CHAPTER 6.

TOTAL QUALITY MANAGEMENNT 165

measured and monitored discomfort to the patient.

improved through QI through QC and QA,
and re-planned through 5 Analytical procedures performed without error and
QP.
These five components with "appropriate" analytical performance
working together in a feedback
loop illustrate how continuous including precision, accuracy, sensitivity, specificity,
how QA is built in a QI is accomplished and and freedom from interferences.
on P 180) laboratory (Refer 'five Q framwork 6 Clear presentation of results and reports, without
ambiguity with reference values, therapeutic
QUALITY CONTROL (QC) intervals and /or decision leveis available as needed.

QC is the of 7 Turnaround time, from submission of the test

setproredures undertaken in
for the continuous assessment of work carried out and
a laboratory request to delivery ofthe repoft, within clinically
evaluation of tests for reliable acceptable limits.
reports. The program
ensures the physicians, quality results 8 Readily available laboratory consultation when
and the patients,
reliable reports. QC emphasizes statistical control
needed; i.e. appropriate interpretation of laboratory
procedures and also inciudes reagent and results presented in a timely manner.
checks, linearity checks etc. QC is standard
used to detect and correct
a technique that is Note
errors before they result in a
defective product or
service QC must be Internal quality control is managed by minimizing-
achievable and affordable./ practical, preanalytical, analytical and postanalytical.errors
Quality control in
Pathology Laboratory can_be
a Control of Preanalytical Variables (Refer Chapter 7, P
conducted by two methods: 1)Internal Quality Control 220)
and 2) External Quality controB.
Monitoring preanalytical variables require the
Internal Q.C. and External Q.C. are coordinated effort of many individuals (and hospital
acttvities7 If Internal QC is necessarycomplementary
for the daily
departments, if the laboratory is in a hospital), each of
which must recognize the importance of these efforts
monitoring of the precision and accuracy of the analytical in maintaining a high.quality service. In order to monitor
method, External QC assessment is important for
and control theae errors, it is essential to understand
maintaining the long-term accuracy of the analytical
methods. the system as a series of following processes, each of
which has potential sources of error.

