We are receiving a lot of questions about relevant analyses in the Analyse-it Method Validation edition to help in evaluating new diagnostic tests in the fight against COVID-19. Below are some quick links that will help, but contact us if you have questions - we are working as normal.

CLSI EP12-A2 for diagnostic accuracy see Sensitivity/Specificity, Predictive values, Estimating the sensitivity and specificity of a binary test and Comparing the accuracy of two binary diagnostic test.
CLSI EP17-A2 for detection capability of molecular methods using Probit analysis, see: Estimating the detection limit using a Probit fit and may be useful for determining the cutoff of a method with an internal continuous response see: Estimating the detection limit of a measurement system and Estimating the detection limit of a measurement system using a precision profile.

Also see our latest blog post: Sensitivity/Specificity and The Importance of Predictive Values for a COVID-19 test

EP12-A2 Example: Sensitivity specificity of 2 diagnostic tests

EP17-A2 Example 1:
LoB and LoD

EP17-A2 Example 2:
LoD from precision profile

EP17-A2 Example 4:
Probit detection limit

Estimating precision for a multi-site study

Estimate the precision of the measurement procedure at multiple sites and samples.

Open the file tutorials\EP05-A3 Appendix B.xlsx.
The worksheet opens showing 4 columns. The Sample column identifies the levels of the analyte in the sample (6 levels). Laboratory identifies the site (3 laboratories), Run identifies the run (5 runs per site). CA19-9 (kU/L) identifies the measured value for 5 replicates of sample in each run.
Click a cell in the dataset.
On the Analyse-it ribbon tab, in the Statistical Analyses group, click Precision, and then click 2 Factor Nested.
The analysis task pane opens.
In the Y drop-down list, select CA-19.
In the Factor A drop-down list, select Laboratory.
In the Factor B nested within Factor A drop-down list, select Run.
In the By drop-down list, select Sample.
In the Estimator drop-down list, select SD.
In the Confidence Interval edit box, type 95%, and then in the Method drop-down list, select Exact / Satterthwaite.
In the Conditions drop-down list, select Reproducibility / Repeatability to label the innermost variance component repeatability and the outermost reproducibility.
On the Analyse-it ribbon tab, in the MSA group, click Variability plot.
On the Analyse-it ribbon tab, in the MSA group, click Identify Outliers, and then on the analysis task panel in the Granularity drop-down list, select Level, Factor A so that outliers are detected for each sample and laboratory.
Click Calculate.
The analysis report opens.

The variability plots show a simple visual assessment of the closeness of agreement between the measured quantity values. The purple lines show the mean of each run, the light blue lines show the mean of each laboratory, and the dark blue line the overall grand mean.

You should observe the scatter of the points to ensure there are no obvious problems. No individual measurements stand out as highly aberrant relative to the bulk of the data and none of the plots exhibit any apparent drift capable of distorting the results.

Note: To see an example of a dataset that contains outliers and how to deal with them see the EP15-A3 tutorial.

The abbreviated variance components table shows the required precision statistics expressed numerically as the standard deviation (SD) and coefficient of variation (CV).

The detailed variance components table show the precision expressed numerically as the chosen measure of imprecision along with a confidence interval for each component.

Note: You can change the imprecision to be expressed as SD, Variance, or CV and you can also change the method of computing the confidence intervals. We prefer the Exact / MLS method as the coverage is closer the nominal level in most cases.

Next topic: Fitting a precision profile