- absolute error : See error.
- acceptance criteria for specimens or samples : Procedures for acceptance or
rejection of specimens or samples arriving at the analytical laboratory. Such
procedures are focused on assessing the adequacy of the chain of custody
- accreditation : Procedure by which an accreditation body formally recognizes
that a laboratory or person is competent to carry out specific tasks.
- accreditation body : Independent science-based organization that has the
authority to grant accreditation.
- accuracy : Ability to get the true result . For quantitative tests the
accuracy expresses the closeness of agreement between the true value and the
value obtained by applying the test procedure a number of times. It is affected
by systematic and random errors,
- action limit : See limit.
- aliquot : Portion of a liquid sample or solution.
- alternative hypothesis : See hypothesis testing.
- analysis : See test.
- analysis of variance (ANOVA) : Statistical technique that can be used to
separate and estimate the different causes of variation .
- analyte or target analyte : Substance to be identified or measured  .
- surrogate analyte : Well-characterized substance that is taken as
representative of the analyte .
- analftical batch : See barch.
- analytical method : See test.
- archive : Collection of documents and records purposefully stored for a
defined period of time .
- arithmetic mean or average : Sum of the individual values in a set divided
by the number of values .
- assay : Quantitative measurement of an analyte.
- assigned value : See value.
- average : See arithmetic mean.
batch or analytical batch : Group of one or more specimens or
samples that are analysed under conditions approaching repeatability. Usually it
should contain calibrators and quality control specimens or samples in addition
to the samples to be analysed.
best fit : See goodness-of-fit.
bias : Difference between the expectation of the test result
and an accepted reference value. There may be one or more systematic error
components contributing to the bias .
binomial distribution : See distribution.
blank : specimen or sample not containing the
blind specimen or sample : specimen or sample that is
analysed by an operator who is unaware at the time of the analysis that the
sample is for control purposes (based on ).
blunder : Big mistake, especially one that seems to be the
result of carelessness or stupidity . See outlier.
- calibration : Set of operations that establishes, under specified
conditions, the relationship between values indicated by a measuring instrument
or measuring system, or values represented by a material measure, and the
corresponding known values of a measurand .
- calibration curve : Relationship between the signal response of the
various concentrations of analyte in a suitable solvent or
- calibration laboratory : laboratory that performs calibrations [91.
- calibration method : Defined technical procedure for performing a
- calibrator : Pure analyte in a suitable solvent or matrix, used to prepare
the calibration curve.
- certification : procedure by which a certifying body formally recognizes
that a body, person or product complies with given specifications.
- certified reference material (CRM) : A reference material one or more of
whose property values have been certified by a technical procedure, accompanied
by or traceable to a certificate or other documentation that has been issued by
a certifying body .
- certifying body : Independent science-based organization that has the
competence to grant certifications.
- chain of custody : Procedures and documents that account for the integrity
of a specimen or sample by tracking its handling and storage from its point of
collection to its final disposition (adapted from ).
- chirality : Property of a molecule not superimposable on its mirror image.
Due to asymmetry in their structures, chiral molecules can exist as different
isomers and will have special optical and biological properties.
- chi-square distribution : See distribution.
- clerical error : See error.
- cochran test : See outlier
- coeffcient of variation or relative standard deviation : Measure used to
compare the dispersion or variation in groups of measurements. It is the ratio
of the standard deviation to the mean, multiplied by 100 to convert it to a
percentage of the average, .
- collaborative studies or interlaboratory test comparisons : Organization,
performance and evaluation of tests on the same or similar items or materials by
two or more different laboratories in accordance with predetermined conditions.
The main purpose is validation of analytical methods or establishment of
reference methods .
- comparison-of-means test : See significance test.
concentration : Amount
of a substance, expressed in mass or molar units, in a unit volume of fluid.
- confidence coefficient : See confidence level.
- confidence interval : Range of value that contains the true value at a given
level of probability. This level of probability is called the confidence
- confidence level or confidence coefficient : Measure of probability, α,
associated with a confidence interval, expressing the probability of the truth
of a statement that the interval will include the parameter value .
- confidence limits : The extreme values or end values in a confidence
interval. See limit.
- confirmatory test : Second test by an alternative chemical method for
unambiguous identification of a drug or metabolite (adapted from ).
- consensus value : See value.
- contractor : Organization that provides a service under contractual
conditions. It should be ensured that the contractor provides services in line
with specified criteria of competence.
- control chart : plot of test results with respect to time or sequence of
measurements, with limits drawn within which results are expected to lie when
the analytical scheme is in a state of statistical control .
- cusum chart : In a cusum chart each result is compared with a reference,
usually the assigned or target value. The differences from the reference are
then accumulated, respecting the sign, to give a cumulative sum of differences
from the standard. The cusum chart has the advantage of identifying small
persistent changes in the analytical scheme faster than the shewhart chart
- Shewhart chart : chart where the variable of interest is plotted against
batch or time. The observed values are compared with the expected or true value.
Lines corresponding to the mean value obtained from repficate analysis of
reference material and warning and action limits are inserted to provide
objective criteria for the interpretation of the chart ,
- control limit : see limit.
controls : specimens or samples used to
determine the validity of the calibration, that is, the linearity and stability
of a quantitative test or determination over time. Controls are either prepared
from the reference material (separately from the calibrators, that is, weighed
or measured separately), purchased, or obtained from a pool of previously
analysed specimens or samples. where possible, controls should be matrix-matched
to specimens or samples and calibrators .
- positive control : Control that contains the analyte at a concentration
above a specified limit.
- negative control : Control that contains the analyte at a concentration
below a specified limit. Usually a drug-free specimen or sample (blank) is used
as a negative control.
- corrective action : Action taken to eliminate the causes of an existing
deviation, defect or other undesirable situation in order to prevent
- correlation coefficient : Number showing the degree to which two variables
are related. Correlation coefficients range from 0 (no correlation) to -1 or +1
- cross-reacting substance : In immunoassays, a substance that reacts with
antiserum produced for the target analyte .
- cusum chart : See control chart.
- cut-off concentration : Concentration of a drug in a specimen or sample used
to determine whether the specimen or sample is considered positive or negative
. In some circumstances it is recommended that the cut-off concentration
should be set equal to the limit of detection. See
- degrees of freedom : Number of independent comparisons that can be made
between the members of a sample .
- detection limit : See limit.
- deviation : Departure from what is considered normal . See standard
- discrimination : Ability to recognize and understand the differences between
two things .
- distribution : A ranking, from lowest to highest, of the values of a
variable and the resulting pattern of measures or scores when they are plotted
on a graph (adapted from ). A frequency distribution, for example, gives the
possible values of a parameter versus the number of times each value occurred in
the sample or population. In many instances it refers to the spread of the
individual values of a sample or population around the mean.
- binomial distribution : Based on the idea that if only one of two possible
outcomes can occur on any one occasion, then the theoretical distribution of the
different combinations of outcomes that could occur can be worked out if the
number of occasions is known. One characteristic of this distribution is that it
consists of a limited or finite number of events, n. When n becomes very large,
tending to infinity, the binomial distribution becomes the normal
- chi-square distribution : This distribution may be considered as that of the
sum of squares of v independent normal variates in standard form. The parameter
v is known as the number of degrees of freedom .
- F-distribution : Theoretical distribution used to study population variance.
It is the distribution of the ratio of two independent variables each of which
has been divided by its degrees of freedom .
- normal distribution : Purely theoretical continuous probability distribution
in which the horizontal axis represents all possible values of a variable and
the vertical axis represents the possible of those values occurring. The scores
on the variable are clustered around the mean in a symmetrical, unimodal pattern
known as the bell-shaped (normal) curve. In a normal distribution, the mean,
median and mode are all the same . The normal distribution is obtained when
the number of events in the binomial distribution n, becomes very large, tending
probability distribution : Distribution giving the probability
of a value of x as a function of x or, more generally, the probability of joint
occurrence of a set of variates x1,….,xp as a function of those quantities
- t-distribution : Theoretical probability distribution used in hypothesis
testing. Like the normal distribution, the t-distribution is unimodal,
symmetrical and bell-shaped .
- theoretical probability distribution : Number of times it can be expected to
get a particular number of successes in a large number of trials . Important
theoretical probability distributions are the normal, t-, chi-square and
- z-distribution : Normal distribution in which the scores are the
- distribution function : The distribution function F(x) of a variate x is the
total frequency of members with variate values less than or equal to x. As a
general rule, the total frequency is taken to be unity, in which case the
distribution function is the proportion of members bearing values less than or
equal to x .
- dixon test : See outlier.
- double blind procedure : Means of reducing bias in an experiment. In the
clinical context, for example, such a procedure ensures that both those who
administer a treatment and those who receive it do not know (are blind to) which
subjects are in the control group and which are in the experimental group, that
is, who is and is not receiving the treatment (adapted from ).
- duplicate samples or specimens : Two aliquots of a sample or specimen
analysed at the same time.
- dynamic range : Range over which a relationship exists between analyte
concentration and assay response .
- Error : Something done that is considered to be incorrect or wrong .
- absolute error : Difference between the analytical result and the true
- clerical error : Mistake made during routine jobs in an office or
laboratory, e.g. a transcription error, a specimen misidentification or a fling
- maximum tolerable error : Extreme values of an error permitted by
specifications, regulations etc. for a given determination. .
- random error : Component of the total error of a measurement that varies in
an unpredictable way. This causes the individual results to fall on both sides
of the average value , .
- relative error : Absolute error of a measurement divided by the assigned
value of the analyte . See coefficient of variation and relative ,standard
- systematic error : Component of the total error of a measurement that varies
in a constant way. This causes all the results to be in error in the same sense
- total error : Sum of random and systematic errors.
- type I error : Error made by wrongly rejecting a true null hypothesis.
If the null hypothesis is that the sample should be negative, a type I error
will generate a false positive result.
- type ll error : Error made by wrongly accepting a false null hypothesis
. If the null hypothesis is that the sample should be negative, a type ll
error will generate a false negative result.
- estimate value : See value.
- evaluation : Systematic examination of the extent to which a product,
process or service fulfiles specified requirements .
- expert witness : Knowledgeable person, for example, a forensic scientist,
familiar with the testing and the interpretation of test results and able to
give an expert opinion based on scientific fact or evidence, e.g. in court or at
a hearing (adapted from ).
- expiration date : Date after which the specified characteristics of a
reagent, solution, specimen, control etc. can no longer be
- false negative : Test result that states that no drug or metabolite is
present when, in fact, such a drug or metabolite is present in an amount greater
than a threshold or designated cut-off concentration (adapted from ).
- false positive : Test result that states that a drug or metabolite is
present when, in fact, it is not present or is present in an amount less than a
threshold or designated cut-off concentration (adapted from ).
- F-distribution : See distribution.
- F-test : See significance test.
- geometric mean : See mean.
- good laboratory practices (GLP) : Organizational process and conditions
under which laboratory studies are planned, performed, monitored, recorded and
reported. Includes a system of protocols (standard operating procedures)
recommended to be followed so as to avoid the production of unreliable and
erroneous data (adapted from  and ).
- goodness-of-fit : How well a model, a theoretical distribution or an
equation matches actual data .
- Grubbs test : See outlier.
- harmonization : Bringing about agreement on terminology, concepts etc. so
that different entities can interact based on the same terms of reference.
- hypothesis test : See significance test.
- hypothesis testing or significance testing : Process of assessing the
statistical significance of a finding. It involves comparing empirically
observed sample findings with theoretically expected findings, expected if the
null hypothesis is true (see significance test). This comparison allows one to
compute the probability that the observed outcomes could have been due to chance
alone . See non-parametric test.
- alternative hypothesis : Hypothesis that must be accepted if the null
hypothesis is rejected .
- null hypothesis (H0 ) : Any hypothesis to be tested. The term null implies
that there is no difference between the observed and known values other than
that which can be attributed to random variation (adapted from [3) and ].
- Imprecision : See precision.
- independent test result : result obtained in a manner not influenced by any
previous results on the same or similar material .
- influence quantity : Quantity, e.g. an environmental condition, that is not
the subject of measurement but that influences the result (adapted from
- in-house reference material : See screening test.
- initial test : See screening test.
- instrument linearity : Straight-line relationship between concentrations of
analyte and instrument response, in which a change in concentration causes a
proportional change in response .
- interfering substance : Substance other than the analyte that gives a
similar analytical response or alters the analytical result .
- interlaboratory test comparisons : See collaborative studies.
- internal standard : Addition of a fixed amount of a known substance that is
not already present as a constituent of the specimen or sample in order to
identify or quantify other components . The physico-chemical characteristics
of the internal standard should be as close as possible to those of the
- Interpretation : Explanation of what analytical results mean based on
chemical, pharmacological, toxicological and statistical principles.
- intralaboratory test comparisons : Organization, performance and evaluation
of tests on the same or similar items or materials within the same or laboratory
in accordance with predetermined conditions .
- laboratory : Facilities where analyses are performed by qualified personnel
using adequate equipment.
- laboratory Information Management System : See LIMS.
- least-squares : Statistical method of determining a regression equation,
that is, the equation that best represents the relationship among the variables
- level of significance : Probability that a result would be produced by
chance alone, i.e. the probability of incorrectly rejecting the null hypothesis.
It is, therefore, the probability of making a type 1 error.
- limit: Prescribed or specified maximum or minimum amount, quantity or
- action limit : Corresponds to a ±3 standard deviation front the mean. If an
observed value falls outside the action limit, the cause must be identified
immediately and remedial action taken.
- confidence limit : The limits of the confidence interval.
- control limit : The limits, on a control chart, that are used as criteria
for action or for judging whether a set of data does or does not indicate lack
of statistical control.
- detection limit : Smallest measured content from which it is possible to
deduce the presence of the analyte with reasonable statistical certainty ,
- quantitation limit : The smallest measured content from which it is possible
to quantitate the analyte with an acceptable level of accuracy and
- warning limit : Corresponds to a ±2 standard deviation from the mean. Even
if the method is under statistical control, approximately 5 per cent of results
may be expected to fall outside the warning limits.
- LIMS (Laboratory Information Management System) : Software package for
collating, calculating, controlling and disseminating analytical data. It can
perform a variety of functions, from specimen or sample registration and
tracking to processing captured data, quality control, financial control and
report generation .
- linear regression : Method of describing the relationship between two or
more variables by calculating a best-fitting straight line or graph .
- logbook : Book that records laboratory activities, e.g. instrumentation,
maintenance of instrumentation, sample preparation and reagents.
- maintenance : Activity of keeping something such as facilities, machines or
instrumentation in good condition by regularly checking it and doing necessary
- matrix : Material that contains the analyte, e.g. urine or blood.
- maximum tolerable error : see error.
- mean : When not otherwise specified, refers to arithmetic mean .
- geometric mean : The nth root of the product of n individual values.
- median : Middle value of a ranked set of data .
- metabolite : Compound produced in the body as a result of biochemical
- method : Detailed (defined) procedure for performing an analysis. See test
- method traceability : Property of a method whose measurements give results
that can be related, with a given uncertainty, to a particular reference,
usually a national or International standard, through an unbroken chain of
comparisons (adapted from).
