What are control charts?
Control charts are a visual depiction of ongoing data points that various statistics are applied to. The EPA used to require a laboratory to adjust its spike acceptance recovery percentage based upon how spikes were recovered over time (typically > 20 recovery events).
The problem in doing this is that it almost always tightens up recovery limits, making it difficult to maintain production levels due to re-batching and re-analyzing client’s samples. When that went away (due to different requirements from Standard Methods and ASTM) then the charts were used to trend data points. Trends were applicable to instrument degradation and the improper construction of the spikes. Generally the only people who see control charts are analysts and anyone assessing their work.
When and how should control charts be used?
Control charts should be viewed on a continuous basis so that any trends can be caught early and rectified.
The main roadblock to control chart usage is when not all of the data points are accounted for, or when obvious statistical outliers are included. Either of these scenarios will skew the data and not let you see trends up or down or trends on either side of the mean. Active data observation will allow for proper control chart construction.
The SVL Analytical difference
Control charting can be done with a LIMS, Excel, or by hand calculations. SVL uses their LIMS to provide real time chart creation. For example, if an LCS came in low, the analyst could pull in all the data since the last LCS was made, to determine if it is a one time event or the spike was made too high or low. If trends show multiple incorrect spike amounts, the trainee will need further training.
Opportunities to learn more
If you’re curious to learn more about control charts, the below resources are a good place to start: