Notification After the Fact

What does your lab do when you are notified about instrument/reagent problems after the fact? Let me give you two recent examples.

A vender sends a recall notice that goes back 6 months on one of their reagents. There has been a problem with a reagent for 6 months, but you are just now being notified. You look and determine that you have run 1000 tests on patients using that reagent during the problematic time period. What do you do?

Or another example, a reference laboratory changes their units of measure and you miss the notification. 23 patients had results come back from the reference lab and the values were entered into the computer with the incorrect units of measure and reference interval in place. The mistake is discovered on the 24th patient.

Laboratories work to have systems in place to prevent these scenarios. We try to carefully watch for changes from reference labs, and we hope that recall notices will not affect many patients. However, despite our best efforts, relatively large-scale lab errors like these happen. Laboratories need to have systems in place for dealing with these events. A few steps that are helpful when this happens include:

1) Determine whether the error is on-going
  a. Are immediate process changes necessary?
2) Fix the error so no new patients are affected
3) Determine the extent of the damage
  a. Often involves a Root Cause Analysis
  b. How many patients were affected
  c. How serious was the effect (changed treatment vs no effect at all)
  d. Do healthcare-providers need to be notified, and if so, who and how to notify
4) Notifications
  a. Healthcare-providers
  b. Hospital/laboratory Quality and/or Risk Departments
  c. Other clients
5) Documentation
6) Is there a way to turn this into an opportunity for improvement?

Occasionally an error can cause more headache than harm. For example, if you report a value of 35 mg/dL with a reference interval is 20 – 80 mg/dL, the result is normal. If the units of measure are supposed to be mg/L on both instead of mg/dL as reported, this is simply a clerical error because the result will still be normal if you change the units on measure on both the result and the reference interval. Unfortunately, it may still result in a corrected report on each patient, which depending on your system, may still flag your ordering physician. If you make this correction on a high volume test, a lot of irate physicians will be calling you regarding the number of corrected reports on their patients. Thus the error causes more headache than harm, but still requires that a procedure be in place to investigate and deal with it. All labs should have such a procedure.



-Patti Jones PhD, DABCC, FACB, is the Clinical Director of the Chemistry and Metabolic Disease Laboratories at Children’s Medical Center in Dallas, TX and a Professor of Pathology at University of Texas Southwestern Medical Center in Dallas.