- mode : In statistics, the value or values occurring most frequently in a set
- negative : Indicates that the analyte is absent or below a designated
cut-off concentration. "Not detected" Is sometimes used as a synonym for
negative, although this Is not recommended.
- negative control : See control.
- none detected : Indicates the absence of an analyte within the
specifications of the test(s) performed.
- non-parametric test : Statistical method that makes no assumptions about the
distribution of the population from which the sample data are taken (adapted
- one-tail test : Hypothesis test stated so that the chances of making a type
1 error are located entirely in one tail of a probability distribution, e.g. it
is applicable if we wish to test only whether method A is more precise than
method B and not whether method B is more precise than method A (adapted from
- two-tail test : Statistical test in which the critical region (the region of
rejection of the null hypothesis) is divided into two areas at the tails of the
sampling distribution, e.g. it is applicable if we wish to test whether methods
A and B differ in their precision (adapted from ).
- normal distribution : See distribution.
- not detected : The use of this term as a synonym for negative is not
- null hypothesis : See hypothesis testing.
- one-point calibration : Simplified calibration procedure using a single
calibrator and a blank.
- one tail test : See non-parametric test.
- organization : Company, corporation or institute (or part thereof, e.g. a
laboratory), private or public, that has its own functions and administration.
Some of the international organizations dealing with quality assurance are the
International Association of Forensic Toxicologists (TIAFT), the International
Federation of Clinical Chemistry (IFCC), the International Olympic Committee
(lOC), the International Organization for Standardization (ISO), the
International Programme on Chemical Safety (IPCS), the International Union of
Pure and Applied Chemistry (lUPAC), and the Organization for Economic
Co-operation and Development (OECD).
- outlier : Result that appears to differ unreasonably from the population of
the other results. Tests for outliers include the following:
- Cochran test : Compares the largest of a set of variances with the other
variances in the set .
- Dixon test : Compares the difference between a measurement and the one
nearest to it in size with the difference between the highest and lowest
measurements in the set .
- Grubbs test : Now recommended to replace the Dixon test or to be used
sequentially after the Dixon test . The single Grubbs test statistic is
calculated as the percentage decrease in the standard deviation of a set of
results following the removal of either the highest or lowest value in the set,
whichever gives the largest decrease in the standard deviation. The pair Grubbs
test statistic is calculated in an analogous manner by removing the two highest,
two lowest or else both the highest and the lowest values in the original set of
results, whichever gives the lowest standard deviation. The presence of an
outlier or a Grubbs outlier pair is indicated if the Grubbs statistic exceeds a
critical value, which depends on the number of results in the set and which is
given by a reference table , .
- parametric test : Statistical techniques designed for use when data have
certain characteristics, usually when they approximate a normal distribution and
are measurable .
- personnel : Persons qualified, by virtue of training and experience to carry
out their assigned functions (adapted from ).
- pharmacology : Study of the interactions of drugs with living systems
- plot : Representation of data on or by a graph.
- population or universe : (Theoretical) entity defined as an entire group of
people, things or events that have at least one trait in common .
- population statistics : Statistical descriptors of the population, e.g.
mean, median, mode or standard deviation.
- Positive : Indicates that the analyte is present at a level above a
designated cut-off concentration.
- positive Control : See control.
- power of test : ,Probability of rejecting the null hypothesis when it is
- Precision : Closeness of agreement between independent test results obtained
under prescribed conditions . It is generally dependent on analyte
concentration, and this dependence should be determined and documented , The
measure of precision is usually expressed in terms of imprecision and computed
as a standard deviation of the test results. Higher imprecision is reflected by
a larger standard deviation. Independent test results refer to results obtained
in a manner not influenced by any previous results on the same or similar
material . Precision covers repeatability and reproducibility .
- presumptive : Describes things that are based on presumptions about what is
probably true rather than on certainty .
- presumptive positive : Specimen or sample that has been flagged as positive
by screening but that has not yet been confirmed by an adequately sensitive
alternate chemical method .
- presumptive negative : Specimen or sample that has been flagged as negative
by screening. Usually no further tests are carried out, so there is no certainty
about its content.
- presumptive test : See screening test.
- preventive action : Action taken to eliminate the causes of a potential
deviation or other undesirable situation in order to prevent occurrence
- Probability : Mathematical measurement of how likely it is that something
will happen, expressed as a fraction or percentage . Values for statistical
probability range from 1 or 100 per cent (always) to 0 or 0 per cent (never)
. The relative frequency obtained after a long run of measurements or
results will give good approximations to the true probability . It is also
understood in other ways: as expressing in some undefinable way a "degree of
belief", or as the limiting frequency of an occurrence in an infinite random
series (adapted from ).
- probability distribution : See distribution.
- probability function : Function of a discrete variate that gives the
probability that a specified value will occur.
- procedure : Specified way to perform an activity. For quality assurance
purposes, procedures should be written .
- test procedure : Total operation necessary to perform the analysis, e.g. the
prearation of the specimen or sample, of the reference materials, or of the
reagents, the use of instruments and of formulas for the calculations (when the
test is quantitative), the preparation and use of calibration curves and the
determination of the number of replicates.
