KLUBSY: QUALITY ASSESSMENT & QUALITY CONTROL Flashcards
Preanalytical, analytical, and post analytical phase is under what quality management?
Quality Assurance
Promotes a philosophy that everyone involved in the laboratory workflow has a responsibility for a quality product
Total Quality Management (TQM)
This encompasses suppliers, health care team, management,
and patientsTQ
TQM
What are the 5 elements of TQM?
- Quality planning
- Quality lab process
- quality control
- quality assessment
- quality improvement
This Element of TQM that strive to continually improve practices and not just meet established quality standards
Continuous quality improvement (CQI)
Its goal is to reduce the number of defects to near zero by systematically identifying and eliminating causes of
defects/errors
SIX SIGMA
Its goal is to reduce variability in processes
Six sigma
What is the tool used in six sigma?
DMAIC
DMAIC stands for:
D - Define
M - Measure
A - Analyze
I - Improve
C - Control
Its goal is to eliminate/ reduce wastes (nonvalued activities) in
the production or manufacturing processes
Lean
What tools are used for Lean?
5S & PDCA
5S in lean stands for:
1-Sort
2- Set in order
3 - Shine
4 - Standardize
5 - sustain
PDCA stands for:
P - plan
D - do
C - check
A - act
What are the 8 WASTES of lean?
- Defects
- Overpopulation
- Waiting
- Non utilized talents
- Transportation
- Inventory
- Motion
- Extra processing
“DOWNTIME”
Selected employees are trained to become experts and
lead efforts
Lean-Six Sigma
Dedicate 100% of their time to QI
projects, proactively addressing
process and quality opportunities for
improvement
Black belt/Project coaches and leaders
Dedicate 20% of their time to
improvement projects while delivering
their normal job functions
green belt/project team members
Includes a full day basic LSS training,
including interactive and didactic
training program, as a prerequisite for
all participation in projects
Yellow belt
Attend a 1-day training class and
usually mid- to senior-level leader who
review the project, remove
organizational barriers and encourage
the team members
Project sponsors/champions
Systematic process that ensures a quality laboratory service
Quality Assurance
A complete system of creating and following procedures and
policies which aims to provide the most reliable patient lab
results and minimize errors in all phases of testing.
Quality Assurance
A program where overall activities conducted by the
institution are direct toward assuring the quality of the
produces and services provided
Quality Assurance
What are the 3 phases of testing?
- Preanalytical
- analytical
- post analytical
Preanalytical testing begins with what?
starts with the physician’s order
Processes prior to the actual testing, starting with the
physician’s order
preanalytical
T/F: 32% to 75% of all testing errors occur in the pre-analytic
phase
TRUE
Statistically, all testing errors occur in the preanalytical testing. What percent?
32% to 75%
Factors that affect the laboratory results due to handling
of the specimen sample prior to analysis
Preanalytical testing
The actual laboratory testing or diagnostic procedures,
processes and products that provide the results
Analytical testing
Includes standardization, preventive maintenance,
operator competence, verification of instrument linearity,
precision, accuracy and reliability, quality control,
Analytical phase
Processes following the examination including systematic
review, formatting, authorization for release, timely and
accurate reporting and transmission of results, and
storage of samples of the examinations
Post analytical phase
Phase of testing: test ordering
Preana
Phase of testing: patient ID & preparation
preana
Phase of testing: spx collection
preana
Phase of testing: spx transport
preana
Phase of testing: spx prep
preana
Phase of testing: spx storage
preana
Phase of testing: instrument selection and assay
analytical phase
Phase of testing: assay validation
analytical
Phase of testing: internal QC
analytical
Phase of testing: external qc
analytical
Phase of testing: preventive maintenance
analytical
Phase of testing: accuracy in transcription of results
Post ana
Phase of testing: content and format of lab narrative report
post ana
Phase of testing: reference interval & therapeutic range
post ana
Phase of testing: timeliness of results reporting
post ana
Phase of testing: TAT
post ana
Phase of testing: patient & physician satisfaction
post ana
a system ensuring accuracy and precision in the laboratory
including: monitoring performance of lab instrument, and
reagents, using quality control reagents in every series of
measurements
Quality control
Detects analytical errors during testing and ultimately
preventing the reporting of inaccurate patient results
Quality control
Part of performance monitoring that occurs after a test has
been established
Quality control
What are the 3 objectives of quality control?
1-check stability of machine
2-heck quality of reagents
3-check technical errors
What are the 8 parameters of QC?
1-Analytical sensitivity
2-Analytical specificity
3-Accuracy
4-precision
5- Practicability
6-Reliability
7-Diagnostic sensitivity
8-Diagnostic specificty
This specimen utilized should resemble the sample with a
known concentration of the analyte being measured in the
testing procedure
Control
These are treated in exactly the same way as any unknown
specimen
Control
Minimum control specimens for general chemistry?
2 levels of
control (run at least once each day of patient testing)
What are the 6 characteristics of an ideal QC material?
1-resembles human sample
2-inexpensive and stable for long periods
3-no communicable diseases
4-no matrix effects/known matrix effects
5-with known analyte concentration (assayed control)
6-convenient packaging for easy dispensing and storage
similar in composition with the specimen
Control
T/F: The
value of control is unknown
False.
Correct ans: KNOWN
Highly purified substances of known
composition
Standard
best way to measure accuracy
Standard
using a standard, this is the best way to measure WHAT?
best way to measure ACCURACY
Best measurements of precision
Control
with a control, this is the best measurement for what?
PRECISION
Used to calibrate an instrument or develop a
standard curve
Standard
T/F:
-Control: precision; standard: accuracy
TRUE
What are the 2 types of QC?
- Internal QC
- External QC
What are the alternate names for internal QC and external QC, respetively?
- Intralaboratory
- Interlaboratory
This is performed by the lab personnel using control materials of
known values and comparing the values to established
acceptable ranges
Internal QC
Control material values are assessed using Levey-Jennings
control charts and Westgard multirules to detect errors
Internal QC
In internal QC, control material values are assessed using what?
Levey-Jennings
control charts and Westgard multirules to detect errors
this is Essential for daily monitoring of accuracy & precision
internal QC
This kind of QC detects random and systematic error
internal QC
Proficiency testing programs that periodically provide
samples of unknown concentrations to participating
clinical laboratories
external QC
Significant in maintaining long-term accuracy of the
analytical method
External QC
Allows laboratory to evaluate the performance by
comparing the mean, SD and CV
EXTERNAL QC
In external QC, THIS Allows laboratory to evaluate the performance by
comparing what parameters?
the mean, SD and CV
In external QC, this result means that laboratory is not in
agreement with other laboratories included in the
program. What is the standard deviation index?
SDI > 2
An SDI of > 2 means what?
means POOR PERFORMANCE in external QC
What is the acceptable SDI for proficiency testing?
SDI of <2
Proficiency testing is a.k.a?
external quality assurance
external quality assurance is a.k.a?
proficiency testing
Administered by an external organization wherein they send
sets of unknown samples (PT samples or challenges) to the
laboratory for analysis
proficiency testing
what is the Gold standard for clinical laboratory proficiency testing?
College of American Pathology (CAP program)
This is the difference between the test and reference method
results
Error
An error is the difference between what?
difference between the test and reference method
results
Total Error is determined by what?
Determined by replication and comparison
What is the formula for total error?
random error + systematic error
Random error detects/measures what?
IMPRECISION
To measure/detect IMPRECISION, what error must be used?
random error
this error is one with no trend or means of predicting because they
have no known pattern
random error
it is due to chance and varies from sample to sample
random error
this type of error is present in all measurements
random error
Test used for determination of random error?
REPLICATION EXPERIMENT
Remedy for random error?
Reassay using the same rgnt
Random error is indicated by what?
STANDARD ERROR ESTIMATES due to instrument, operator, and environmental conditions
Systematic error detects/measures what?
INACCURACY
To detect/measure INACCURACY what type of error is used?
SYSTEMATIC ERROR
This type of error is predictable and cause a constant difference in results in one direction (positive or negative) or stay the same
Systematic error
this type of error is repeating
systematic error
this type of error is easily predicted
systematic error
T/F: Systematic error is seen as a trend on data
TRUE
Systematic error is determined by what?
comparison method
systematic error is assessed by what?
stepwise evaluation of procedures
this is the usually the first step taken in systematic error
recalibration of the assay
Systematic errors are due to what? (3)
- calibration problems
- deterioration of rgnts and control materials
- contaminated solutions
what are the 2 types of systematic errors?
1- constant
2- proportional
RE or SE: Pipetting error/reagent
dispensing
re
RE/SE: Sample evaporation
RE
RE/SE: Mislabeling of samples
RE
RE/SE: Improper mixing of sample and rgnt
RE
RE/SE: Temperature fluctuation
RE
RE/SE: Voltage fluctuations not
compensated for by the
instrument circuitry
RE
RE/SE: Electro-optical
mechanism
RE
RE/SE: Environmental
conditions
RE
RE/SE: Variation in handling
techniques: pipetting,
mixing, timing,
RE
RE/SE: Variation in operators
RE
RE/SE: Improper calibration
SE
RE/SE: Deterioration of reagents & control materials
SE
RE/SE:Improperly made standard solutions
SE
RE/SE: Contaminated solutions
SE
RE/SE: Wear and tear of
instruments
SE
RE/SE: Sample Instability
SE
RE/SE: Unstable and inadequate rgnt blanks
SE
RE/SE: Leaky ion selective
electrode (ISE)
SE
RE/SE: Failing instrumentation
Instrument drift &
malfunction
SE
RE/SE: Poorly written
procedures
SE
What are the measures of central tendency?
1-mean
2- median
3-mode
What are the measures of spread or dispersion?
1-range
2-Standard deviation
3- coefficient of variation
4-variance
this is the Average of the individual values in the sample set
Mean
how to calculate for the mean?
To calculate, divide the sum of all the individual values by
the number (n) of values in the set
Number in the middle ofan ordered data set/ midpoint of
the distribution
median
What is the first step in determining the median?
Rank the data set in numerical order to determine the
middle point
Most frequently found number(s) in a data set
mode
Used to describe data that seem to have two centers
mode
Referred to when in reference to the shape of the data
mode
this is the difference between the highest and lowest score in a
data
range
good measure of dispersion for small samples of data
range
what is the disadvantage of RANGE?
susceptible to extreme values or outliers
This measure of spread or dispersion is susceptible to extreme values or outliers
range
Most frequently used measure of variation
standard deviation
measure of the dispersion of values from the mean
standard deviation
this helps describe normal curve
standard deviation
this is a measure of distribution
range
standard deviation
aka Relative Standard Deviation (RSD)
Coefficienct of variation (CV)
Coefficient of variation is a.k.a
Relative standard deviation
this is the percentile expression of the mean
CV or RSD
this is an INDEX of PRECISION
CV
this is used extensively to summarize QC
data
CV
Formula for CV?
CV= (SD/Mean) x 100
this is the standard deviation squared (S)
Variance
this is the “average” distance from the mean and every value in
the dataset.
variance
represents difference between each value and the
average of the data or average ofthe squared distances of
all values from the mean
variance
formula for variance?
V= (SD)²
2 types of inferential statistics?
1-parametric analysis
2- nonparametric analysis
A Statistical test that assumes the
observed values, or some mathematical
transformation of those values, flow a
(normal) Gaussian distribution
Parametric analysis
What type of inferential statistics is MOST preferred in the laboratory?
Parametric analysis
A Statistical test that makes no specific
assumption about the distribution of
data
nonparametric analysis
T/F: Nonparametric methods rank the
reference data in order of decreasing
size
FALSE: increasing size
This assesses the accuracy using the mean
t-test
this assesses the precision using the standard deviation
f-test
t test is used to assess what?
assess accuracy using the mean
“M - mean A - accuracy T - t test” (MAT)
f test is used to assess what?
assess precision using the standard deviation
“S- standard devi P - prevision F - f test” (SPF)
Used to observe values of control materials over time to
determine reliability of the analytical method
Quality control charts
Utilized to observe and detect analytical errors such as
inaccuracy and imprecision
QC charts
types of QC charts?
1-cumulative sum graph (Cusum)
2-Tonks Youden Twin Plot
3-Levy Jenning
this systematic error Refers to the difference
between the target value and
assayed value
constant
this systematic error is INDEPENDENT on sample conc
constant
T/F: a constant systematic error is dependent on sample conc.
FALSE. INDEPENDENT
This systematic error occurs when y is NOT equal to 0
constant
Test used for determination of systematic error
interference experiment
T/F: Proportional systematic error occurs when y is NOT equal to 1
FALSE.
Occurs when SLOPE is NOT equal to 1.
This Results in greater devlation
from target value due to
higher sample concentration
prorportional
This type of systematic error is dependent on sample conc
proportional systematic error
this type of systematic error occurs when SLOPE is NOT equal to 1
proportional
Test used for determination of proportional error
recovery experiment
2 types of QC charts?
1-Cumulative Sum Graph (CuSum)
2- Tonks Youden Twin Plot
3- Levy Jenning
used for the earliest detection of systematic errors
CuSum
this type of QC chart needs to be
computerized
CuSum
*C for computerized
this QC chart is used to determine the difference between QC results and the target means
CuSum
What is the indication of an OUT OF CONTROL CuSum?
When the slope exceeds 45 degrees or
+2.7SD is exceeded
this qc chart compare results obtained on a high and low control serum
from different laboratories
Youden
Twin plot is a.k.a
youden
Plotting mean values for one specimen on the ordinate axis
(y) and the other specimen on the abscissa (x)
youden
in the twin plot, this suggests a PROPORTIONAL error
Points falling from a center but on 45° line
in the twin plot, this suggests a CONSTANT ERROR
points falling from the center but not
on the 45° line
What error is this suggesting? points falling from the center but not
on the 45° line.
constant error
What error is this suggesting? Points falling from a center but on 45° line suggest
proportional error
proportional error
Multi-rules were developed by?
James Westgard
this is used to determine whether analytical methods are in control
westgard multirules
What are the 3 Westgard warning rules?
- 1 2s
- 2 2s
- 4 1s
What are the 3 westgard mandatory rules?
- 10x
- 1 3s
- R 4s
One control value above or below +/-2 SD
1 2s
this is a warning rule
1 2s
2 consecutive controls above or below +/-2SD
2 2s
1 control above or below +/-3SD
1 3s
One control value above +2SD and one control value below -2SD in the same run
R 4s
10 consecutive controls on same side of mean
10x
4 consecutive controls above or
below +/-1SD
4 1s
T/F: All westgard rules except for 12s should be REJECTED
TRUE
For biological studies, control confidence limits are usually
established at??
+/- 2SD
What are the systematic errors in the westgard rules?
- 2 2s
- 4 1s
- 10x
*All EVEN numbers
what are the random errors in the westgard multirules?
- 1 3s
- R 4s
*Odd numbers
Are trends and shifts a systematic or random error?
systematic error
Levy Jenning is a.k.a?
Shewhart
is representative of the data used to complete the normal or gausslan (bell-shaped) curve
LJ chart
Most widely used chart that allows laboratorians to apply
multirules without the use of computers
LJ chart
What parameters should be noted on the LJ chart?
Mean and SD
A Minimum of how many measurements to calculate for the
mean and SD?
min of 20 measurements
What are the factors needed to construct a LJ chart?
- Mean
- SD
- Control values (min. of 20)
Y/N: If controls are within +/- 2SD, the run is “in control” and
should be accepted
Yes
T/F: When values fall outside +/- 2SD, procedure is out of
control
true
What is the protocol in addressing out of control results?
1- If any rule is violated, patient results should be held until
problem is resolved
2- Once problem is resolved, patient samples from
previous runs may need to be retested
3-All steps taken to resolve QC errors should be
documented
What are 2 errors seen in an LJ chart?
1-Trend
2-Shift
Abrupt, sudden, and sustained change in one direction in
control sample values.
