(CVA) 2 )), where CVI represents the within-subject BV, and CVA represents the analytical variation in a given laboratory for a given method. CVA was obtained from data collected from an internal quality control process of the laboratory over a period of six months prior to the beginning of the study, using the following formula: CVA = (SD/mean) × 100.…”

Automated data analytics workflow for stability experiments based on regression analysis

“…In its mathematical essence, the problem of estimating stability times reduces to solving a quadratic equation and obtaining at least one real-valued, positive solution. The solutions for the quadratic equation in (13) are presented in [1] , [6] . However, none of the aforementioned publications provides guidance on how to verify if the mathematically valid solution for (13) also corresponds to the solution of the underlying substantial problem which is finding the correct stability time.…”