INTERNAL QUALITY CONTROoL 1 Test ordering and utilization: Sources of error may
Through internal quality control, the laboratory can be introduced due to inappropriate tests, bad
ensure that the results being issued by laboratory are handwriting (which may not be legible), wrong
reliable Internal quality control is the study of errors patient identification and due to delayed order of
and introduction_of procedures to récognize and testsand inappropriate control of 'stat' tests.With
minimize them The errors include all those arise within proper measures, sources of error arising due to
theTaboratorybetween the teceipt of the specimen and above-mentioned aspects can be controlled.
dispatch of the report./A laboratory that meets quality 2 Patient preparation: Laboratory tests are affected
requirements has fewer reruns and complaints, and this by many factors such as recent intake of food, al-
saves money Meeting quality requirements satisfies cohol, drug etc., food intake (3 days prior to test).
clinicians (users of the laboratory), giving the laboratory Smoking, exercise, stress, sleep and posture dur-
a greater competitive edge and providing job security ing specimen collection also can affect laboratory
for the senior staff. It also leads to better and more tests. Laboratory management must define the in-
efficient patient care. structions and procedures for patient preparation.
These procedures should be included in hospital
For internal quality control, it is necessary to understand
the following
procedure manuals and should be transmitted to
requirements of clinicians patients in both oral and written instructions and
1 An adequate test menu to meet clinical decision- Compliance with these instructions can be moni-
making needs. tored by the phlebotomist.
2 Information and education about the tests that are 3 Specimen collection, acquisition and patient iden-
appropriate for a particular clinical problem, tification: Various problems can arise due to in-
3 Correct tube (or container), incorrect order of
Instructions on preparing the patient prior to
specimen collection and on the appropriate vacutainers, incorrect anticoagulant, incorrect
specimen for an assay. patient identification, inadequate volume, hemo-
4 Phlebotomy performed safely and efficiently without yzed, lipemia or icteric sera, specimens collected
at wrong time and due to improper transport
166 TEXTBOOK OF MEDICAL LABORATORY
TECHNOLOGY
containers
conditions. Prolonged tourniquet in making plastic
application the plasticizers used materials
causes local anoxia to cells and
excessive venous interfere with drug
analysis. Some plastic such as
back pressure. The anoxia causes small solutes amounts of analyses
some
adsorb trace are not
(such as potassium) to leak from cells and the
parathyroid
hormone. Glass Containers
and trace
venous pressure concentrates
cells, proteins, and suitable for the determination
ofcalcium
substances bound to proteins (such as
Blood collected from an arm into which calcium). elementS.
an
venous infusion is running can be diluted or
intra
Variables
taminated. con Control of Analytical
to
variables, it is necessary
The greatest For the control of analytical
promise for correct identification of terms:
understand the following
patient specimens is the introduction of the laser between the
beam wand, which can read bar code or optical Trueness is closeness of
agreement
series of test
characters from labels. Various measures are obtained from a large
average value
suggested in chapter No. 7 to prevent hemolysis of reference value.
results and an accepted
sera and use of between a
specimen blanks or dichromatic of agreement
Accuracy is the degree Statistically it is
readings to tackte problems arising due to measured valyeand its true value^
hemolyzed, icteric or lipemia sera. which is defined as
represented by analytical bias,
between measured
the percentage of the difference
Note method gives
value and true value. An accurate
1 As much as 84 percent of laboratory errors can be correct result.
to the
attributed to the preanalytical phase of specimen Precision refers to reproducibility. It refers
measurements: It is
testing. agreement between replicate
Standard Deviation
2 A quality test result always starts with a quality quantitatively expressed as
or more precisely as Coefficient of
Variation
sample. (S.D.)
(C.V.) of the results in a set of replicate
3 High sensitivity assays that measure-smallamounts measurements. Hence good precision means least
such as TSH, troponin,hCG, troponin are more C.V. A precise method gives same results, if
Susceptible to errors
repeated.
4 Citical care staff such as phlebotomists, nurses A sensitive method measures low concentration
and collection centre technicians should be well of analytes. Sensitivity is a measure of the incidence
trained for coFrect specimen collection, separation of positive results in patients known to have a
of serum/plasma, preservation and transportation
condition, thatis True Positive (TP). A sensitivity of
ofspecimen. 90% implies that only90% of people known to have
5 inadequate clotting time leads to formation of the disease would be diagnosed as having it on the
micro-clots in the specimen and instrument probe basis of that test alone, 10% would be False
obstruction. Negative (FN).
Inadequate mixing of anticoagulated sample in the TP
tube also leads to formation' of microclots, latent Sensitivity = x 100
All with disease + FN
fibrin formation and platelet clumping. This may
lead to decrease in hematocrit, hemoglobin and A specific method is not subject to interference
also sample probe obstruction by other substances. Specificity is a measure of
theincidente of negative results in persons known
Inadequate spinning of sample tube leads to to be free of disease, that is
leakage of red blood cetts, which leads to *True Negative' (TN).
A specificity of 90% implies
contamination of serum with potassium, inorganic that
free people would be classified as
10% of disease-,
phosphorus, LDH and SGOT. on the basis of
having the diseasee
arrive in the
test result; 10% would have a 'False
8 Laboratory logs: When the specimens Positive' result.
laboratory, various logging and monitoring systems
TN
are necessary. Specificity =
All with 100
9 Specimen transport: The stability of specimen disease + TN
during transport from the patient to the laboratory
should be monitored carefully. Particularly for Note
specimen collected at collection centers and those Factors which increase the specificity of a test tend
sent to specialized laboratory should be transported to
decrease the sensitivity and vice versa, as there is
at 2-8°C. almost always an overlap between test results seen in
Collection tubes,
10 Careful specimen separation: health and in disease.
tubes are sources of
pipettes and aliquot determinations. Some of
contamination for many Whether to maximize specificity or
sensitivity depends
 CHAPIER6 TOTAL QUALITY MANAGEMENT 167

A Number
of B Number
results Method 1
of
Method 3 Method 4
results

Method 2
Mean Test result True Mean Test result
value value value
Method 1 is
precise compared to method 2 Accuracy
Methods 3 and 4 are equally precise but method
4 does not offertrue value
Fig. 6.1A & B: Precision and
accuracy
on the nature of the condition.
For example,
is very important in a sensitivity and issued by an organization are accepted as
screening test for
condition, since it is possible to a harmful

investigate further the technically competent.

false negative results.
for a trial of a new However, in selecting
patients Note
treatment, highly specific test is
more appropriate to ensure
a
that the treatment is Proper choice of calibrating materials establishes the
given only to patients who have a particular disease being
"traceability" of the analytical results to the "true value"
(Fig.6.1A & B). through the structure of the measurement system.
Reference and Control Materials
Analytical variables
These are highest quality control materials and are used
to validate reference methods and
In analytical measurements, problems can arise due
primary reference to various reasons such as
selection of inacCurate
materials.
methods, specimen mix-up, instruments with poor
precision, uncalibrated instruments, inçorrect votume
Primary reference material is the highest quality of specimens and presence of interfering substance in
material and should be used in the development the-speeimens. The analytical variables can be
and validation of reference methods. It is also used controlled by considering following various aspects of
to calibrate definitive and reference methods and Internal Quality Control:
production of secondary reference materials. 1 Selection of analytical methods: The ideal analytical
Reference methods should be used to validate field method should be, accurate, precise, sensitive
(routine) methods. and specific.
Reference Materials (RM): These materials are used 2 Validation of
for the calibration of an
apparatus or for the
glassware, reagents, equipment and
instruments: It is necessary to get specific glass-
verification of a method. One or two properties of ware, reagents, diagnostic kits, equipments and in-
these materials are well established for the purpose
of calibration.
struments validated (standardized) by Government
certified validation centre before starting the labo-
Secondary reference materials should be used to ratory practice. This is to ensure Quality Assurance
provide working calibrators for field methods, and and enhance Internal quality control.
also to assign values to control materials. Validation of methods: Validation of an analytical
Control material: It is a specimen, or solution, which method is the process by which it is established,
S analyzed only for quality control purposes and by the laboratory studies that the performance char-
not used for calibration purposes. acteristics of the method meet the requirements
for the intended analytical application. The perfor-
Control materials are used to monitor field (routine)
methods. mance characteristics of a method are
accuracy,
Calibration and Test Material (CTR): This is a
precision, specificity, sensitivity, linearity and rug-
gedness. It is essential to validate standard, run
reference material or solution with which the test
standard and new test and /or calibration methods
sample is compared.
to confirm that the methods are suitable for the
Certified Reference Material (CRM): This isa intended use. The technique used for validation of
material, accompanied by, or traceable to, a
certificate stating the property value(s) concerned à method should be one or a combination of the
following:
 168 TEXTBOOK OF MEDICAL LABORATORY TECHNOLOGY
instrumentation.
Calibration using references or reference labeling and errorsin
unpredictably. It
materials. haphazardly
and
errors occur specimen
or by
Comparison of results achieved with other Random patient's
wrong
methods. be using
byreconstituting reagent.
may
incorrectly