Shift
*GTAS
G-radual T-rend
A-brupt S-hift
6 or more consecutive control values on same side of the
mean but maintain a constant level
shift
Most common cause of SHIFT errors
Improper calibration of instrument
This is the gradual change in the control sample results
Trend
*GTAS
G-radual T-rend
A-brupt S-hift
control values increasing or decreasing for 6 consecutive
means
trend
T/F: Shift is a random error and trend is a systematic error
FALSE.
Both are systematic errors
Most common cause of trends?
deterioration of rgnts
2nd most common cause of trends?
deterioration of control samples
Uneven distribution of control values outside +/-1 SD but
within +/-3 SD
Dispersion or “scattered”
A dispersion has uneven distribution of control values outside what values but
within what values?
Values outside +/-1 SD but
within +/-3 SD
T/F: Increased dispersion represents loss of precision without
loss of accuracy
TRUE
Increased dispersion = loss of what?
loss of precision
T/F: increased dispersion = loss of precision and accuracy
FALSE.
Loss of precision without loss of accuracy
Control results outside established limits
Outliers
May be due to chance or indicate problem in test system
outliers
If outliers occur more than once in how many successive runs,
invetrgation must be carried out?
20 successive runs
This include all the data points that
define the range of observations. A pair of medical decision
points that span the limited of results expected for a defined
healthy population
reference interval (ranges)
All normal ranges are reference intervals, but …
but not all
reference intervals are normal ranges
T/F: All reference ranges are normal intervals, but not all
normal intervals are reference ranges
FALSE.
All normal ranges are reference intervals, but not all
reference intervals are normal ranges
Reference intervals are calculated using what mean values?
+2SD and -2SD
Require from 120 to as
many as 700 study
individuals
ESTABLISHING a reference interval
to establish a reference interval, how many study individuals is required?
120 to as many as 700
to verify a reference interval, how many study individuals are needed?
20 study individuals
Require as few as 20 study
individuals
VERIFYING a reference interval
A bell-shaped curve and focuses on the distribution of error from the analytical
method
Normal distribution/Gaussian Curve
the VERIFICATION of a reference interval is a.k.a?
Transference studies
What are the characteristics of a gaussian curve?
1- Mean = media = mode
2- Symmetric distribution
3- Total area under the curve = 1.0/100%
What is the “rule” adhered to in normal distribution?
68-95-99 rule
What is the normal distribution if all numbers will fall between -1SD and +1SD
68%
68% of all numbers will fall between??
-1SD and +1SD
What is the normal distribution if all numbers will fall between -2SD and +2SD?
95%
95% of all numbers will fall between??
-2SD and +2SD
What is the normal distribution if all numbers will fall between -3SD and +3SD?
99%
99% of all numbers will fall between??
-3SD and +3SD
This CI l is used in part to account for certain unavoidable error
caused by sampling variability and imprecision of the
methods themselves.
99%
Determine how good a given test is at detecting and
predicting the presence of disease (or physiologic condition)
Diagnostic EFFICIENCY
what are the measures of diagnostic efficiency?
1-diagnostic sensitivity
2-diagnostic specificity
3-Positive Predictive Value (PPV)
4- Negative Predictive Value (NPV)
Diagnostic sensitivity and specificity, respectively, is a.k.a?
clinical sensitivity and specificity
Proportion with the
disease who have a
positive test result
D. sensitivity
Proportion without the
disease who have a
negative test result
D. specificity
Proportion with a
disease who have a
positive test result
compared with all
individuals who have a
positive test result
PPV
Proportion without a
disease who have a
negative test result
compared with all
individuals who have a
negative test result
NPV
What is the formula for D. sensitivity?
TP/TP+FN x100
what is the formula for D. specificity?
TN/TN+FP x 100
What is the formula for PPV?
TP/TP+FP X 100
*Lahat ng may positive
What is the formula for NPV?
TN/TN+FN X 100
*Lahat ng may negative
This measure of diagnostic efficiency prevents FALSE NEGATIVES?
Sensitivity
This measure of diagnostic efficiency prevents FALSE POSITIVES
Specificity
