Intro To Stat Concepts & QC Flashcards
Quantitative techniques & procedures which monitor performance parameters
Quality control (QC)
A broad spectrum of plans, policies, & procedures which provide for achievement of quality goals in the lab
Quality assurance (QA)
List pre-analytical factors
Ordering of tests, patient prep, patient ID, specimen collection, specimen transport to the lab, specimen handling in transport and in the lab
List analytical factors
Use of integral and external quality control programs, labeling and use of reagents, calibration procedures, maintenance procedures, monitoring of equipment
List post-analytical factors
Verification of calcs, review of results for possible error, reporting of test results, interpretation of test results, competency and adequacy of staff
Defined ranges determined by the manufacturer; used for periodic accuracy checks
Assayed controls
Each laboratory determines it’s own particular range; used to check accuracy and precision of daily patient results
Unassayed controls
3 purposes of QC programs
1 assess accuracy
2 access precision
3 identify any analytical errors which may be present that may jeopardize patient care
The measure of the correctness of a result; how close the result comes to a true value
Accuracy
A measure of reproducibility; the measure of the variability present in an analytical process
Precision
Those errors which occur without prediction or regularity ; affect SD
Random error
Those error which occur within the test system or methodology that occur regularly; affects accuracy
Systematic error
Those errors that are proportional to analyte conc; error increases in magnitude as conc increases
Proportional systematic error
Those errors which are constant or in the same amount over the entire range of analyte conc
Constant systematic error
A sample of known conc that is used to calibrate or set an instruments parameters; has a specific conc, not a range if concentrations; may be primary or secondary
Standard
A sample with a range of concentrations either determined by the manufacturer (assayed) or determined by the particular lab (Unassayed); used to determine if we may be reasonably assured that patient values are accurate and precise and may be reported
Control
A field of study concerned with the organization and summarization of data and drawing of inferences about a body of data when only part of the data is observed
Statistics
Used to summarize the important features of a group of data; concerned with the mean, range, variability, and distribution of a data set
Descriptive statistics
With enough data points, the plot will be a predictable, bell shaped curve
Gaussian (Normal) distribution
Center of distribution determined from the sum of the values divided by the number of values. Average, expressed as x, reported in same units as test. Representative of the central tendency of population only if histogram is symmetric about the mean.
Mean
Most frequently obtained value
Mode
The value in the middle of the distribution
Median
The difference between the highest and lowest values in a set
Range
A measure of the average squared distance of data points from the mean
Variance
Describes the spread of data about the mean.
Standard deviation
The size of the standard deviation in relationship to the mean; expressed as percent; useful to compare deviations of different pools and different data
Coefficient of variation
Limits between which a specified proportion of a population is expected to lie
Confidence limit (or interval)
% within +- 1 SD
68.2%
% within +- 2 SD
95.5%
% within +- 3 SD
99.7%
Causes of constant error
Presence of interfering substances in all samples or reagents; presence of interfering substances in reaction between the analyte and reagent; presence of interfering substances that destroy or inhibit the reagent
Causes of proportional error
Incorrect assignment of the amt of substance in the calibrator; occurrence of a side reaction of the analyte
Lab confidence limits
95% confidence or a mean of +- 2 SD, 1 out of 20 will be out
Purpose of internal QC
To ensure accuracy and precision of testing methodology
CLIA ‘88 requirements
Two levels of QC must be analyzed every 24 hours for each test performed that day
Time frame assessed in internal QC
Daily, “real time”
The period of time between which the instrument or procedure is expected to be stable or period of time between QC analyses; run 2 levels at least every 24 hours for daily tests
Analytical run
Purpose of external QC
To compare quality of our lab to those in other parts of the state/country
External QC time frame assessed
2-3 months before results are compiled
Process of external QC
Pooled samples distributed (CAP); analyzed like any patient sample
What do you do if QC is out?
1 check lot # 2 rerun same bottle of QC 3 run another bottle of same level QC 4 run another level of QC 5 check levey-Jennings charts 6 run assayed control 7 recalibrate or perform maintenance
QC #1 ⬆️ QC #2 ok
Repeat QC #1
Repeated QC #1 and still ⬆️
See if bottle expired, new bottle, can rule out reagent or instrument
Both QC #1 & 2 are ⬇️
Instrument or reagent problems, repeat first
QC #1 ⬆️ QC #2 ⬇️
Instrument or reagent problem, calibrate, look at levey-Jennings
