How to Construct a Quality Control Chart for Clinical Chemistry

Virginia Gaces's image for:
"How to Construct a Quality Control Chart for Clinical Chemistry"
Image by: 

Quality assurance encompasses four stages of laboratory analyses: patient identification, patient and sample preparation, testing or assaying and reporting or recording of results.

Quality control specifically used in the third stage of quality assurance testing, is a valuable tool in the proper performance of laboratory tests. This program monitors the reliability of results and identifies both the random and systematic errors committed, thereby helping the technologist correct the error.

Quality control as part of Quality Assurance is very vital in determining whether results obtained are accurate or not.

In the observance of a Quality Control Program, Quality Control Charts are prepared manually by clinical laboratories that do not have an automated means of doing it.

The following are the steps adapted in constructing your QCC.

1. Compute for the mean (X) by getting the sum of all the individual control values (x) and dividing this by the number of values or tests.

N.B. Control values are obtained from pooled sera or commercial sera. Aliquots of these are transferred into test tubes and tested simultaneously with the unknown and the standard.

2. Compute the difference of each value from the mean (X – x).

3. Compute for the squared difference of each value (X – X) 2.

4. Get the sum of the squared differences. ∑(X –X)2.

5. Divide the result of computation number 4 with the total number of values or tests minus 1. (N-1)

6. Compute for the square root of the quotient.

7. The value obtained is your standard deviation (sd, SD)

8. Solve for ±1SD, ±2 SD, ±3 SD.

9. Plot these SD values in a linear graphing paper. Assigning the number of tests on the x-axis and the control values on the y-axis.

10. Plot your control values.

11. Interpret the QCC, making use of the Westgard Rules. Westgard Rules include 1(2S), 1 (3S),   4 (1S), 2 (2S), 10(X), R (4S).   1(2S), R (4S) and 1(3S) indicate random errors or technical errors while 2(2S), 10(X) and 4(1S) indicate systematic errors.  A proper interpretation would help identify where the error is coming from.

12. If there are violations of these rules, then reject the procedure/method.

13. If there are no violations, then accept and use the procedure/method.

Preparing a QCC manually, maybe a tedious, but necessary process if there are no computer- based or automated methods to do it. This would ensure the reliability of clinical laboratory results.

More about this author: Virginia Gaces

From Around the Web