Statistics graphics for the laboratory Applications Analytical

Зарегистрируйтесь, чтобы просмотреть полный документ!

Statistics & graphics for the laboratory Applications Analytical quality Dietmar Stöckl Dietmar@stt-consulting. com Linda Thienpont Linda. thienpont@ugent. be In cooperation with AQML: D Stöckl, L Thienpont & • Kristian Linnet, MD, Ph. D Linnet@post 7. tele. dk • Per Hyltoft Petersen, MSc Per. hyltoft. petersen@ouh. fyns-amt. dk • Sverre Sandberg, MD, Ph. D Sverre. sandberg@isf. uib. no Statistics & graphics for the laboratory. Courses-Overview

Prof Dr Linda M Thienpont University of Gent Institute for Pharmaceutical Sciences Laboratory for Analytical Chemistry Harelbekestraat 72, B-9000 Gent, Belgium e-mail: linda. thienpont@ugent. be STT Consulting Dietmar Stöckl, Ph. D Abraham Hansstraat 11 B-9667 Horebeke, Belgium e-mail: dietmar@stt-consulting. com Tel + FAX: +32/5549 8671 Copyright: STT Consulting 2007 Statistics & graphics for the laboratory 2

Content Metrology Analytical quality specifications ("Goals") Method evaluation/method comparison Statistics & graphics for the laboratory 3

Content Detailed content Metrology • • Terminology & definitions Metrological concepts (Error, Accuracy, Uncertainty) Measurement traceability Conclusion Annex • • • Glossary Système International d'Unités: base units Metrological concepts Traceability requirements of Directive 98/79/EC How to meet the traceability requirement? The elements of a reference measurement system Validation of metrologically traceable calibration What if SI-traceability does not apply? Traceability – to which extent? Additional references Analytical quality specifications ("Goals") • Introduction • Concepts – Clinical concepts – Questionnaires to clinicians – Goals from biology – Goals from experts – State-of-the-art • Comparison of goals • Comparison "state-of-the-art" with goals • Analytical goals – Translation into practice • Goals – future vision • Outlook • References Statistics & graphics for the laboratory 4

Content Detailed content Method evaluation/method comparison Introduction • The analytical quality triangle • Purpose of method evaluation Performance characteristics of a method. Precision • Limit of detection • Working range … Method evaluation strategies Assessment of performance characteristics • Specific protocols • Stability/ruggedness • Multifactor protocols • Method comparison • Summary of protocols, statistics & graphics • Method comparison – The stable basis Exercises with EXCEL-file References Interactive part • Factors that influence the interpretation of a method comparison: qualitative use of a difference (Bland & Altman) plot Final remark • When to use regression-based interpretation Exercises • Case studies 1 - 5 Statistics & graphics for the laboratory 5

Content Metrology • Terminology & definitions • Metrological concepts – Error – Accuracy – Uncertainty – Nature and measures of error • Measurement traceability Statistics & graphics for the laboratory 6

Terminology and definitions For understanding metrology ("science of measurement"), one needs to know a lot of definitions and to use the correct metrological terms (quantity; unit; traceability; accuracy; uncertainty; …). A few of them will be presented in this part, however, most of them can be found in the Annex. The measurement Measurement (in general) • process of experimentally obtaining one or more quantity values that can reasonably be attributed to a quantity Quantity and value i. Quantities are length, mass, amount-of-substance, time, temperature, etc. ii. The value of a quantity is expressed by both a number and a unit The actual measurement We determine the value of a particular quantity. Example of a particular quantity: amount-of-substance concentration of glucose in plasma. The particular quantity is specified by 3 elements: • System: Plasma • Component (also called analyte): Glucose • Kind-of-quantity: Amount-of-substance concentration NOTE: The particular quantity subject to measurement is called measurand. Measurand: quantity intended to be measured. A full measurement report The full report of a measurement result for glucose would read: “the amount-ofsubstance concentration of glucose in plasma was 5. 2 (= number) mmol/L (unit)” Measurement units We have to distinguish between SI-units and International units. Système International d’Unités (SI): • Only meaningful in connection with components (analytes) whose elementary entity can be recognized by full physicochemical characterization. • For components where the elementary entity is not known (identity and/or purity not known), arbitrary units have to be used to express quantities. International Units (IUs): • In case that an internationally accepted calibrator and/or measurement procedure available, the quantity of the above components can be expressed in “IUs”. • Characteristically, IUs depend on the measurement procedure and the calibrator used, in contrast to SI units, which are independent thereof. Units must be "materialized" in standards ! Statistics & graphics for the laboratory 7