Inter-laboratory comparisons. include misreading

of
Systematic assessment of the factors influenc of random
errors
and
Other types
calculation
in
ing the result. chance errors
digits.
instruments, transposing of
Assessment of the uncertainty of the results including
transcription error,
based on scientific understanding of the and these are
errors
random
theoretical principles of the method and Tmprecision is caused by random variation
Sources of
practical experience. nherent in all analyses.
differently; the result may
measurement

The aim of the validation of test and calibration influence each should be and may vary
methods than it
be higher or lower
is to demonstrate that the method is fit for the little in magnitude.
intended Considerably or
purpose and that the results have an acceptable
reliability. could occur due to wrong
errors
Systematic analytical
and incorrect
procedure, incorrect standard
Sources of Variation in Analytical Phase of Laboratory
Tests standardization procedure.
Error is a difference between a single result estimate Note
of a quantity and its true value. This difference or de- 1 Determinate (systematic) error (Fig. 6.2A & B)
viation (positive or negative) may be expressed either
in the units in which the quantity is measured or as a can be traced to a specific cause. Common causes
of inaccuracy include constant or proportional
percentage of the true value. Many kinds of error may
errors. Constant systematic error is an error that
skew the findings of laboratory tests. These include
random error, random variation, constant and propor- is always in the same direction and of the same
tional determinate (systematic) error, and methodologi- magnitude even as the concentration of the analyte
cal or physiologic interference. changes. Proportional systematic error is an error
that always occurs in one direction.
Analytical phase can be affected by various types of A classic example of systematic error is a method
errors, imprecision and interference of certain related error caused by the
Components in the sample such as drugs. presence of interfering
substances in the specimen These interfering
substances may be endogenous metabolites, such
Analytical Errors: The errors are classified into random asbilirubin, chylomicrons, drugs-received by the
analytical errors and systematic analytical errors.
Random analytical errors indicate poor precision and patient or excess of anticoagulant
systematic errors indicate poor accuracy. (Refer to Drug interference may be methodologic (in vitro)
figures 6.10-6.20). or
physiologic (in vivo) in
interference occurs when origin. Methodology
the drug is actually
Random analytical errors include pipetting error, detected during measurement as if it were the
transcription error, wrong sample numbering and substance of interest. Physiologic interference,

A
40
40+
Constatemor-
301
30

Ne
o ro
Proportionalerror
20
20

10
10-

(0,0) 0 20 0 40
(0,0) 10 20
Reference values 30 40
Reference values
Fig. 6.2A & B: (A) Systematic errors and (B) Random errors.
 MANAGEMENT 169
CHAPTER 6: TOTAL OUALITY

actually represents biological variation when a drug9 average value. The formula
for finding the S.D. is,
received by the patient induces an increase in the
substance of interest being measured. (K-x)
3 Random and systematic errors can be S.D.
detected by n-1
using quality control charts such as "Levey-Jenning
and Cusum charts (Figs. = Sum of
6.10-6.22). x Averagevalue
X = Any single observed value
Postanalytic Errors
n = Total number of observed values
These are mainly transcriptional errors. To avoid these
errors in the results of the test, the reporting
or signatory Note
analyst should verify the results entered manually or control chart, the first
In order to prepare a quality
through on line instrument interfaces before the results the S.D. for the procedure
are reported or dispatched. thing to do so is to determine
it is necessary to
to be controlled. In a practical basis,
At the end of the month,
run a control serum each day.
FORMULATING QUALITY CONTROL CHARTS at least 20 values of various analytes
can be obtained
which can be used for the preparation
of Levey-Jenning
Following are the additional "Definitions" needed in
formulating Quality Control charts such as "Levey- chart for each analyte.
Jenning and Cusum charts:
EXPERIMENT NO. 6.1