- processed data : Raw data that have been acted upon to make them clearer or
more readily usable.
- proficiency testing : Ongoing process in which a series of proficiency
specimens or samples, the characteristics of which are not known to the
participants, are sent to laboratories on a regular basis. Each laboratory is
tested for its accuracy in identifying the presence (or concentration) of the
drug using its usual procedures. An accreditation body may specify participation
in a particular proficiency testing scheme as a requirement of accreditation.
- qualitative test : Test that determines the presence or absence of specific
drugs or metabolite in the specimen or sample .
- quality assessment : Overall system of activities whose purpose is to
provide assurance that the overall quality control job is being done
effectively. It involves a continuing evaluation of the products produced and of
the performance of the production system .
- quality assurance (QA) : System of activities whose purpose is to provide to
the producer or user of a product or a service the assurance that it meets
defined standards of quality with a stated level of confidence .
- quality assurance management : All activities of the overall management
function that determine and implement quality policy, objectives and
responsibilities (adapted from ).
- quality assurance programme : Internal control system designed to ascertain
that the studies are in compliance with the principles of good laboratory
quality audit : Systematic and independent examination to determine whether
quality activities and related results comply with planned arrangements and
whether these arrangements are implemented effectively and are suitable to
achieve the objectives .
quality control : Overall system of activities whose purpose is to control
the quality of a product or service so that it meets the needs of users. The aim
is to provide quality that is satisfactory, adequate, dependable and economic
- external quality control : See proficiency testing.
- internal quality control : Set of procedures undertaken by a laboratory for
continuous monitoring of operations and results in order to decide whether the
results are reliable enough to be released. Quality control of analytical data
primarily monitors the batchwise trueness of results on quality control
specimens or samples and precision on independent replicate analysis of test
- quality management : That aspect of the overall management function that
determines and implements the quality policy .
- quality manual : Document stating the general quality policies, procedures
and practices of an organization .
- quality policy : Statement by top management regarding the laboratory's
adherence to principles of quality. It may set forth codes of practice or
- quality system : The organizational structure, responsibilities, procedures,
and resources for implementing quality management . In a
laboratory it refers to the total features and activities of a laboratory aimed
at producing accurate work and a high-quality product .
- quantitation limit : See limit.
- quantitative test : Test to determine the quantity of drug or metabolite
present in a specimen .
- random error : See error.
- random sample : Sample taken in such a way that all the members of the
population have an equal chance of being included, that is, each is chosen
entirely by chance .
- range : Concentration interval for which acceptable accuracy and precision
can be achieved (adapted from ). Statistically it is the difference between
the minimum and the maximum values of a set of measurements .
- raw data : Data that are in their original state and have not been processed
- record : Document that furnishes objective evidence of activities performed
or results achieved .
- recovery : Percentage of the drug, metabolite or internal standard
originally in the specimen or sample that reaches the end of the procedure 
- reference material : Material or substance one or more properties of which
are sufficiently well established to be used for calibrating an apparatus,
assessing a measurement method or assigning values to materials .
- in-house reference material: Material whose composition has been established
by the user laboratory by several means, by a reference method or in
collaboration with other laboratories .
- reference method or standard consensus method : Method developed by
organizations or groups that use collaborative studies or similar approaches to
validate it. Its value depends on the authority of the organizations that
sponsor it (adapted from ).
- reference preparation : Processed reference material.
- reference standard : A standard, generally of the highest quality available
at a given location, from which measurements made at that location are derived
- regression analysis : Method of explaining or predicting the variability of
a dependent variable using information about one or more independent variables.
Also, techniques for establishing regression equations .
- regression curve : Curve that comes closest to approximating a distribution
of points in a scatter diagram.
- relative error : See error.
- relative frequency : Number calculated by dividing the number of values with
a certain characteristic by the total number of values . Also, the frequency
of an event that would occur in the long run given the probability of the event
- relative standard deviation : See coefficient of variation.
- reliability : Extent to which an experiment, test or measuring procedure
yields accurate results in repeated trials .
- repeatability : Closeness of the agreement between the results of successive
measurements of the same analyte made under repeatable conditions, e.g. same
method, same material, same operator and same laboratory and carried out in a
narrow time period (adapted from , ). Results should be expressed in
terms of the repeatability standard deviation, the repeatability coefficient of
variation or the confidence interval of the mean value.
- replicate analysis : Multiple analysis of separate portions of a test
material using the same test method under the same conditions, e.g. same
operator, same apparatus, same laboratory ,
- report : Document containing a formal statement of results of tests carried
out by a laboratory. It should include the information necessary for the
interpretation of the test results .
- representative sample : Statistically, a sample that is similar to the
population from which it was drawn. When a sample is representative, it can be
used to make inferences about the population. The most effective way to get a
representative sample is to use random methods to draw it , Analytically, it
is a specimen or sample that is a portion of the original material selected in
such a way that it is possible to relate the analytical results obtained from it
to the properties of the original material .
- reproducibility : Closeness of the agreement between the results of
successive measurements of the same analyte in identical material made by the
same method under different conditions, e.g. different operators and different
laboratories and considerably separated in time (adapted from , ).