Terminology and definitions What is a standard? (Measurement) standard: realization of the definition of a given quantity, with stated quantity value and measurement uncertainty, used as a reference. Usually, measurements have to be calibrated with standards. Calibration of measurement Calibration: operation that, under specified conditions, in a first step establishes a relation between the quantity values with measurement uncertainties provided by measurement standards and corresponding indications with associated measurement uncertainties and, in a second step, uses this information to establish a relation for obtaining a measurement result from an indication. >Calibration relates the measured values of particular quantities with the values defined by standards. Example: Amount-of-substance concentration of serum sodium. Proper calibration is the major factor for "overall" trueness. The main input elements for proper calibration are analyte preparations with certified content and matrix-free or matrix-corrected calibration solutions. The actual calibration needs to be controlled by: • Written calibration protocols • Sufficient calibration intervals • Documentation of calibration status These elements should be covered by the quality system of a laboratory. Statistics & graphics for the laboratory 8

Metrological concepts Quality of measurements The 3 metrological concepts from ISO Error concept: BIPM, IEC, IFCC, ISO, IUPAC, IUPAP, OIML. Vocabulaire International des Termes Fondamentaux et Généraux de Métrologie. 3 rd ed. Geneva: ISO, 2007. Accuracy concept: International Organization for Standardization. Accuracy (trueness and precision) of measurement methods and results. Part 1: General principles and definitions, ISO 5725 -1. 1 st ed. Geneva: ISO, 1994. Uncertainty concept: International Organization for Standardization. Guide to the expression of uncertainty in measurement. 1 st ed. Geneva: ISO, 1993. Important terms related to the 3 concepts Graphical presentation of measurement quality Because the uncertainty concept is relatively new in analytical chemistry, some of its main features will be explained below. Statistics & graphics for the laboratory 9

Metrological concepts Measurement uncertainty (GUM) Uncertainty Parameter characterizing the dispersion of the quantity values being attributed to a measurand, based on the information used. The GUM philosophy • Deprecates the terms "error" and "truth" • Deprecates the distinction between "systematic" and "random" • Requires the correction of all "biases" Systematic versus random Every calibration introduces a small systematic error. Over time, however, these errors should be Normal distributed and can be viewed as random calibration bias: >Classification systematic/random may depend on the observation time! Propagation of systematic error GUM Correct: after correction, [random] uncertainty remains Classical Errors are summed (sign is respected) Sum 1 2 3 What is GUM all about? It is about propagation of random errors! Calculation may require statistical experience (see below). For a detailed treatise of the GUM concept, inclusive examples see: EURACHEM/CITAC Guide: Quantifying uncertainty in analytical measurement (http: //www. measurementuncertainty. org/mu/guide/index. html). Statistics & graphics for the laboratory 10

Metrological concepts Rules for propagation of random error The standard deviation (s) of calculated results (propagation of s) 1. Sums and differences y = a(±sa) + b(±sb) + c(±sc) sy = SQRT[sa 2 + sb 2 + sc 2] (SQRT = square root) Do not propagate CV! 2. Products and quotients y = a(±sa) • b(±sb) / c(±sc) sy/y = SQRT[(sa/a)2 + (sb/b)2 + (sc/c)2] 3. Exponents (the x in the exponent is error-free) y = a(±sa)x sy/y = x • sa/a For GUM, see also: http: //physics. nist. gov/cuu/Uncertainty/index. html http: //physics. nist. gov/Pubs/guidelines/ http: //www. measurementuncertainty. org http: //www. westgard. com/guest 41. htm Thienpont LM. Calculation of measurement uncertainty–why bias should be treated separately. Clin Chem 2008; 54: 1587 -8. Uncertainty of the final result Input variables • Calibration traceability & lot variation • Linearity limits • Recovery limits • Reagent lot-/instrument variation • Drift limits (recommended IQC procedure) • Specificity (“cross-reactivity”) • Interference limits (all kinds; continuous update): variation in sample matrix; common (lipemia, etc. ); drug effects; auto-/heterophilic antibodies; genetic variants Note that, currently, NOT all can be included in a GUM uncertainty statement! In particular, statistical concepts are missing of how to deal with unspecificity and interferences. Statistics & graphics for the laboratory 11

