Accessible Unlicensed Requires Authentication Published by De Gruyter September 21, 2011

The exponentially weighted moving average (EWMA) rule compared with traditionally used quality control rules

Kristian Linnet

Abstract

Background: The exponentially moving average (EWMA) rule for internal quality control is a well-known type of control rule in industry. Here, the power of the EWMA rule is evaluated to outline the potential of this type of control rule in clinical chemistry.

Methods: Using simulations, the power of the EWMA rule was explicitly compared with that of commonly used rules in clinical chemistry. The type I error levels were standardized to common values to achieve unbiased comparisons.

Results: For small to moderately large errors (systematic errors up to 2–3 standard deviations), the EWMA rule outperforms simple rules (N=1) and multi-rules (N=2–6). For example, for a systematic error of 2s, the EWMA rule equivalent to the 13s rule has a power of 0.30, whereas the 13s rule only displays a power of approximately 0.15. For N=4, comparison was carried out with the 13s/22s/R4s/41s rule. Here the common type I error level is 0.017. At all error levels, the EWMA rule is superior to the multi-rule. For example, given a 1s systematic error, the EWMA rule has a power (0.4) of twice the value of the multi-rule (0.2).

Conclusion: The EWMA rule is an efficient control rule with regard to systematic errors that should be considered for general application in the field of clinical chemistry.


Corresponding author: Kristian Linnet, MD, PhD, Department of Forensic Chemistry, University of Copenhagen, Frederik V's Vej 11, 2100 Copenhagen, Denmark Phone: +45-35326100, Fax: +45-3532 6085,

References

1. Roberts SW. Control chart tests based on geometric moving averages. Technometrics 1959; 1:239–50. Search in Google Scholar

2. Crowder SV. Design of exponentially weighted moving average schemes. J Qual Technol 1989; 21:155–62. Search in Google Scholar

3. Cembrowski GS, Westgard JO, Eggert AA, Toren EC Jr. Trend detection in control data: optimization and interpretation of Trigg's technique for trend analysis. Clin Chem 1975; 21:1396–405. Search in Google Scholar

4. Neubauer AS. The EWMA control chart: properties and comparison with other quality-control procedures by computer simulation. Clin Chem 1997; 43:594–601. Search in Google Scholar

5. Jansen RT, Laeven M, Kardol W. Internal quality control system for non-stationary, non-ergodic analytical processes based upon exponentially weighted estimation of process means and process standard deviation. Clin Chem Lab Med 2002; 40:616–24. Search in Google Scholar

6. Westgard JO. Internal quality control: planning and implementation strategies. Ann Clin Biochem 2003; 40:593–611. Search in Google Scholar

7. Linnet K. Mean and variance rules are more powerful or selective than quality control rules based on individual values. Eur J Clin Chem Clin Biochem 1991; 29:417–24. Search in Google Scholar

8. Winkel P, Zhang NF. Serial correlation of quality control data – on the use of proper control charts. Scand J Clin Lab Invest 2004; 64:195–204. Search in Google Scholar

9. Marquis P. [A new software for quality control at a laboratory]. Ann Biol Clin (Paris) 2000; 58:233–6. Search in Google Scholar

Received: 2005-12-20
Accepted: 2006-1-5
Published Online: 2011-9-21
Published in Print: 2006-4-1

©2006 by Walter de Gruyter Berlin New York