Standard: This is a substance of constant composition Aim

of sufficient purity to be used for comparison purposes,
or standardization. (e.g. glucose, cholesterol, etc.). Preparation of Levey-Jenning chart

Control (Control serum): A sample with a known Requirements

value or range of values (e. g, concentrations of glucose, 1 Normal and abnormal control sera

creatinine, SGPT, SGOT, etc.), that is tested in the same 2 Glassware to perform tests
tested in order to ensure that
way patient samples are 3 Equipments (e. g. push button pipettes,
water bath,
This
methodology of a laboratory is working accurately. centrifuge, photometer or spectrophometer, etc.).
is a sample (control serum) that is chemically and 4 Diagnostic kits (for the determination of glucose,
physically similar to the unknown specimen. urea nitrogen, creatinine, etc.).
in the name of 5 Pencils, rulers and graph papers
Analyte: It is a component represented
a measurable quantity (i. e.
creatinine in serum, glucose
in plasma, or, urea in control, etc.). Procedure
1 Use a control sample for various tests such
serum
recorded values
Median: A middle value of a set of as glucose, urea, creatinine, uric acid, SGPT,
etc.
above as below (Refer
with an equal number of values every day.
to L-J chart, Fig. 6.4). 2 Note control values for various tests every day.
from the 3 At the end of the 3 weeks, at least 20 values of
Drift: The condition of values moving away various tests will be obtained.
mean line over a period of
time (Refer to L-J chart,
4 Prepare "Levey-Jenning chart for every test
Fig.6.6). to following "calculation procedure"
according
values suddenly move (Table 6.1).
Shift: The situation where control
from clustering around the mean
line to a line above or 5 Example- Test: Glucose. In the column "A" of
Table
below the median (Refer to L-J chart, Fig. 6.7).
listed.
6.1, 20 values of control glucose are
value
i. e. a b Calculate average control serum glucose
Bias: It is a measure of degree of trueness, (Equation 1: It is 100 mg/dl).
measure of systematic error (Fig. 6.14 to the "calculation
7 Prepare column "B" according
from procedure" in Table 6.1.
outliers: Results which appreciably different
are
the "calculation
the general cluster of values (Fig. 6.12). 8 Prepare column "C" according to
procedure" in Table 6.1.
of the dispersion of
randem 8 in Table
Imprecision: A measure 9 Prepare "Equation 2" according to step
errors. 6.1
10 Calculate "Standard Deviation" (1SD,
2SD, 3SD)
Standard deviation: This a statistical expression values using following formula:
around a central
of the scatter or dispersion of values
 172 TEXTBOOK OF MEDICAL LABORATORY TECHNOLOGY

Table 6.2: Determination of Cusum value Example

Difference Cusum In the case of Plasma Urea nitrogen determination, if
Date Mean Individual
value True value (average) of "Control"
=
35 mg/dl and
value test values
Reported value (average) by the laboratory = 32 mg/
+2 8.5 %.
12.2.89 100 98 dl, % Bias =(35 32) X 100/ 35
13.2.89 100 101
14.2.89 100 104 Note
100 97 +3 0
15.2.89
Systematic of urea nitrogen test is 8.5%. It is
error
100 106 6
16.2.89 significantly high and need immediate corrective
0 - 6 measures, i.e. checking qualities of reagents used,
17.2.89 100 100
pipetting and dispensing systems, various functions
IMPORTANCE OF DETERMINATION OF CV of automated system, primary standards and

Frequently it is necessary to compare the relative controls.

variability between various sets of data and methods.
This can be done by examining the mean and standard +35D
deviation of two groups for they may be of different +25D
magnitude or of different units. To make such 1SD 441
comparisons the standard deviations (S) are expressed Mean 4300
as percent of the means. This gives a measure of
-1SD 34100
relative yariability and called as Coefficient of Variation 2SD 32 100
(CV) 3SD 28300

S.D. 3R 240

CV 100 2P 160
mean (x) 1R

Example
1 Glucose determination: Method: Folin-Wu.
If = 98 mg/dl
S.D. = t 2.5

2.5 x 1000
C.
98
2.
2 Glucose determination: method: Glucose oxidase
If 78 mg %
=