Metrological concepts Metrological concept used in the course Here, we apply the error concept, including the statistical uncertainty of the experimental estimates of error [Stöckl D. Scand J Clin Lab Invest 1996; 56: 193 -7] Nature & measures of error Random error (RE) • Quantitative expression: SD/CV Systematic error (SE) • Quantitative expression: Bias – Constant – Proportional Random & systematic error • Quantitative expression: Total error (TE)/Uncertainty Statistics & graphics for the laboratory 12

Metrological concepts Random error Nature of imprecision • Imprecision is an inevitable characteristic of each measurement. It results in a characteristic dispersion of results (usually Gaussian) of repeated measurements, even when carried out under apparently identical test conditions. • The standard deviation (or coefficient of variation) is a direct measure for imprecision. Note: Characteristic of imprecision is that results always lie on both sides (±) of a mean value. Input factors for good precision are: • Sufficient pipetting volume • Positive displacement volumetric equipment • Gravimetric control of volumes • Good signal-to-noise ratio • Adequate scales • Sufficient reported digits • Automatisation (and quality thereof) • Control of environment Total analytical imprecision (sa, tot) comprises • Sampling (ssam), • Sample preparation (sprep), and • Measurement (smeas) Total analytical imprecision is calculated as: • sa, tot = SQRT[(ssam)2 + (sprep)2 + (smeas)2] (SQRT = square root) The total imprecision is dominated by the step with the highest imprecision For automatic analyzers, these 3 are not distinguished: then, imprecision is understood to consist of all 3 components. Evaluation of measurement imprecision makes a difference between: • Total • Between-day (between-run) • Within-day (within-run) Note: To be representative, the within-day imprecision has to be calculated from cumulated data of several days. SD and CV in analytical practice Common situation CV constant/SD decreasing down to a certain concentration, then SD constant and CV increasing Statistics & graphics for the laboratory 13

Metrological concepts Systematic error (Un)trueness (systematic error) of analytical methods may have a variety of origins: • Calibration errors • Unspecificity • Interferences • System drift/shift • Carryover A measure for (un)trueness/ systematic error is the bias, observed as deviation of the actual measurement result from a target (or expected value, or reference value). Control of systematic errors • Reference methods and materials • Apply proper calibration (major factor for overall analytical trueness) • Adequate data processing (e. g. , calibration function) • Select measurement principle that is insensitive to interferences and specific and/or apply sample purification • Select stable, high quality instrumentation (low drift/shift/carryover) • Apply internal quality control Note: Characteristic of (un)trueness is that results tend to lie on either side of a target value (either + or -). Opposite to imprecision, bias can be obviated. Types of systematic error • Constant & • Proportional Statistics & graphics for the laboratory 14

Metrological concepts Total error Graphical presentation of the total error From: Westgard JO, Barry PL. Cost-effective quality control: managing the quality and productivity of analytical processes. 4 th printing. Washington: AACC Press, 1995 Components of total error (TE) TE comprises • Imprecision • Calibration errors (including non-linearity) • Unspecificity • Interference (all kinds) • Carryover • System instability (drift; shift) Measures of total error (TE) We can distinguish 2 situations for the measure of TE TE for a single measurement: TE = SE + z • RE, or TE = SE + z • s z is usually set to 1. 96 or 2. 58, encompassing 95% or 99% [two-tailed] of a gaussian distributed population. Note: s shall be sufficiently reliable (n 30) TE for multiple measurements: TE = SE + z • [s/ n], or TE = SE + t n-1 • [s/ n] Note that the formula uses the confidence interval for the actual number of measurements (n) (instead of z • s). If s is derived from at least 30 replicate measurements, z (at a given confidence level) is used; if s is derived from the n actual measurements, tn-1 = the student's tvalue for (n-1) degrees of freedom is used. Statistics & graphics for the laboratory 15

