POST ANALYTICAL PHASE OF TOTAL TESTING PROCESS

LABORATORY TESTING

INTRODUCTION
POST ANALYTICAL QUALITY
• This is the ultimate check on the
consistency of pre-analytical and
analytical quality and can be considered
as the overall quality.

• It ties together:
✓ The analytical quality achieved
✓ The context of the patient
✓ The perceived abilities of the POST ANALYTICAL PROCEDURES
physician to interpret and utilize the
laboratory information • The post-analytical procedures
performed within the laboratory
• Similar to the pre-analytical step, the include:
post-analytical phase can be ✓ Verifying laboratory results
subdivided into: ✓ Feeding them into the laboratory
✓ One phase performed within the information system
laboratory ✓ Communicating them to the
✓ Post-post analytical phase clinicians in a number of ways
- Where clinicians receive, ▪ Producing a report and making
interpret, and react to laboratory any necessary oral
results communications regarding
FACTORS INFLUENCING QUALITY “alert” or “panic” values
✓ Report generation without any
Pre- Post- transcription errors
Analytical
analytical Analytical ✓ Double checking of printed report
Right Laboratory and counter signed by a pathologist
Recording
Specimen professionals
or senior laboratory scientist
Right
Reagents Interpretation ✓ Report dispatch to right person (kasi
collection
meron diretso sa physician or
Right Turnaround
Equipment diretso sa patient [outpatient
labeling time
Right Selection of Report to right usually])
quantity test - SOP user ✓ Storage of reported material
Right (usually 1 copy for patient, 1 copy
Records for physician, 1 copy for laboratory)
transport
Right ✓ Disposal of specimen
Biosafety
storage ✓ Monitor of turnaround time
 • Review and evaluate the following:
✓ Effectiveness of the corrective
actions
✓ Procedures and policies to prevent
recurrences
✓ Accuracy and completeness of
results and reports
✓ Disposition of unacceptable
specimens
✓ Turnaround times
✓ Referral specimens and their
reports
POST ANALYTICAL ERRORS ✓ Corrected reports
✓ Procedures for notification of test
• Post-analytical causes of errors results with statistics
accounting for 18.4 – 47% of total ✓ Assurance of confidentiality of
laboratory errors patient’s information
× Transcription errors
× Wrong validation ASSESSMENT OF ANALYTIC
× Excessive delay in reporting values CORRECTNESS OF RESULTS
× Incorrect reference values ALARMS AND FLAGS
× Physician not notified of a panic or
critical value o Automated analyzers can flag
× Incorrect interpretation of lab results specimens that require additional or
by physician repeat testing before results are
× Incorrect data entry released by specialized middleware
or by the laboratory information
VALIDATION system.
MANUAL TEST VALIDATION
o This is a time-consuming process o Flags can indicate a problem with the
with large inter-individual variation specimen (e.g., the presence of an
o It slows down the response of the interfering substance) or an issue
laboratory, thus causing delay in the with the result (e.g., a numeric value
diagnostic and therapeutic process outside the analytic range of the
o Example: method, or the need for confirmation
- Urinalysis tapos ikaw medtech by an additional assay).
sa clinical microscopy. Before
releasing results, somebody FLAGS FOR PROBLEM SPECIMENS
must countersign, esp if marami
abnormalities parang sistema ng o Many automated instruments can
feu measure the amount of sample
present in a collection tube and flag
POST ANALYTICAL ACTIVITIES samples that contain amounts that
are inadequate for a reliable
• Automated Validation System analysis.
o High sensitivity
o High specificity o Another frequent cause of
inadequate samples is the presence
 of high concentrations of interfering immediately to a health care provider
substances in the specimen, who can provide necessary medical
▪ Lipids (lipemia), hemoglobin interventions.
(Hb) (hemolysis),
paraproteins o The laboratory then has to document
(gammopathies), or bilirubin the event, including the name and
(icterus). title of the caregiver who is notified,
▪ The mechanism for this the time and date of notification, and
interference is dependent on the read-back by the care provider.
the substance and the
analytic method. REFERENCE RANGES

o Automated analyzers are also able o Reference intervals are usually

to detect troublesome levels of defined as the range of values into
interfering substances, even when which 95% of non-diseased
they are not apparent to the (“normal”) individuals will fall; the
laboratorian at the macroscopic corollary of this definition is that 2.5%
level. of non-diseased individuals will have
laboratory results below the
o Automated systems can measure reference range, and 2.5% of non-
the concentrations of bilirubin, lipid, diseased individuals will have
and hemoglobin in samples and can laboratory results above the
report the degree of interference as reference range (CLSI, 2008).
an index.
FACTORS THAT INFLUENCE
REFERENCE RANGES
DELTA CHECKS
✓ Different laboratory methods
o Delta checks are defined as often yield significantly different
comparing a current laboratory result results and therefore require
with results obtained on a previous different reference ranges.
specimen from the same patient.
✓ Differences in age, genetic
o Delta checks can detect pre-analytic background, or exposure to
(e.g., mislabeling of specimens) and environmental factors, different
analytic issues (e.g., aspiration of populations may need different
insufficient sample volume by the reference ranges for certain
instrument sample probe). laboratory analytes.

