Corrective action is a beautiful thing. When quality control is out of limits, running a fresh control that is acceptable saves time, avoids an irritating sample look back, and reassures the tech. When a weak blood bank reaction in all tubes disappears by recollecting a sample and letting it clot in the refrigerator, a simple fix is found. And when an instrument problem is identified and repaired before a service call, everyone is happy.
Corrective action should always be this easy, but I’ve seen QC problems ignored, instruments run despite problems, refrigerator temperatures ignored, expired reagents and controls run, and out of control values repeated multiple times until they just squeak in. I suspect this is common in all labs to some degree.
One solution is for any laboratory instrument to lock out the operator if conditions aren’t met, similar to bedside glucose meters. This is possible with information systems, too. I don’t know why that doesn’t happen. In the CPSI system, for example, one can post unacceptable quality control and still report patients. That seems to me a trivial software engineering problem.
But the problem is not just engineering. We all naturally take the shortest route to the finish, which explains why nurses will hoist a person in bed and get a back injury instead of using a lifting device, why people arrive at work in the winter in high heels and walk down an icy hill around a dumpster, and why lab techs repeat a control three times until it is within limits. Most of the time, this kind of behavior doesn’t cause a problem.
“Near miss” behavior is the darling of quality geeks, who think if we can eliminate it errors will never happen. There won’t be lifting injuries, there won’t be falls, patient results will be improved, and we’ll all save money. Of course, these are the same people wearing the heels, eating the fast food, typing with the keyboard on top of a bookcase, etc. We are all carelessly expeditious.
What we need is corrective action to fix corrective action. Any ideas?
NEXT: We All Want Simple Solutions