Measurement traceability From the definition of a measurement: • "Determine the value of a quantity" • The value of a quantity is expressed by both a number and a unit • The unit must be meaningful one can infer that an actual measurement result should be traceable [preferably] to the SI-unit. Recently, the process of establishment of traceability has been formalized. Measurement traceability – The regulations European Directive 98/79/EC • Traceability requirement EN/ISO 17511: 2003. In vitro diagnostic medical devices – Measurement of quantities in biological samples – Metrological traceability of values assigned to calibrators and control materials • Traceability establishment Selected contents of EN/ISO 17511 4 Metrological traceability chain and calibration hierarchy 6 Expression of uncertainty of measurement 5 Calibration transfer protocols 7 Validation of metrologically traceable calibration Traceability chains ISO 17511 elaborates 5 different traceability models (see table), depending on the definition of the analyte and the existence of a reference measurement system. Statistics & graphics for the laboratory 16

Measurement traceability Metrological traceability chain (SI) (ISO 17511) The metrological traceability chain starts with the definition of the analyte and its unit. The unit must be materialized in a standard. The unit is traced to the routine measurement procedure by a cascade of hierarchically different materials and methods. The methods are used for value assignment and the materials are used for calibration according to the above scheme. Example testosterone Statistics & graphics for the laboratory 17

Measurement traceability Testosterone - Outcome Calibration traceability Method comparison with 50 human samples Assess regression estimates vs 17511 specifications: • Slope = 1 • Intercept = 0 (“some tolerance”) Result traceability Variation around line: “ 17511 alternative” Manufacturer limit: maximum allowable deviation Analytical specifications from biological variation (www. westgard. com) • CV = 4. 7% • Bias = 6. 4% • Total error = 14% (z = 1. 65 for imprecision) Suppose the manufacturer selects the TE goal for the validation study = 14% Variation around line, manufacturer limit • 14% (“biology”) down to 5 nmol/l • At 5 nmol/l constant absolute deviation, i. e. 14% of 5 nmol = 0. 7 nmol Statistics & graphics for the laboratory 18

Measurement traceability Traceability – To which extent? Currently, no generally accepted answer, only models how to generate the "numbers". Models, in hierarchical order# • Clinical concepts • Concepts based on biological variation • Expert opinion • Regulations • "State-of-the-art" #Consensus Statement (Stockholm 1999). Scand J Clin Lab Invest 1999; 59: 585. Statistics & graphics for the laboratory 19

Conclusion Which quality? • Document your own quality • Compare it with the “state-of-the-art” Identify therefrom “problem analytes” • Compare it with international goals Identify therefrom “problem analytes” or “problem goals” • Identify “problem analytes” by communication with the laboratories/clinicians • Identify “possible” problem analytes by comparison with “biological” specifications: biological variation, reference intervals • Verify “possible” problem analytes by comparison with “clinical” specifications • Check whether you want to improve the quality of some tests, independent of proposed goals Conclusion For valid measurements … adhere to the analytical quality triangle! Statistics & graphics for the laboratory 20

Notes Statistics & graphics for the laboratory 21

Annex - Metrology Content Glossary Système International d'Unités: base units Metrological concepts Traceability requirements of Directive 98/79/EC How to meet the traceability requirement? The elements of a reference measurement system Validation of metrologically traceable calibration What if SI-traceability does not apply? Traceability – to which extent? Additional references Statistics & graphics for the laboratory 22

Annex - Metrology Glossary Metrology [1] field of knowledge concerned with measurement Measurand [1] quantity intended to be measured Quantity [1] property of a phenomenon, body, or substance, to which a number can be assigned with respect to a reference Measurement [1] process of experimentally obtaining one or more quantity values that can reasonably be attributed to a quantity Notes: • Quantities are length, mass, amount-of-substance, time, temperature, etc. • The value of a quantity is expressed by both a number and an unit • The full specification of the quantities measured in the medical laboratory comprises three elements: System (e. g. , blood plasma) Component (also called analyte) (e. g. , glucose) Kind-of-quantity (e. g. , amount-of-substance concentration) The full report of a glucose measurement would read: “the amount-of-substance concentration of glucose in blood plasma was 5. 2 mmol/L” Measurement unit [1] scalar quantity, defined and adopted by convention, with which any other quantity of the same kind can be compared to express the ratio of the two quantities as a number Value of a quantity [1] number and reference together expressing magnitude of a quantity EXAMPLE: Length of a given rod: 5. 34 m Measurement standard [1] realization of the definition of a given quantity, with stated quantity value and measurement uncertainty, used as a reference EXAMPLE: 1 kg mass standard. Statistics & graphics for the laboratory 23

Annex - Metrology Glossary Error [1] difference of measured quantity value and reference quantity value Systematic error [1] component of measurement error that in replicate measurements remains constant or varies in a predictable manner Bias [1] systematic measurement error or its estimate, with respect to a reference quantity value Random error [1] component of measurement error that in replicate measurements varies in an unpredictable manner Trueness [1] closeness of agreement between the average of an infinite number of replicate measured quantity values and a reference quantity value Accuracy [1] closeness of agreement between a measured quantity value and a true quantity value of the measurand Precision [1] closeness of agreement between indications obtained by replicate measurements on the same or similar objects under specified conditions Repeatability condition [1] condition of measurement in a set of conditions that includes the same measurement procedure, same operators, same measuring system, same operating conditions and same location, and replicate measurements on the same or similar objects over a short period of time Reproducibility condition [1] condition of measurement in a set of conditions that includes different locations, operators, measuring systems, and replicate measurements on the same or similar objects Uncertainty [1] parameter characterizing the dispersion of the quantity values being attributed to a measurand, based on the information used [Metrological] Traceability [1] property of a measurement result whereby the result can be related to a stated reference through a documented unbroken chain of calibrations, each contributing to the measurement uncertainty Statistics & graphics for the laboratory 24

Annex - Metrology Glossary Commutability [of a reference material] [1] property of a reference material, demonstrated by the closeness of agreement between the relation among the measurement results for a stated quantity in this material, obtained according to two given measurement procedures, and the relation obtained among the measurement results for other specified materials Matrix effect [2] Influence of a property of the sample, other than the measurand, on the measurement of the measurand according to a specified measurement procedure and thereby on its measured value [2] Influence quantity [1] quantity that, in a direct measurement, does not affect the quantity that is actually measured, but affects the relation between the indication and the measurement result Note: Specificity & Interference are not yet unequivocally defined by ISO. Selectivity [1] capability of a measuring system, using a specified measurement procedure, to provide measurement results, for one or more measurands, that do not depend on each other nor on any other quantity in the system undergoing measurement (= specificity in chemistry) Interference [in analysis] A systematic error in the measure of a signal caused by the presence of concomitants in a sample (http: //goldbook. iupac. org) specific [in analysis] A term which expresses qualitatively the extent to which other substances interfere with the determination of a substance according to a given procedure. Specific is considered to be the ultimate of selective, meaning that no interferences are supposed to occur (http: //goldbook. iupac. org). Calibration [1] operation that, under specified conditions, in a first step establishes a relation between the quantity values with measurement uncertainties provided by measurement standards and corresponding indications with associated measurement uncertainties and, in a second step, uses this information to establish a relation for obtaining a measurement result from an indication Sensitivity [1] quotient of the change in the indication and the corresponding change in the value of the quantity being measured Statistics & graphics for the laboratory 25

