The Weakest Link

Aviation safety experts will tell you that the weakest link in any system is the human component. They spend their days devising new ways to remove reliance on humans beings performing error-free tasks. They know that there is no such person. All humans make mistakes.

The question is why do we rely on humans to behave in a 100% error-free fashion in the vastly more complex area of medicine?

Last month, a nurse is reported to have administered a drug at ten times the prescribed dose, killing the patient, 60-year-old Shirley Curtis. The nurse has been stood down. I am hoping she has been stood down because she is too upset to work, rather than being suspended because she made an error. Because the fault for this drug overdose does not lie with her error, but in the archaic systems that allowed her to make it. Nevertheless, criticism will solely revolve around the personnel involved – the nurse herself, the nurse who checked the dose with her and, possibly, the doctor who wrote the prescription.

But this was not some callow nurse just out of university, this was an experienced nurse. One who had probably dispensed this drug hundreds of times. And yet she still made a fatal error.

Surely someone might ask why?

The MacDoctor can tell you right away that the very experience of this nurse counted against her. Her error is a common one of familiarity. The drug in question, metoprolol, is normally given in slow release format, in doses similar to the final dose she gave. Mrs MacDoctor actually takes 190mg of this drug, so a dose of 125mg (125ml of the 1mg/ml syrup) is not unreasonable. Except that the syrup is NOT slow-release, so the entire dose is released into the blood stream rapidly, instead of in dribs and drabs. The nurse would have dispensed hundreds of doses of the controlled release form, but far fewer doses of the syrup form.

Combine this with an unusually small dose of the drug being prescribed and you have the setting for a fatal error.

Dr David Galler from the Health Quality and Safety Commission thinks he has a solution:

“Dr Galler said the new National Medication Chart which is being rolled out at DHBs across the country will make mistakes far less common.”

While this might be generally true, this particular error would probably not have been avoided. There is no evidence that the nurse had problems reading the prescription. Any experienced nurse would have immediately rung the doctor if she had been unable to read the prescription clearly. It was also not a “decimal point error” which is a calculation error, rather than a legibility error. This was a pure perception error. The nurse expected a prescription with a dosage of at least 100mg (the standard dose of metoprolol) and so saw exactly that. It is possible, but by no means certain, that a large pre-formatted decimal point might have helped. But it also may not have. Paper-based forms always rely on the human element (doctors and nurses) to perform flawlessly, every time. This is a simply unrealistic expectation.

What is actually required here is a system where error is actually impossible, or, at least, so difficult that one has to make a deliberate effort to cause an error.

This technology is already available to us. Order-entry dispensing systems have been around for sometime and are now virtually bullet-proof. There are systems that make it impossible to give intravenous (into the vein) drugs into a spinal tap (a common error). Pre-filled syringes and dial up auto-dispensing syringes stop injectable drug errors. Drug dispensing systems such as PixSys, PillPick, MedDispense and the like make it extremely difficult to give the wrong tablet or the wrong dose. Software warns doctors of drug interactions, contraindications and allergies right at the moment of typing in the prescription. RFIDs and barcoding make “wrong patient” errors almost non-existent.

All of these solutions cost money, of course. However, all the current research shows that the ROIs on these systems are very short as they stop stock wastage, overprescription and drug errors. It is time that we stopped relying on the flawless operation of the weakest link in our hospital safety systems – our staff. People like Shirley Curtis deserve better. And so does the unfortunate nurse whose error ended Mrs. Curtis’ life.

Related posts: