Accuracy - unina.it

Download Report

Transcript Accuracy - unina.it

Palermo – April 26, 2014
Controlling the quality of
diagnostic tests
E. Bravi
Palermo – April 26, 2014
Precision & Accuracy
 In the fields of science, engineering, industry, and statistics,
the ACCURACY of a measurement system is the degree of
closeness of measurements of a quantity to that quantity's actual
(true) value.
 The PRECISION of a measurement system, related
to reproducibility and repeatability, is the degree to which repeated
measurements under unchanged conditions show the same results.
 Although the two words, precision and accuracy, can be
synonymous in colloquial use, they are deliberately contrasted in the
context of the scientific method.
Palermo – April 26, 2014
Precision and accuracy
Precise and
accurate
Accurate
Precise
Neither precise,
nor accurate
Palermo – April 26, 2014
Purpose of the exercise
To check how the performance of different test
results may influence the follow-up of patients
submitted to a Gluten Free Diet in various clinical
settings.
Palermo – April 26, 2014
Materials available




7 vials containing ready-to-use samples.
Samples must not be diluted
Each sample has a different anti-tTG concentration
Anti-tTG concentration is unknown
Palermo – April 26, 2014
Protocol
1. Each centre will be assigned a code by MediCel
2. Register data results into the provided spreadsheet
3. Complete the spreadsheet and send it back to the
evaluation centre
4. The evaluation centre will provide all participants with
overall and their results
Palermo – April 26, 2014
Assay Protocol
1.
2.
3.
4.
Allow all reagents to reach room temperature
Run your own assay as usual
Include unknown samples as ready-to-use specimens
Register data results into the provided spreadsheet
Palermo – April 26, 2014
Palermo – April 26, 2014
Palermo – April 26, 2014
How to report results
Palermo – April 26, 2014
Interpretation
Results will be used as follows:
 All values:
 To generate a consensus Mean.
 To define acceptability range (Mean ± 2 St. Dev.)
and for each method:
 To calculate Mean and Standard Deviation
Palermo – April 26, 2014
Overall data - Example of graph
+2 St Dev
*
*
Mean
*
*
*
*
-2 St. Dev
Lab #
*
*
1
*
2 3 4 5 6 7 8
Cumulative Mean
*
*
Your lab
9 10
Palermo – April 26, 2014
Questions?
Thanks for your attention