✓ Many other factors can affect

ASSESSMENT OF CLINICAL reference ranges, including
sample collection container (e.g.,
SIGNIFICANCE OF RESULTS
glass vs. plastic tubes), sample
CRITICAL VALUES transport (e.g., by messenger or
pneumatic tube), the time
o A critical value (also known as panic
between specimen collection and
value) is a laboratory result that may
analysis, and sample storage
represent a life-threatening situation
before analysis.
that may not otherwise be readily
detectable; it must be reported
 DETERMINATION REFERENCE ▪ Examples:
RANGES
✓ diurnal variations in
✓ Because many factors can affect cortisol levels
reference ranges, laboratories are ✓ estrogen levels that vary
strongly encouraged to perform their
with the menstrual cycle
own studies to establish reference
✓ seasonal variations of
ranges for all analytes they report,
usually by testing at least 120 vitamin D
samples from non-diseased ▪ Many other analytes show
individuals in each “partition” (e.g., some biologic variability,
gender, age group).
including changes related to
✓ If this is not possible, the laboratory exercise or food intake.
can verify a reference interval that it
has previously established for a
different method by transference (i.e., GENERAL PRINCIPLES FOR THE
demonstrating that the new method
yields identical results to the previous INTERPRETATION OF LABORATORY
method). RESULTS
✓ The laboratory can verify another
laboratory’s or the manufacturer’s
reference interval if the analyte was
not previously tested for in the
laboratory.

VARIABILITY OF LABORATORY
RESULTS
✓ Random variability
▪ Is the sum of analytic and
intraindividual variability. • A test with perfect diagnostic accuracy
✓ Analytic variability could determine the presence or
absence of disease with certainty
▪ Is the result of assay • The established cutoff point would
imprecision. perfectly separate diseased from non-
▪ It is usually determined during diseased populations
validation studies for a new
method by running the same • diagnostic accuracy of a test
sample multiple times and is o determined by comparing the test’s
expressed quantitatively as ability to discern true disease from
the coefficient of variation non-disease as determined by a
(CV). diagnostic gold standard
✓ Intraindividual variability
• true-positives (TPs)
▪ Is due to biologic changes
o patients correctly classified as
that cause analyte levels to
abnormal
fluctuate over time.
• true-negatives (TNs)
o patients correctly classified as
normal
• false-positives (FPs) SENSITIVITY AND SPECIFICITY
o patients incorrectly classified as
➢ are measures of the diagnostic accuracy
abnormal
of a test; they are indicators of a test’s
• false-negative (FNs)
ability to distinguish between disease
o Patients incorrectly classified as
and absence of disease at a chosen
normal
cutoff

❖ These true results are the ▪ Sensitivity

nonoverlapping areas of the two o is the ability of a test to detect
patient distributions. False results disease and is expressed as the
occur because the two populations proportion of persons with
overlap (i.e., because a test cannot disease in whom the test is
completely discriminate all abnormal positive
patients from normal ones)
▪ Specificity
o is the ability to detect the
absence of disease and is
expressed as the proportion of
persons without disease in whom
the test is negative

• Predictive value of a positive test or

Positive predictive value

o May be understood as the

probability that a positive test
indicates disease.
o It is the proportion of persons with
a positive test who truly have the
disease.

• Predictive value of a negative test or

Negative predictive value

• As seen in Figure 7-3, where for ease of o Is the probability that a negative
illustration a single cutoff is used to test indicates absence of disease.
discriminate disease from normal o It is the proportion of persons with
populations, varying the cutoff changes a negative test who are truly
the numbers of true and false results in a without disease.
given population.
• False results can be produced when an ❖ The predictive value of a positive test
analyte has two relevant cutoffs (e.g., is highly dependent on the
thyroid-stimulating hormone), with prevalence of the disease being
overlapping populations at both the low tested.
end and the high end. ❖ The higher the prevalence, or pretest
probability, the higher the posttest
probability, or predictive value of a
positive test.
 • Improved accuracy (sensitivity and
specificity) enhances the predictive value • P(D)
of a test. o known as clinical suspicion,
• The formula for predictive value shows prevalence, or pretest probability
that sensitivity and specificity influence - probability of disease before the
the predictive value. test result is obtained
• The predictive value of a positive test • P(D | T)
increases with increasing prevalence o posttest probability
and improved accuracy. - probability of the disease after
the test result is known
• Specificity has the biggest impact on the
• P(D | T)
predictive value of a positive test
o TP (true positive) rate
• Sensitivity determines the predictive
- probability that the test is
value of a negative test.
positive when the disease is
• The number of FPs directly influences present
the predictive value of a positive test, • P (T | 𝑫)
whereas the number of FNs has the o FP (false positive) rate
same effect on a negative test.
- probability (posttest) of disease
or no disease is calculated
BAYES THEOREM || PREDICTIVE REPORTING OF RESULTS
VALUE THEORY • Release of reports only to authorized
➢ describes the relationship between person
posttest and pretest probability of
• Timely release of provisional and final
disease or no disease based on the
sensitivity and specificity of the test. report
• Any value which exceeds the normal limit
must be clearly published, understood
and conveyed verbally, electronically or
printed form when, where, what and to
whom was reported document it
RESULT FORMAT
o Name and address of laboratory
o Name of patient with gender
o Laboratory ID number
o Date and time of receipt of sample
o Type of sample
o Name of test requested with a brief
clinical background
o Results with the units
o The normal or reference range of
the test