Annex - Metrology Glossary Linear range Concentration range over which the intensity of the signal obtained is directly proportional to the concentration of the species producing the signal (http: //goldbook. iupac. org). Linearity (generic) Ability of an analytical procedure to produce test results which are proportional to the concentration (amount) of an analyte, either directly or by means of a well-defined mathematical transformation. Working interval [1] set of values of the quantities of the same kind that can be measured by a given measuring instrument or measuring system with specified instrumental uncertainty, under defined conditions Limit of detection (in analysis) The limit of detection, expressed as the concentration, c. L, or the quantity, q. L, is derived from the smallest measure, x. L, that can be detected with reasonable certainty for a given analytical procedure. The value of x. L is given by the equation x. L = xbi + k • sbi, where xbi is the mean of the blank measures, sbi is the standard deviation of the blank measures, and k is a numerical factor chosen according to the confidence level desired (http: //goldbook. iupac. org). Limit of detection [1] measured quantity value, obtained by a given measurement procedure, for which the probability of falsely claiming the absence of a component in a material is β, given a probability α of falsely claiming its presence Ruggedness (generic) Ability to reproduce the method in different laboratories or in different circumstances. Ruggedness (USP) Degree of reproducibility of the results obtained under a variety of conditions, expressed as %RSD. These conditions include different laboratories, analysts, instruments, reagents, days, etc. Robustness (ICH Q 2 A 1995) The robustness of an analytical procedure is a measure of its capacity to remain unaffected by small, but deliberate variations in method parameters and provides an indication of its reliability during normal usage. [1] BIPM, IEC, IFCC, ISO, IUPAC, IUPAP, OIML. Vocabulaire International des Termes Fondamentaux et Généraux de Métrologie. 3 rd ed. Geneva: ISO, 2007. [2] EN/ISO 17511: 2003. In vitro diagnostic medical devices – Measurement of quantities in biological samples – Metrological traceability of values assigned to calibrators and control materials. [3] See also: www. clsi. org>Harmonized Terminology Database Statistics & graphics for the laboratory 26

Annex - Metrology Système International d'Unités: base units Quantity Name Symbol Length meter m Mass kilogram kg Time second s Electric current ampere A Thermodynamic temperature kelvin K Amount of substance mol Luminous intensity candela cd Catalytic amount katal kat Metrological concepts • Error concept [1] • Trueness concept [2] • Uncertainty concept [3] [1] BIPM, IEC, IFCC, ISO, IUPAC, IUPAP, OIML. Vocabulaire International des Termes Fondamentaux et Généraux de Métrologie. 3 rd ed. Geneva: ISO, 2007. [2] International Organization for Standardization. Accuracy (trueness and precision) of measurement methods and results. Part 1: General principles and definitions, ISO 5725 -1. 1 st ed. Geneva: ISO, 1994. [3] International Organization for Standardization. Guide to the expression of uncertainty in measurement. 1 st ed. Geneva: ISO, 1993. Statistics & graphics for the laboratory 27

Annex - Metrology Traceability requirements of Directive 98/79/EC Some excerpts ANNEX I: Essential Requirements A. General requirements 3. The devices must be designed and manufactured. . . taking account of the generally acknowledged state of the art. They must achieve the performances, in particular, where appropriate, in terms of analytical sensitivity, diagnostic sensitivity, analytical specificity, diagnostic specificity, accuracy, repeatability, reproducibility, including control of known relevant interference, and limits of detection, stated by the manufacturer. The traceability of values assigned to calibrators and/or control materials must be assured through available reference measurement procedures and/or available reference materials of a higher order. B. Design and manufacturing requirements 4. Devices which are instruments or apparatus with a measuring function 4. 1. Devices which are instruments … must be designed and manufactured in such a way as to provide … accuracy of measurement within appropriate accuracy limits, taking into account … available and appropriate reference measurement procedures and materials. 8. Information supplied by the manufacturer 8. 7. Where appropriate, the instructions for use must contain the following particulars: (h) the measurement procedure to be followed with the device including as appropriate: - … information about the use of available reference measurement procedures and materials by the user; (k) information appropriate to users on: the traceability of the calibration of the device; ANNEX III: EC Declaration of Conformity 3. The technical documentation. . . must include in particular: - adequate performance evaluation data showing the performances claimed by the manufacturer and supported by a reference measurement system (when available), with information on the reference methods, the reference materials, … Statistics & graphics for the laboratory 28

