Which quality control (QC) rule would be broken 1 out of 20 times by chance alone?

In quality control for clinical chemistry, the 12s rule is based on the idea that a single control value exceeds the designated mean by more than 2 standard deviations. This approach is designed to detect random error; when using this rule, there is a probability of breaking it 1 time out of 20 due to chance alone, which corresponds to a significance level of 0.05. This means that in a statistically controlled process, if everything is functioning normally, you would expect to see a violation of the 12s rule about 5% of the time purely due to random variation, which translates to 1 out of 20 tests.

Other rules, such as 22s, 13s, and 14s, typically involve cumulative or multi-directional checks and are designed to catch different types of systematic errors or shifts in control. They have different statistical thresholds that result in different probabilities of detecting non-random errors, often corresponding to stricter criteria that lead to less frequent violations under random circumstances. Therefore, the 12s rule specifically characterizes the chance of random control variation and fits the criteria of being broken 1 out of 20 times.