Results should be expressed in terms of the reproducibility standard deviation,
the reproducibility coefficient of variation or the confidence interval of the
resolution : Ability to distinguish meaningfully between closely
adjacent values .
- result : Information obtained from a test or series of tests. Usually it
refers to processed data.
- retention sample or specimen : Amount of material equivalent in quantity to
the assay specimen or sample and taken from the consignment in a manner similar
to that used to assay the sample or specimen. It should be stored under
specified conditions .
- review : Evaluation of laboratory results to ensure that they have been
- robustness or ruggedness : Capacity of a test to remain unaffected by small
variations in the procedures. It is measured by deliberately introducing small
changes to the method and examining the consequences . See influence
- sample : Analytically equivalent to specimen, it is a representative portion
of the whole material to be tested. Statistically, it is a set of data obtained
from a population .
- sample statistics : Statistical descriptors of the sample, e.g. mean,
median, median, mode, standard deviation, range or size.
- Sampling : Analytically, the whole set of operations needed to obtain a
sample or specimen, including planning, collecting, recording, labelling,
sealing, shipping etc. Statistically it is the process of determining properties
of the whole population by collecting and analysing data from a representative
segment of it .
- scatter diagram or scatter plot or scattergram : Pattern of points that
results from plotting two variables on a graph. Each point or dot represents one
subject or unit of analysis and is formed by the intersection of the values of
the two variables .
- screening test or initial test or presumptive test : First test carried out
on a specimen or sample for the purpose of determining a presumption of a
positive or negative assay. Usually, presumptive positives are followed by a
- selectivity : Extent to which a method can determine particular analyte(5)
in a complex mixture without interference from the other components in the
mixture  A method that is perfectly selective for an analyte or group of
analytes is said to be specific.
- sensitivity : Difference in analyte concentration corresponding to the
smallest detectable difference in the response of the method. It is represented
by the slope of the calibration curve. Sometimes it is used, erroneously, to
mean detection limit.
- shewhart chart : See control chart.
- significance test or hypothesis test : Statistical test whose purpose is to
draw a conclusion about a population using data from a sample. It is used to
determine the likelihood that observed characteristics of samples have occurred
by chance alone in the population from which the samples were selected,
- significance tests include the following :
- comparison-of-means test or t-test : Compares the mean of the results from
one sample taken from a given population with the mean of the results from a
second sample taken from the same population, with the two sets of results
having, for example, been produced by different analytical methods. It answers
the question, "Are the two means significantly different?" The null hypothesis
is that the means are not significantly different and that the samples are
therefore part of the same population. It is assumed that the variances of the
samples are the same and that the samples are representative of the whole
population. The larger the number of results for each sample, the more likely
this is to be true. Statistical comparison of the means will indicate whether
any differences between the samples could have arisen by chance alone. The
t-fest is used under particular circumstances, for example, when the size of the
samples is small (usually less than 20) or when a single sample is taken from a
population for which the variance is unknown.
- variance-ratio test or F-test : In the comparison-of means test it is
assumed that the variance of each sample is the same. A variance-ratio test is
used to check if this assumption is reasonable.
- significant figures : Number of f4gures that are consistent with the
precision of the test.
- skewness : Said of measures or scores that are bunched on one side of a
central tendency parameter (mean, median, mode) and trail out on the other. The
more skewness in a distribution, the more variability in the scores . Also
used to refer to asymmetry in, for example, a chromatographic peak shape
("tailing" and "fronting").
- solution : Liquid in which a solid substance or a gas has been dissolved
- specification : Statement of requirements, usually in written form.
- specificity : See selectivity.
- specimen : Analytically, equivalent to sample. In the context of this
Glossary, any biological material for examination, study or analysis.
- spiked sample : A test material containing a known addition of analyte
(adapted from ).
- split-level model : Statistical model that splits the study sample according
to a predetermined assumption so that only a portion of the cases falls into the
category of interest, e.g. so that only some of the specimens that were positive
for group A of drugs will be positive for group B.
- split specimen or sample : Practice of dividing a specimen or sample. A
urine specimen, for example, may be divided into two portions, one of which may
be submitted for analysis and the other preserved by freezing for confirmatory
analysis or reanalysis (adapted from ),
- stability : Resistance to decomposition or other chemical changes, or to
physical disintegration .
- standard addition : The addition of a known amount of a pure component
supposed to be present as a constituent of the specimen or sample in order to
verify and quantitate this component . Operationally, a measurement is made
on the specimen or sample, a known amount of the desired constituent is added,
the modified specimen or sample is remesured, and the amount of the constituent
originally present is determined by proportionation .
method : See reference method.
- standard deviation : A statistic that shows the spread or dispersion of
scores in a distribution of scores. It is calculated by taking the square root
of the variance , It is applicable to all kinds of repeated measurements,
e.g. between-batch, withinbatch, repeatability and reproducibility.
- standard operating procedures (SOP) : Written procedures that describe how
to perform certain laboratory activities .
- standard solution : Solution of known concentration prepared from
- statistical control : A procedure is in statistical control when results
consistently fall within established control limits, that is, when they have
constant mean and variance (adapted from ). Statistical control should be
monitored graphically with control charts.
- statistical correlation : Extent to which two or more things are related to
one another. This is usually expressed as a correlation coefficient .
- statistical significance : Said of a value or measure of a variable when it
is larger or smaller than would be expected by chance alone. Statistical
significance does not necessarily imply practical significance .
- stock solution : ·Concentrated standard solution used to prepare
- study : Experiment or set of experiments performed to obtain information on
a particular subject.
- surrogate analyte : See analyte.
- surveillance : Monitoring of certain activities to ensure that specified
requirements have been fulfilled (adapted from ).
- survey : Study conducted among organizations to collect information on their
activities or performance.
- systematic error : See error.
target analyte : See analyte.
target value : See value.
t-distribution : See distribution.
test : Technical operation to determine one or more
characteristics of or to evaluate the performance of a given product, material,
equipment, organism, physical phenomenon, process or service according to a
specified procedure (adapted from ).
test linearity : Ability within a given range to obtain test
results directly proportional to the concentration (amount) of analyte in the
specimen or sample.
test procedure : See procedure.
theoretical probability distribution : See
threshold : A particular, significant amount, level or limit,
at which something begins to happen or take effect. See cut-off
total error : See error.
traceability : Ability to trace the history, application or
location of an entity by means of recorded identification . See also chain
trueness : Closeness of agreement between the average value
obtained from a large series of test results and an accepted reference or true
true value : See value.
t-test : See significant test.
type I errors and type ll errors : See
uncertainty : A parameter, associated with the result of a
measurement, that characterizes the dispersion of the values that could
reasonably be attributed to the anatyfe.
universe : See population.
- validated method : Method whose performance characteristics meet the
specifications required by the intended use of the analytical results (adapted
from ). Some of the performance characteristics to be evaluated are limit of
detection, limit of quantitation, linearity, precision, range, ruggedness,
selectivity and specificity, and trueness.
- validation : Confirmation by examination and provision of objective evidence
that the particular requirements for a specific intended use are fulfilled
- value : The expression of a quantity in terms of a number and an appropriate
unit of measurement  .
- assigned value : Best available estimate of the true value .
- consensus value : Value produced by a group of experts or referee
laboratories using the best possible methods. It is an estimate of the true
- estimate value (statistical) : Value(s) of population characteristic(s)
obtained from sample data.
- target value : Numerical value of a measurement result that has been
designated as a goal for measurement quality .
- true value : Value that characterizes a quantity perfectly defined in the
conditions that exist when that quantity is considered. The true value of a
quantity is an ideal concept and, in general, cannot be known exactly
- variability : Spread or dispersion of scores in a group of scores·, the
tendency of each score to be unlike the others .
- variable : Generally, any quantity that varies. More precisely, a quantity
that may take any one of a specified set of value .
- variance : Statistic that shows the spread or dispersion of scores in a
distribution of scores. It is calculated as the sum of the squares of the
differences between the individual values of a set and the arithmetic mean of
the set, divided by one less than the number of values ,
Variance-ratio test : See significance test.
- variate : In contradistinction to a variable, a variate is a quantity that
may take any of the value of a specified set with a specified relative frequency
or probability. It is often known as a random variable .
- verification : Confirmation by examination and provision of objective
evidence that specified requirements have been fulfilled .
- warning limit : See limit.
- working standard solutions : Standard solutions prepared by diluting the
stick solution containing the concentrations used to establish the calibration
- z-distribution : See distribution.
- z-score : Number of standard derivation units that separate a value from its
mean (based on ).
1. United Nations Drug Control Programme, "Report of the Consultative Meeting
on Quality Assurance and Good Laboratory Practices", Glasgow, 2-6 November
2. National Institute for Drug Abuse, Urine Testing for Drugs of Abuse,
Research Monograph 13 (Rockville, Maryland, Department of Health and Human
3. J. C. Miller and J. N. Miller, Statistics for Analytical Chemistry
(Chichester and New York, Ellis Horwood, 1984).
4. R. Wennig,,Practical Compendium for Health Professionals : Drugs of Abuse
Currently Used in Europe, Publication CEC/V/E/I/Lux 92 (Luxembourg, Commission
of the European Communities, Health and Safety Directorate, 1992).
5. International Olympic Committee/Reference Materials Committee of the
International Organization for Standardization, "Quality control of analytical
data produced in chemical laboratories", Publication 271, draft protocol
presented to the Fifth International Symposium on the Harmonization of Internal
Quality Assurance Schemes for Analytical Laboratories, Washington, D.C.,23 July
6. PONS English Language Dictionary (London, Collins Cobuild.,1987).
7. F. M. Garfield, Quality Assurance Principles for Analytical Laboratories
(Arlington, Virginia, Association of Official Analytical Chemists,
8. American Society of Crime Laboratory Directors, Laboratory Accreditation
Manual (Norfolk, Virginia, 1992).
9. International Organization for Standardization/International
Electrotechnical Commission, Guide 2 : General Requirements for the Competence
of Calibration and Testing Laboratories (Geneva, 1990).
10. W. p. Voight, Dictionary of Statistics and Methodology: A Non-techniccal
Guide for the Social Sciences(Thousand Oaks, California, Sage Publications,
11. International Organization for Standardization/international
Electrotechnical Commission, Guide 2.: General Terms and Their Definitions
concerning standardization and Related Activities (Geneva, 1991).
12. M. G. Kendall and W. R. Buckland, A Dictionary of Statistical Terms
(London, Longman Group, 1976).
13, C. Visher, A C0mporiron of Urinalysis Technologies for Drug Testing in
Criminal Justice (Washington, D.C., National Institute of Justice and Bureau of
Justice Assistance, 1991).
14. Society of Forensic Toxicologists Inc./American Academy of Forensic
Sciences, Forensic Toxicology Laboratory Guidelines 1991.
15. K. Eckschlager, Errors, Measurement and Results in Chemical Analysis
(London, Van Nostrand Reinhold, 1969).
16. International Union of Pure and Applied Chemistry, Spectrochemica Acta,
17. International Organization for Standardization, International Vocabulary
of Basic and General Terms used in Metrology (Geneva, 1984).
18. National Measurement Accreditation Service, NIS 46, Accreditation for
Forensic Analysis and Examination (Teddington, Middlesex, 1992).
19. B.S.Finkle, R.V. Blanke and J. M. Walsh, eds., Technical, Scintific and
Procedural Issues of Employee Drug Testing (Rockville, Maryland, NIDA,
20. J. A. Timbrell, Introduction to Toxicology (London, Taylor and Francis,
21. Organization for Economic Co-operation and Development, The OECD
Principles of Good Laboratory Practice, Environmental Monograph No.45, Paris,
22. D. McCormick and A. Roach, Measurement, Statistics and Computation
(Chichester, Wiley, 1988).
23. Webster’s Third New International Dictionary of the English Language
(Springfield, Massachusetts, Merriam, 1971).
24. Food and Agriculture Organization of the United Nations/World Health
Organization, Codex Alimentarius Commission, Criteria to Limit the number of
False Positive and False Negative Negative Results for Analytes Near the Limit
of Determination, CX/MAS 92/15, Budapest, 1992,
25. EURACHEM/WELAC Chemistry Working Group, Accreditation for Chemcal
Laboratories : Guidance on the Interpretation of the EN45000 Series of Standards
and ISO/IEC Guide 25 (Teddington, Middlesex. EURACHEM/WELAC, 1993).
26. R. Caulcutt and R. Boddy, Statistics for Analytical Chemists (New York,
Chapman and Hall, 1983).
27. W Horwitz, "Protocol for the design, conduct and interpretation of
collaborative studies", PURE AND applied Chemistry vol. 60 (1988), pp.
28. p. C. Kelly, "Outlier detection in collaborative studies", Journal of the
Association of Official Analytical Chemists, vol. 73 (1990), pp. 58-64.
29. B. G. Katzungt and A. 1. Trover, Pharmacology (London, Prentice Hall
30. European Community, Guideline Criteria for Reference Methods, BNL
SP/Lab/div (92)5 (1992), p. 27.
31. International Organization for Standardization/Development Information
System 8402, Quality management and Quality Assurance Vocabulary (Geneva,
32. F. H. C. Harriet and M. G. Kendall, A Dictionary of Statistical Terms,
5th ed. (Mallow, Essex and New York, Longman Scientif'4c & Technical/wiley,
33. J. K. Taylor, quality Assurance of Chemical Measurements (Boca Raton,
Florida, Lewis, 1987).
34. L. Stebbing, Quality Assurance : The Route to Efficiency and
Competitiveness, 3rd ed. (Chichester and New York, Ellis Horwood, 1993).
35. A. C. Moffatt and M. D. Osselton, "Characterization of quantitative
methods for the analysis of drugs in blood in forensic toxicology", Bulletin of
the International Association of Forensic, vol. 20, No. 3 (1990), pp.
36. World Health Organization, Good laboratory Practices in Governmental Drug
Control Laboratories, WH0/PHARM/84.5121Rev.2,1984.
37. 1. M. Kolthoff, p. J, Riving and E. J, Meehan, eds., Treatise on
Analytical Chemistry, 2nd ed. (New York, John Wiley, 1978).
38. Commission of the European Communities, Directorate 111, Addendum,
Guidelines on the quality, safety and Efficacy of Medical Products for Human
Use, vol. 111, Publication CEC/11HEN/9104/90 (Brussels, 1990).
39. Analytical Methods Committee, "Prof'4ciency testing of analytical
laboratories: organization and statistical assessment", Analyst, vol. 117
(1992), pp. 97-117.
40. International Organization for Standardization/REMCO, "The international
harmonized protocol for the proficiency testing of (chemical) analytical
laboratories", Publication 263, Geneva, 1992.
41. W. Horowitz and R. Albert, "Experience with the lUPAC-1987 Harmonized
Protocol for Method-Performance Studies: suggestions for revision and
application to internal quality control systems", Quality Assurance for
Analytical Laboratories, M. Parkany, ed. (Cambridge, Royal Society of Chemistry,
자료 출처 : 식품의약품안전청