Accuracy and precision = Reliability
Reliable results from patient samples, regardless of whether they are analysed at the bedside by POCT or in the laboratory, and regardless of which methodology is used, must be accurate and precise.
Pre-analytical factors such as the quality of the sample itself or the competence of the operators account for most POCT errors. However, analysers must also work reliably to get correct results.
Accuracy definition
If you analyse a sample and you get the correct result you may say that you demonstrate good accuracy.
Precision definition
If you repeat the analysis a number of times and get more or less the same result each time, i.e., with little variation of results, you may say that you demonstrate good precision.
There are four variations possible:
Good Accuracy Good precision: Correct results with acceptable variation.
This is the only acceptable option!
Poor Accuracy Good Precision: Incorrect results but an acceptable degree of variation.
Good Accuracy Poor Precision: Correct results but unacceptable degree of variation.
Poor Accuracy Poor Precision: Incorrect results with unacceptable degree of variation.
A graphical way of looking at this would be to imagine a whole series of darts hitting a dart board.
 |
 |
 |
 |
|
Good Accuracy Good Precision |
Poor Accuracy Good Precision |
Good Accuracy Poor Precision |
Poor Accuracy Poor Precision |
The calculation of mean (average) values, Standard Deviations (SD) and Coefficients of Variation (CV%) will be sufficient to determine acceptable levels of Accuracy and Precision. Normally these calculations will need to be done only when setting up a new test.
For information about daily QC requirements in order to ensure continuing Accuracy and Precision, please refer to the QC page.
If you would like to know more about this fascinating subject, please contact the POCT Coordinator.
