We need Lean, not more $tudies

Report: Drug errors injure more than 1.5M

There are many lean lessons and methods that can be applied to fixing pharmacies and drug dispensing. One method is the use of electronic prescriptions. Errors are still possible with electronic systems (clicking on or entering the wrong drug), but it eliminates the problem of messy handwriting and confusing one drug for another for that “failure mode.” Bar-coding and patient wristbands can help (although it's possible for a patient to get the wrong wrist-band put on their wrist).

We're human. We make errors. That's a core tenet of the Toyota Production System and it applies in healthcare also.

“…the new report highlights how the nation's fragmented health care system is conducive to drug errors, said Dr. Donald Berwick, a Harvard professor who heads the nonprofit Institute for Healthcare Improvement.

“This isn't a matter of doctors and nurses trying harder not to harm people,” Berwick cautioned. “Safety isn't automatic. It has to be designed into the system.””

That's absolutely right. We can't just tell people — doctors, nurses, phamacists, lab technicians — to just be “more careful”. As Dr. Deming would have said, the last thing we need is everyone doing their best. We need systems with designed-in error proofing and designed-in quality, rather than relying on inspection or being careful.

Being careful still counts. Gross incompetence might still defeat an error proofing device. But we still have to try so we can prevent as many errors as possible.

I don't know if I would agree with all of the report's recommendations. Investing in technology is one approach. But, how about instead of investing $100m in researching drug interactions (that's a lot of money!!), how about we spend that money teaching everyone lean methods.

The DVD “Good News… How Hospitals Heal Themselves” points out how improving quality doesn't have to cost a lot of money. Using lean methods NOW will improve quality and reduce errors without wasting $100m investigating and studying the problem. Error proofing methods (and standard work) like this costs very little to implement and has a huge impact.It's frustrating how we want to study problems instead of taking little practical steps to fix them.