The Westgard rules can be very helpful in determining errors, but can be confusing. Here are some hints and guidelines on using the Westgard rules:

• Run at least two controls, one normal and one abnormal. Each should be plotted on its own chart.
• The Westgard rules call for accepting a run if the control measurements are within 2 standard deviations. However, it is still possible for all measurements to be within this limit, and still violate rules 10_x or 4_1S. You may want to check for violation of these two rules, even if the run passes rule 1_2S. The 1_2S rule is meant to simplify and speed up error-checking, and using it may result in fewer errors detected. Visit the www.westgard.com for more information.
• For the 2_2S, 4_1S, and 10_x rules, make sure you review the normal controls, the abnormal controls and a combination of the two. For example, the 10_x rule applies if the past 3 normal controls and the past 7 abnormal controls have all been above their respective means.
• For the rules that look back over several runs, it may be necessary to look at the control charts for previous months.
• The rule that is broken provides a clue as to whether the error was systematic or random. This can aid in diagnosing the problem with the procedure.
• If any rule is broken, do not report patient results until the problem, if any, has been resolved.
• Once the problem has been resolved, it may be necessary to redo patient samples from previous runs, especially if the error was systematic.