S. D. = 0.9 mg %

0.9 x100
C.V.=
78
1.2%
Fig. 6.10: Display of quality control charts (L-J) on
autoanalyzer screen
Interpretation
Thus the variation inherent in the reducing sugar method IMPORTANCE OF REFERENCE RANGE
(Folin-Wu) is more than twice as great as that in the (NORMAL
glucose oxidase method. VALUES)
The college of American
standards for their Pathologists, in setting
IMPORTANCE OF DETERMINATION OF % BIAS
inspection and Accreditation Program
for Clinical Laboratories, indicates
that reference values
Determination of percent bias of a control test result (i.e., normal values, the term used in the
test should be provided with past) for each
gives idea of any systematic error present in a particular mind two modern
the report by keeping in
batch analysis. An estimate of bias is the average value laboratory realities:
of a series of measurement minus a reference value. 1 A reference range is required for the
of
Percent Bias gives idea of percent degree systematic
formula:
of most laboratory tests, and
interpretation
calculated using following
error. It is 2 It may not be
Reported value X 100/ True value
possible to provide an
appropriate
% Bias True value
=
- reference range in all cases.
 170 TEXTBOOK OF MEDICAL LABORATORY
TECHNOLOGY
t 6.0
Sum 2nd S.D.
=

of squared differences 9.0

(column c) from average 3rd S.D. =
t
Standard deviation =y Levey-Jenning
Chart and the
No. of values 1 Interpretation of the
Gaussian Curve
(Fig. 6.3)
11 Draw "Levey-Jenning chart by plotting Test days" distribution curve, or
Gaussian
o b t a i n e d on a
curve

on "X" axis and "Glucose values" on The values

"Y" axis. Draw 68% of all
within 1st S.D. from
indicate that about
a linerepresenting "average glucose value" would fall
of control s e r a 95% would be
(median) and corresponding standard
deviation
group
or average
value. About
would be
values (t 1SD, 2SD, 3SD) the mean
and 99,7%
charts in Figs. 6.4-6.7,
(Refer Jevey-Jenning expected to fall
within 2 S.D. The values when
6.12-6.22). S.D. vàlues./
expected to fall within 3 deviation range indicate
first standard
Note observed between methodology.
the specific
g00d control over
Plot "Levey-Jenning Charts" for other
tests such as
urea, creatinine, uric acid, SGOT, etc., this using
procedure (steps 1-12).
Table 6.1:
Calculating standard deviation 68.7
Calculation Procedure Number A B C
of tests
List values in Column A 5.4%
95 25
Add column A 100
Divide total of Column A by 101 +1
number of values.(see equation:1) 102 99
This is the average or mean 97
value
In column B lists the difference in 103 +3
values of column A from the 101 +1
Concentration
average value (100). Disregard 99 1S1S
plus or minus signs 9 98 2S 25
Multiply each difference in value 10 100 0 3S 3S
by itself (square each value 11 95
and place in column C) 12 01 +1
The normal distribution curve
Add values in column C 13 105 5 25
Divide total of column C by number 14 100 0 Fig. 6.3: The Gaussian curve.
of values minus 1 (see equation 2)| 98 4

9 Determine the square root of 8.37 16 101 +1

(a number that when multiplied by 97 Various examples of L- charts (Figs. 6.4-6.8)
itself will equal 8.37, see equation 3) 18 106 +6 |36
|10 2.89 x 2.89 8.35 19 100 Note
11 Standard deviation 2.89 20 102+2 1 It is necessary to take prompt remedial action in
Rounded off to nearest significant
the laboratory, once Levey-Jenning chart indicates
number standard deviation 2001 159 any specific error. This action helps to correct the
results of a specific batch-analysis before the
2001 reports of the patients are dispatched.
Equation: 1: = 100.05
20 2 Levey-Jenning charts and Cusum charts
correct postanalytical errors
help to
159
Equation: 2: 8.37
20 1 Median
Levey-Jenning chart
Sum of squared differences +3 SD
+ 2SD
(column c) from average
Standard deviation =V 1SD
No. of values 1
-1SD
= 8.37 2SD
-3 SD
S.D. = 2.89

3.0 LuuuuLLLILLYLLLLLLLLLLLILLLLLLuLLL+ Days

5 10 15 20 25
Nearest significant number 30 35 40 45 50
1st S.D. = t 3.0 Fig. 6.4: L-J
chart
 CHAPTER6 TOTAL OUALITY MANAGEMENT 171

Bias in control values aftor Shift in control values

day 9
Random error on day 3
vey Jeinitg Chnrt
120
Determination of plasma glucose
120
100
110

100

90 1 5 6 7&9 10 11 12 13 14
80 Days
1 2 3 45 6 7 8 9 10 11 12 13 14 15 This type of shift is observed in the case
Fig. 6.8: Interpretation:
Days of reagent or standard deterioration of concentration at one point of
Mg/dl time and later on there was no rurtner deterioration in reagent or
Note: Average value of QC serum
glucose =
100 mg/d standard.
Standard deviation (S. D.) =+10
mg/d
Fig. 6.5: Interpretation: Random error is seen on
mistake committed by the laboratory technician. Afterday
4, due to
day 9 higher
detect systematic errors. When a systematic error
readings of glucose are observed. Jhis may be due to use of is present, the
decreased concentration (all low values or all high values) cusum
of
primary standard (may be due to values will steadily increase. At a certain state the error
deterioration) will be too large to be accepted and method will be
considered out of controlCusum technique generally
Excessive scatter provides better determination of systematic errors than
can be obtained by Levey-Jenning charts. A Cusum chart
can be prepared as follows (Refer to the glucose
120
determination values used for the Levey-Jenning chart
Table 6.1).
100

Specimen analyzed: Quality control serum.