Annex - Metrology How to meet the traceability requirement? EN/ISO 17511: 2003 In vitro diagnostic medical devices – Measurement of quantities in biological samples – Metrological traceability of values assigned to calibrators and control materials Some excerpts related to the traceability chain and how it works Introduction Objective of applying the traceability chain … is to transfer the degree of trueness of a reference material, and/or reference measurement procedure, to a procedure that is of a lower metrological order, e. g. a routine procedure Trueness of measurement … depends on the metrological traceability of the value through an unbroken chain of alternating measurement procedures and measurement standards (calibrators), usually having successively decreasing uncertainties of measurement. Uncertainty of the value … depends on the stated metrological traceability chain and the combined uncertainties of its links. The measurement of quantities in biological samples requires reference measurement systems including: • the definition of the analyte in the biological sample with regard to the intended clinical use of the measurement results; • a reference measurement procedure for the selected quantity in human samples; • suitable reference materials for the selected quantity, e. g. primary calibrators and secondary matrix-based calibrators that are commutable. • To ensure the validity of a metrological traceability chain, the quantity shall be the same at all levels. Statistics & graphics for the laboratory 29

Annex - Metrology The elements of a reference measurement system EN/ISO 17511: 2003 4. 2. 2 a) Definition of measurand including SI unit of measurement 4. 2. 2 b) Primary reference measurement procedure$ shall be based on a principle of measurement proved to be analytically specific, providing metrological traceability to an SI unit of measurement without reference to a calibrator for the same quantity, and having a low uncertainty of measurement (isotope dilution-mass spectrometry, coulometry, gravimetry, titrimetry). $ Typical examples of measurement principles Method principle Analyte ID-MS Ca, K ID-GC/MS Cholesterol, glucose, creatinine, … steroid- & thyroid hormones, drugs FAES K, Na AAS Ca, Mg, Li, Ni Ion chromatography Phosphate Ion exchange-gravimetry Na LC-MS, CZE after HPLC Hb. A 1 c Potentiometry p. H Coulometry Cl Absorption photometry Hb, bilirubin, cholesterol, protein See also: IFCC. www. ifcc. org. BIPM. Joint Committee on Traceability in Laboratory Medicine (JCTLM) . www 1. bipm. org/en/committees/jc/jctlm. 4. 2. 2 c) Primary calibrator that is an embodiment of the unit of measurement with the smallest achievable uncertainty of measurement. The primary calibrator shall have its value assigned either directly by a primary reference measurement procedure or indirectly by determining the impurities of the material by appropriate analytical methods. The material usually is highly purified containing a physico-chemically well-defined analyte, examined for stability, compositional integrity, and accompanied by a certificate (certified reference material, CRM). Statistics & graphics for the laboratory 30

Annex - Metrology The elements of a reference measurement system (ctd) 4. 2. 2 d) Secondary reference measurement procedure shall describe a measuring system which is calibrated by one or more primary calibrators… See, also: EN/ISO 15193: 2002 In vitro diagnostic medical devices – Measurement of quantities in biological samples – Requirements and layout of reference measurement procedures [Secondary] Reference measurement procedure Thoroughly investigated measurement procedures shown to have an uncertainty of measurement commensurate with their intended use, especially in assessing the trueness of other measurement procedures for the same quantity and in characterizing reference materials. Validation of metrologically traceable calibration EN/ISO 17511: 2003 7. 1 The following conditions shall apply to the concept of metrologically traceable calibration: a) The reference and routine measurement procedures measure the same quantity. b) The mathematical relationship between the measurement results generated by the routine procedure and the measurement results generated by a higher order measurement procedure, is the same for all relevant human samples. c) The mathematical relationship between the measurement results generated by the measurand in a given calibrator using the reference and the routine procedure is the same as the relationship expected for measurands in routine human samples. This assumption has been termed commutability of the reference material (see 3. 9). NOTE 1 The object of the use of metrologically traceable calibrators in routine measurement procedures, such as those of in vitro diagnostic medical devices, is to produce a result of measurement of the measurand that is as close as required to that which would have been obtained if the reference measurement procedure to which the calibrators are metrologically traceable had been applied to the samples. Thus, the trueness of results given by a calibrated routine measurement procedure derives from that of the reference measurement procedure when such is available. NOTE 2 When the conditions a), b), and c) do not apply, the use of a manufacturer's product calibrator with assigned value cannot guarantee that the routine results are metrologically traceable to the reference measurement procedure. 7. 2 The commutability of the manufacturer's working calibrator(s) … shall be assessed by the manufacturer applying both reference measurement procedure … and the routine measurement procedure …to the manufacturer's working calibrator and to a set of relevant human (routine) samples. Statistics & graphics for the laboratory 31

Annex - Metrology Validation of metrologically traceable calibration (ctd) 7. 3 The commutability of the manufacturer's product calibrator, shall be demonstrated by comparing the results of measurements, made by both the reference procedure and the calibrated routine procedure on a set of actual samples of a type to which the routine measurement procedure is intended to be applied. The samples shall be authentic, preferably single-donation and unspiked human samples … 7. 4 …For metrological traceability to be achieved, the results by the routine procedure shall be related to those of the reference procedure by, e. g. a linear regression of unit slope and zero intercept with a stated probability. NOTE If linear regression is used, the observed value of the slope should be stated, including its uncertainty. A unit slope is expected but a deviation from unit slope within a stated interval of quantity values may be tolerable. … The observed value of the intercept should be stated. If a value significantly different from zero at a given probability is considered tolerable, the reasons for this shall be stated…An intercept on the axis of the routine measurement procedure significantly different from zero can indicate a difference of analytical specificity between the two procedures, which could invalidate the principle of metrological traceability. The expected variability of comparison around the regression line (prediction limits) may be estimated at a given probability on the basis of the number of samples and the respective uncertainties of the two measurement procedures. Variations greater than this indicate an aberrant-sample-dependent variability in the inter-procedure relationship that invalidates metrologically traceable routine results for certain samples. Alternatively, a limit of maximum allowable relative variation between results by the reference and calibrated routine procedures may be specified by the manufacturer…. 7. 5 If a panel of human samples is used as part of the process of assigning a value to the manufacturer's product calibrator, the same panel shall not be used also to validate metrological traceability. Statistics & graphics for the laboratory 32

Annex - Metrology What if SI-traceability does not apply? EN/ISO 17511: 2003 Introduction … but the selection of steps and the level at which metrological traceability for a given value stops, depend on the availability of higher order measurement procedures and calibrators. In many cases, at present, there is no metrological traceability above the manufacturer's selected measurement procedure or the manufacturer's working calibrator. In such cases, trueness is referred to that level of the calibration hierarchy until an internationally agreed reference measurement procedure and/or calibrator becomes available. Laboratory medicine routinely provides results for 400 to 700 types of quantity… Depending on the possibility of metrological traceability to SI and on the availability of various metrological levels of measurement procedures and calibrators, the following five typical upper ends of the metrological traceability chain can be identified. a) Quantities for which results of measurements are metrologically traceable to SI. b) Quantities for which results of measurements are not metrologically traceable to SI (4 cases)… Traceability – to which extent? No numbers available! • … generally acknowledged state of the art. They must achieve the performances, …, stated by the manufacturer (IVD-Directive 98/79/EC) • … will depend on the state of development of methods of measurement and the medical uses to which the results are to be applied (EN/ISO 17511: 2003) • … should reflect the medical use (e. g. based on biological variation or other means) and the "state-of-the-art" of the quality of the IVD MDs (EN/ISO 14136: 2004) Additional references • Petersen PH, Stöckl D, Westgard JO, Sandberg S, Linnet K, Thienpont L. Models for combining random and systematic errors. Assumptions and consequences for different models. Clin Chem Lab Med 2001; 39: 589 -95. • Thienpont LM, Van Uytfanghe K, De Leenheer AP. Reference measurement systems in clinical chemistry. Clin Chim Acta 2002; 323: 73 -87. Statistics & graphics for the laboratory 33

Notes Statistics & graphics for the laboratory 34

Скачать презентацию Statistics graphics for the laboratory Applications Analytical

03f4f22aae00f1ef8b29ac41ac03e6c4.ppt

Количество слайдов: 34