Why run controls at more than one concentration?

Choose levels around the results you need to monitor

Start with the concentrations at which patients’ results are used and where a change in analytical performance could matter. A control near a relevant decision concentration can help monitor that part of the range. Another level may help detect changes elsewhere. The choice also depends on the method, control materials available and the laboratory’s QC plan. [1] [2]

Using a second concentration answers a different question from measuring the same control twice. A repeat tells you more about that material at its own concentration. Another level adds a different view of the measuring range. Give each level its own target and limits, based on the procedure and conditions it is meant to monitor.

Read the two charts together

A fictional laboratory monitors one method with controls near 20 and 100 units. It records an analyser service between runs 5 and 6. From run 6, the low control results are 22.0, 22.2 and 22.1 units, each above its established upper limit of 21.8. The high control continues to give results close to 100 units.

The chart displays each result as a number of SDs from its own target. This makes the patterns easier to compare even though the concentrations and SDs differ. The high control gives a useful observation about its own level. It does not resolve the signal at the low level.

A useful control level needs to be measurable reliably and suitable for the procedure. Its behaviour may differ from a patient specimen, so the level and material both matter. If the available low control sits well away from the concentration used for a decision, record the limitation and consider what other evidence covers that part of the range. [2]

Decide which part of the range needs attention

The pattern gives the laboratory a reason to look at the low level and the events around the service. Possible explanations include the control material, its preparation, calibration or method behaviour at that concentration. The chart alone cannot distinguish these. Comparing the records at both levels helps define the question to investigate. [1] [2]

Describe what each level showed

The two levels provide evidence at two points. They do not sample every concentration in between. Where performance near another decision level matters, consider whether the available controls and other evidence give you an adequate view of that part of the range. Record why you selected the levels and what they help you monitor. [1] [2]

Example wordingAfter service, the low control exceeded its upper limit in runs 6–8. The high control remained near its own target. The signal was confined to the low control in the data reviewed, and its cause and relevance to patient results require follow-up.