Basics of quality assurance laboratory services

BASICS OF QUALITY ASSURANCE
LABORATORY SERVICES

1. Diagnosis purposes (to confirm or rule out a
disease.
Purposes of laboratory investigations
- LFT for patients with jaundice to confirm
hepatitis
- Brucella or widal test to rule out brucellosis
or typhoid fever.
2. Prognosis purposes
- Tumor markers for early detection of
malignancy.
- Microalbuminuria for early detection of
diabetic nephropathy

3. Follow-up purposes
- Increasing titre.
- Decreasing titre.
- Stable titre.
4. Screening purposes
5. Research purposes

The types of laboratory investigations
1. Qualitative investigations.
2. Semi quantitative investigations.
3. Quantitative investigations.

1. Patient samples:
Blood, serum, plasma, urine, stool, CSF,
swabs for cultures, biopsy, body fluids.
There are certain specifications for the
collection, handling and storage of each
specimen, e.g.:
- 24 hours urine sample.
- Fresh early morning sample.
- Random urine sample.
Samples of laboratory investigations

2. Standard sample:
Is the substance of sufficient purity and
stability and of fixed concentration used for
comparison purpose

3. Control sample:
Is the substance which is chemically and
physically similar to the unknown sample and
of known concentration.
it is essential that lyophilized control samples
be reconstituted and handled with good
quantitative technique and must be allowed
to dissolve completely.
Reconstituted control samples must be
protected from deterioration due to bacterial
action (glucose), expose to light (bilirubin)
and evaporation (all analytes)

Variations in repeated investigations are due
to errors that may take place at any stage
involved in requesting, performing and
evaluating these investigations.
Variations in repeated investigations

1. Pre-instrumental sources:
Sources of variations
- Preparation of the patient.
- Obtaining the specimen.
- Processing the specimen
- Specimen interference.
- Storing the specimen prior to the
measuring step.

2. Instrumental sources:
- Dispensing a sample aliquot into a reaction
vessel.
- Combining the sample with one or more
reagents.
- Recording some physical-chemical
consequence of the reaction.
- Calculating the value of the quantity measured.

3. Post-instrumental sources:
- Accepting the result of the test by the
technician as being of good quality.
- Sending the report of the test to the requesting
physician.

It is a statistical study of errors that could be
due to patient preparation, laboratory
personnel, laboratory technique, or used
reagent in order to recognize and minimize
them. It is the study of imprecision and
inaccuracy.
Quality Control

Factors affecting quality control
1. Control on laboratory supplies.
2. Control on quality of patient specimen.
3. Selection and continuing education of
laboratory personnel.
4. Selection and maintenance of equipment.
5. Communications.
6. Evaluation of new laboratory procedure.
7. Implementation of quality control programme.

1. Mean value (X):
It is the total score of all measurements
divided by the number of measurements.
2. Standard Deviation (SD):
It is the degree of dispersion of results of the
repeated laboratory investigations around the
mean value, i.e. the degree of reproducibility.
3. Coefficient of variation (CV):
It is the standard deviation expressed in
percentage.
Statistical features involved in quality
control implementation

4. Allowable limit of error (ALE):
It is the accepted error expressed in
percentage.
5. Deviation index (DI):
It is the difference between the individual
laboratory’s result and the mean calculated
from the results of all laboratory.

1. Imprecision:
The degree of dispersion of repeated
laboratory results around their mean value,
expressed as: SD, CV or DI.
2. Inaccuracy:
The degree of dispersion of repeated
laboratory results around the true value.
Interpretation of quality control results

3. Random errors:
Are those errors in the results of the repeated
analysis of the same sample due to variations
in apparatus, temperature, weighing, … etc.,
which are either positive or negative errors,
and are usually equally distributed.
These random errors can be reduced by more
précised method and more accurate
equipment, but can never be completely
avoided.

4. System errors:
Are those errors in the results of the repeated
analysis of the same sample due to variations
in analytical method, technical performance,
reagents, equipment or technicians which
may cause too high or too low, continuous
increasing or continuous decreasing results,
and are generally avoidable and detectable.
5. Trend in the quality control chart:
Is the gradual decrease or increase in the
results of the repeated analysis of the control
sample.

6. Shift in the quality control chart:
Is the sudden consistent change in the
results of the repeated analysis of the control
sample in either side of the mean.
7. Dispersion in the quality control chart:
Is a wide range of scatter of the results of the
repeated analysis of the control sample
around their value due to random errors.

8. Specificity:
Is the ability of method to determine only the
substance which is supposed to measure.
The greater the specificity of a test, the fewer
the number of false-positive results.
Total number of negative results
Specificity = --------------------------------------------------
Total number of uninfected patients
95
= ------------------------------------------------- = 95%
100

9. Sensitivity:
Is the ability of method to determine the
lowest single value after which the lower
value is not detected. The greater the
sensitivity of a test, the fewer the number of
false-negative results.
Total number of Positive results
Sensitivity = ------------------------------------------------
Total number of infected patients

10. Analytical range:
The lowest and highest values that method
can measure.
11. Detection limit:
The smallest single result that can be
distinguished from blank.
12. Linearity:
The minimal and maximal values at which
the method can produce a reliable result due
to linear reaction without sample dilution or
concentration.

A variety of statistical control techniques have
been used in clinical laboratories.
Tabular records with appropriate calculations
can be used to implement the techniques, but
graphical displays are often easier to interpret.
Therefore, control charts have been accepted as
a more effective way to implement most control
techniques.
The Levey-Jennings chart has been the most
widely used technique.
Interpretation of quality control chart

Levey-Jennings Control Chart:
The control results are plotted on the Y-axis
versus time on the X-axis. This chart shows the
expected mean value by the solid line in the
centre and indicates the control limits or range
of acceptable values by the dashed lines.
The usual way of interpreting this control chart
is to consider the run to be in control when the
control values fall within the control limits, and
to be out of control when a result exceeds the
control limits.

When changes in control data indicate that the
performance of an analytical method has
deteriorated, the analyst must determine the
cause of the problem.
It is generally useful first to try to classify the
error as random or systematic, because the
different kinds of errors suggest different
sources. Random errors show a wider range of
scatter of the points on the control chart, while
systematic error can be seen when the points
drift or shift to one side of the central line.
Further information on the nature of the
systematic and random errors were previously
mentioned.

+2 SD
+1 SD
X
-1 SD
-2SD
Days
Interpretation of quality control charts
Figure 1 - Levey-Jennings chart showing the analysis is in control

+2 SD
+1 SD
X
-1 SD
-2SD
Days
Figure 2 - Levey-Jennings chart showing a sudden shift in values

+2 SD
+1 SD
X
-1 SD
-2SD
Days
Figure 3 - Levey-Jennings chart showing a shift of more than 5
values on the same side of the mean and therefore out of control

+2 SD
+1 SD
X
-1 SD
-2SD
Days
Figure 4 - Levey-Jennings chart showing a gradual trend toward
lower values.

HIGH PRECISION AND POOR ACCURACY
⚫ = Target (True Value) X = Shots (Observed Values)

IMPRECISED AND INACCURATE

HIGH PRECISION AND HIGH ACCURACY

POOR PRECISION AND POOR ACCURACY

Internal quality assurance
Make sure that:
1. Normal and abnormal control samples are
available as frozen aliquots.
2. Control charts are available and updated.
3. Control results are analyzed by the technician
on monthly basis.

External quality assurance
1. Should be performed on quarterly basis.
2. Results are to be discussed with technicians.
3. Should be reported to SLSO.

Basics of quality assurance laboratory services

More Related Content

What's hot (20)

Similar to Basics of quality assurance laboratory services (20)

Recently uploaded (20)

Basics of quality assurance laboratory services