Westgard rules

Westgard rules are statistical decision rules used to interpret quantitative control results. They examine distances from the control mean and patterns across observations to identify possible analytical error. Familiar examples include a result beyond three standard deviations, consecutive results beyond two standard deviations on the same side, or patterns suggesting a persistent shift. Exact interpretation depends on the configured rule set.

A laboratory should specify which rules apply, how many control levels and observations are considered, and whether each signal is a warning or a rejection. Not every rule is appropriate for every method. Selecting too many sensitive rules can create frequent false rejections; selecting too few may miss clinically important errors. Staff need clear instructions for investigating signals and managing patient results.

For accreditation, the important evidence is a justified control strategy and consistent implementation, not simply naming a familiar rule set. Document the selection rationale, configuration, training, and response to violations. Verify that instrument or software settings match the approved procedure, including how results from different levels and runs are combined in rule evaluation.