80
1 2 3 4 5 6 7 8 9 10 11 12 Determination: Glucose by glucose oxidase method.
Mean glucose value: 100 mg/dl
Days

Fig. 6.6: Interpretation: Excessive scatter may be observed due to Calculation of Cusum
inconsistent ways of working of laboratory technician.
Refer to Table 6.2. The Cusum is calculated by adding
the values in column 4 (Difference), i.e.Cusum for first
Drift or trend in the control values day is (average value:100 individual test value, 98 =
2), Cusum for day 13.2.89 = +2 -1 = +1. Cusum for
day 15.2.89 = sum of +2, -1, -4, +3 = 0. The Cusum
120 for the next day can be calculated by adding the new
difference to the previous Cusum. The Cusum chart is
100 drawn by plotting days on X-axis and Cusum values on
Y-axis (refer to the graphic presentation).

1 2 3 4 5 67 89 10 11 12 13 14
+10
Fig. 6.7: Interpretation: In type of drift or trend in the control
values may be seen either due to deterioration of reagent or due to
increased concentration of primary
standard.
0.0
5 7 8 9 10
Cusum Charts (Fig. 6.9) Days
In order to check the accuracy of the analysis run and Drift or trend after the day 5
also the quality control sera, a Cusum chart is plotted
from the quality control sera results. Such chart can Fig. 6.9: Cusum chart.
 MANAGEMENT 173
CHAPTER 6: TOTAL QUALITY

Following conditions are mandatory to make comparison

of a patient's laboratory result with reference values No of
possible and valid: 60001 results
1 All groups of reference individuals should be
clearly
defined.
2 The patient examined should
sufficiently resemble
the reference individuals. 4000
3 The conditions under which the
samples were
obtained and processed for analysis should be
known
4 The specimens compared should be of same type. 2000
The International Federation of Clinical Chemistry (IFCC)
recommends the following terms:
A Reference individual: A healthy individual selected
for comparison using defined criteria.
50 60 70 80 90 100 110
A Reference value: A value obtained by observation Glucose Mg/d
or measurement of a particular type of quantity on a 6.11: Determination of reference range.
Fig.
reference individual.
of a particular analyte (of various reference individuals):
An Observed value: It is defined as a value of
particular type of quantity, obtained by observation or Reference range = + 2 S.D.
measurement and produced to make a medical decision = Mean value
after comparing with a reference value. S.D. = Standard deviation.

Note This is based on inter-percentile interval and defined

as an interval bounded by two percentiles of the
1 Direct selection of reference individual is the only
method that agrees with the reference values concept
reference distribution.

as recommended by IFCC. The disadvantages

of this
Note
method are the problems and costs of obtaining a
1 It is necessary to define partition criteria for the
representative group of reference individuals.
sub-classification of the set of selected reference
2 The indirect method is based on the observation
in individuals into more homogeneous groups. Age
that the majority of analysis results produced and sex are the most frequently used criteria for
"Normal'. The
the clinical laboratory seems to be
that the peak subgroups.
assumption of the indirect method is
and shape are not too far from that of a
Gaussion 2 Age may be categorized by equal intervals and by
distribution and the peak is composed mainly
of intervals that are narrower in the periods of life,
"Normal values" (Fig. 6.11). where greater variation is observed. It is convenient
to use qualitative age groups such as postnatal,
examples of exclusion criteria
for health infancy, childhood, prepubertal, puberty, adult,
Following are
premenopausal, menopausal or geriatric.
- associated reference values:

3 Height and weight could be used as criteria for

Disease categorizing children.
and risks
R i s k factors such as obesity, hypertension, Use following procedure for the determination
from environment and occupation and genetically
of reference range of a particular test:
determined
a) Example: Test: Fasting plasma glucose
Intake of pharmacologically active agents such as determination by glucose oxidase method.
oral contraceptives, alcohol, tobacco and drugs
(treatment for disease or suffering) b) Selection of normal adults: Age: 18-55
Specific physiological states, such as pregnancy, years, Height: 1.50-1.80 meters.
stress and excessive exercise. c)Collect and note data of fasting blood
glucose values of 100 normal adults
Determination of Reference Range d) Calculate average blood glucose value and
SD by using formulae given in the prep-
Following formula is used to obtain reference range of
aration of L-J chart.
a particular test, after determining the various values
 MANAGEMENT 179
QUALITY
CHAPTER 6: TOTAL

