Statistics in Clinical Chemistry and Quality Control Flashcards
Used with 2 or 4 control measurements per run
Westgard Rules
Limited number of values
Categories
Sum of all results divided by the number of results
Mean
Amount of variation
Coefficient of variation
Protein in urine
Ordinal scale
Uses a single criterion or single set of control limits
Single-rule Quality Control
Value that occurs most frequently in the data set
Mode
Locator
Finds a requested percentile
Warning rule to trigger careful inspection of the control data in Westgard multirule Quality Control
1(2s)
Things we measure and count
Variables
Variables can take on values in a quantitative range with defined differences between points
Interval scale
50th percentile
Median
Reject when 10 consecutive control measurements fall on one side of the mean
10(x) violation
Probability that the true value falls into a category
Confidence interval
Lowest value that an assay can reliably detect
Sensitivity
Reject when 2 consecutive control measurements exceed the same +/- 2s control limit
2(2s) violation
Accuracy is tested by ______ and ______.
Assaying patient specimens
Assaying interlaboratory survey materials with known values
Analytic specificity is characterized by _____ and _____.
Determination of interferences
Non-specific responses to other analytes
Ability to measure only the sought for analyte
Analytic specificity
Uses a combination of decision criteria
Multirule Quality Control
Scale with specific values but without equivalent distances between categories
Ordinal scale
Control rule when the control limits are set as the mean +/- 3s
1(3s)
Extent of random variation
Standard deviation
Decision criteria
Control rules
Variables can take on a limited number of values
Nominal scale
Chart with control limits mean +/- 2s or mean +/- 3s
Levey-Jennings Chart
Spread of data in which elements are distributed symmetrically around the mean
Gaussian distribution
Reject when 1 control measurement in a group exceeds the mean + 2s and another exceeds the mean - 2s
R(4s)
4(1s) violation rejects
when 4 consecutive control measurements exceed the same mean +/- 1s control limit
The ability to correctly detect and quantify and analyte
Accuracy
Precision is
the reproducibility of an assay