VARIOUS WAYS OF MAINTAINING INTERNAL 1 Preventive phase

QUALITY CONTROL
2 Retrospective phase
Following are the various ways of maintaining Internal preventive
quality control in a laboratory: 1 Preventive phase: In this phasevariouS
stages
precautions are taken at the following
1 Employment of conscientious and well trained staff 7).
of specimen analysis (Refer Chapter
2 Use of standardized glassware, and reagents, equip- a) Collection of specimen
ment. 30
(preferably within
3 Maintenance of- b) Separation of s e r u m
minutes of blood collection)
a) Proper analytical skill throughout the test.
c) Specimen analysis
b) Required quality of reagents.
d) Photometric readings and
c) Desired performance of the instruments.
e) Calculations of test values etc.
4 Selection of accurate and precise methods. includes the
2 Retrospective phase: This phase
5 conditions variance
Routine use of various primary standards, quality Comparison between optimum
control sera, reference sera, previously analyzed conditions variance (RCV).
(OCv) and routine
specimen, and statistical methods to evailuate the results obtained under
OCV refers to the test
obtained results. conditions i.e. by using (i) Freshly
optimum standardized
prepared reagents and (ii) By using
Quality control in specimen analysis is observed as
A' grade glassware.
follows: routine
RCV refers to the results obtained by using
1 At least one primary standard is included with each i.e. using routinely stored reagents
requirements, by
batch of unknown specimen analysis. and glassware in regular use

2 Occasionally a different primary standard of high

er concentration is included to check the reliability Note
of routine primary standard. not be
The difference between OCV and RCV should
3 At least one normal or abnormal quality control more than 3%.
sample is included with each batch of unknown
specimen analysis. The analytical procedure variance may be observed
values of due to the following reasons.
4 The batch results are accepted if the
control sera are within first standard deviation. 1 Deterioration of reagents.
The random and systematic errors can
be detected Inadequate mixing of reagent and sample.
5 2
Levey-Jenning and Cusum charts.
by 3 Variation in temperaturee control.
4 Exposure of light sensitive colored end products to
Different types of errors observed during QC program
are:
light.
1 Intrinsic :a) Due to an imprecise and inaccurate To control analytical variance appropriate precautions
method are taken. The other factors, which also may be
blank values (in terms
b) Due to high responsible for the analytical procedure variance are:
of O.D.)
1 Environmental factors (unfavourable)
2 Systematic a) All high values
2 Qualification and training of staff (not complying
b) All low values and
3 Random with requirements)
:Technicalerrors
3 Work load (excessive)
1 Intrinsicerrors can be eliminated by (a) selecting 4 Staff management (mismanaged) and
and accurate methods and (b) by using fresh
precise 5 Disorganized documents of laboratory proceduress
batches of the reagents.
errors can be corrected by using
different
2 Systematic Other Aspects of Internal Quality Control
concentrations of primary standards and also by
a control serum. Following aspects are also important in maintaining good
analyzing
Internal quality control-
The technician with a proper analytical application
can correct random errors. DDocumentation of analytical protocols
While performing analytical determinations, it is
Quality Control
Two Phases in Internal necessary to follow a procedure step by step. Over
are two phases in internal QC a long period of time, to obtain reliable results,
Following
 180 TEXTBOOK OF MEDICAL LABORATORY
TECHNOLOGY
some professional
(particularly when different technicians work on the sponsored by m a t e r i a l s . The
External QC program is control
same procedure) m a n u f a c t u r e r s of
consistency is required. It can be Societies or by involves following
maintained by preparing written protocols, standard basic operation of these programs
operation procedures (SOPs) or procedure manuals.
steps the
Manuals should be reviewed annually and revised laboratories daily analyze
1 All the participating
material.
whenever changes OCCur. It is also good
retain outdated
practice to same lot of control
procedures in an archival file. sent to the
monthly and
O Inventory control of materials 2 The results are tabulated
analysis.
for the data
Sponsoring groups
Formal inventory control procedures will program
help minimize are prepared by the
required storage by initiating orders when the stock 3 Summary reports all participating
distributed to
reaches a certain predetermined level. sponsor and are
The laboratories.
quality
of materials should be monitored
when reference laboratories is
they are
received. The adequacy of materials should 4 The mean of values of all
"true" or correct value and is used for
be tested prior to their introduction
for routine use. taken as the
individual laboratory reported
Important part of input control is the proper labeling comparison with the
of reagents and materials. values.
Proper ic ntification should value and
include the name, lot number, date If the difference between the reported
received, then the
expiration date, and recommended storage conditions. true value is statistically significant,
is alerted, since their results
Monitoring method changes reporting laboratory
are biased compared to the results
of most of the
It is necessary to maintain a record or
log of all other laboratories.
changes and problems occurring with a method.
These logs should include the date, time,
analyst, Proficiency Testing Programs
and changes in the lots of reagents, materials or
calibrators. All instrument maintenance should be Proficiency testing (PT) is a process in which simulated
recorded along with work performed by service patient specimens made from a common pool are
personnel from outside the laboratory. The analyzed by the laboratories. A commonly used
Occurrence of problems should be indicated along evaluation criterion has been the comparison of PT test
with actions employed to resolve them. These steps results with those of pee groups. All values that exceed
are helpful in preventive maintenance program. 2 S.D. are considered "unacceptable".

Control of Analytical Quality using Patient Data To be totally "successful" in a given category, a
Intra-laboratory duplicates: Samples can be divided laboratory must produce correct results on four out of
into two aliquots and analyzed and the duplicate five specimens for each of the
analytes in that category
can be used for control purposes. and have an over all score of at least 80% for three
consecutive challenges.
Delta Checks with previous test: Certain errors,
particularly errors in specimen identification, can
FIVE Q' FRAMEWORK
be detected by comparing many laboratory test
results with values obtained on previous specimens The "Five Q" framework also defines how
from the same patient (except some tests like blood be managed objectively using the quality can
"Scientific method or
and urine glucose and cardiac enzyme profile tests). the PDCA cycle, i.e., Plan, Do,
Check, ACT. By means of
Limit Checks: A patient's test result should be "Quality planning" program, Quality laboratory
reviewed to check that they are within the processes" can be
implemented and maintained at
physiological ranges compatible with life. The limit excellent level using "Quality control"
checks are helpful for detecting clerical errors such "Quality assessment' measures, defectsmeasures. By
in "Quality
as transposed digits or misplaced decimal points. processes" can be rectified and improved by "Quality
Mean of normal: Determination of mean of normal improvement" program and accordingly, the
values of various tests can be useful in the detection planning" is revised. "Quality
of systematic errors.
Quality planning
External Quality Control (QP)
Internal QC and External QC are complementary Quality improvement
activities. If internal QC is necessary for the daily Quality laboratory
(Q)
monitoring of the precision and accuracy of the analytical processes (QLP)
method, external QC assessment is important for
maintaining the long term accuracy of the analytical Quality assessment
(QA)
Quality control
methods. (QC)
 182 TEXTBOOK OF MEDICAL LABORATORY
TECHNOLOG3Y member position in eac
ach
between
staff
in order to work with other relationship everyone involved
people to get things done. section. This
helps
It requires laboratory or command (authority)
people-oriented and task-oriented leaders the chain of
understand
who have well-developed
management skills. These to
skills include, prompt decision
making, good verbal and (Fig. 6.24).
written
Department chairman
communication, negotiation, sensitivity to the
need of
(Laboratory Directory)
others, and being visionary. Continuing
education provides ample opportunities to
acquire,
improve, reaffirm and maintain current knowledge and
skills in leadership and
management. Following are the Department manager
basic managerial responsibilities:
Offering quality leadership
Laboratory medicine Support
Personnel management Anatomical services
directors
pathology
Directing preparation of policy manual
Directing preparation of standard operation Central processing
procedures (SOPs) of management and laboratory Histology supervisor Chemistry Phlebotomy
Autopsy service Coagulation
procedures Hematology Clerical services
Laboratory information
Directing criteria-based job descriptions Immunology services
Microbiology
Managing recruitment and staffing Toxicology
Managing in-service and continuing education chart.
Fig. 6.24: Generalized organizational
Conducting staff meetings
Controlling personal records Good Management Skills and Personal Characteris-
Maintaining performance evaluation/appraisals tics
Maintaining discipline The personal characteristics and management skills of
Caring legislative and regulatory issues the laboratory manager determine the daily work
Financial management environment of the clinical laboratory. The summary
chart of CLIA88 personnel qualifications for a laboratory
Effective marketing
director for moderate complexity testing (refer to
Managing medico-legal concerns chapter 1) are as follows:

Organization of the Laboratory Personnel Qualifications

Organization of the clinical laboratory includes both
Laboratory Director
structure and process. Structure refers to framework
or
M.D. (Pathology) with 1 year
and process to interactive relationship and systems. directing supervisin9
non-waived laboratory testing
Three important elements within the laboratory are
Or
the work place, staff and tasks to be performed. A cost
effective and successful laboratory service depends on Ph.D. in chemical,
physical, or clinical
careful planning and design to meet current and laboratory sciences and Certifiedbiological
foreseeable needs of personnel, equipment and space. or ABMLI with 1
by ABMM, ABCC, ABB
year directing or supervising non-waived
Following are the main facility design processes: laboratory testing
Preparation: This includes, assessment of require- Or
ments and recruitment of staf, and awareness of Master's degree in chemical,
Current and anticipated technological changes. biological or
clinical laboratory sciences and physical,
one
Schematic design: This aspect considers, structural training and one year Supervisory year of laboratory
design, architectural design, system options, which Or laboratory experience.
include- plumbing, electricity, heating, ventilation
and air conditioning. Bachelor's degree in
chemical, physical, biological
science, clinical laboratory
Design development: Interior design, colors, fabric, science, medical laboratory
textures, etc.; are considered for this type of process. technology with 2 years laboratory
training and 2 years
supervisory laboratory experience
Construction: Bidding, negotiating contracts, legal
documentation, completion and occupation
Reference: This is an
procedures and related documents are considered federal abridged version of the code of
for construction related processes.
regulations, chapter 42, section 493.
should have a generalized
Pathology department
that shows the reporting
Qualifications for laboratory director for High
organizational charts laboratory are as follows: complexity