The Need for Lean Healthcare
Here is another sad example of the need for lean methods for quality improvement in healthcare. The story here is about a pregnant woman who died after being adminstered too much of a drug, magnesium sulfate, during labor. If you have a loved one who is pregnant, please read this and be on the lookout for this mistake.
“One person's error killed Elisha Crews Bryant, hospital officials said last week: a miscalculation by a nurse that overdosed the pregnant 18-year-old with a drug meant to slow her labor.
But the drug that killed Bryant, magnesium sulfate, is a known hazard. At least 52 overdoses have occurred in recent years, including seven cases in which the patient died or remains in a persistent vegetative state, according to a widely cited nursing journal article.”
One common theme I've read about in my study of lean healthcare is that “good doctors” (or in this case, a “good nurse”) can too-easily make a deadly mistake in the middle of a poorly designed process. This was a key point in the excellent book, Wall of Silence.
“It's a reminder that the most careful, vigilant, knowledgeable practitioner can make errors,” nurse Judy Smetzer, vice president of the institute, said of Bryant's death.
It's cruel to call this the mistake of “one person.” Yes, it was human error. The hospital, and I'm sure the nurse, would admit as much. But, the fact that at least 52 other cases like this have been reported (with 7 fatalities) in recent years should tell us that it's a systemic problem, or at least a problem that requires a systemic solution. 52 other nurses made that same mistake. Was the nurse the ONLY one involved in the health care system? I think not.
Telling people to “be careful” isn't enough to assure quality, in a factory or a hospital. All of the warnings and cautions in the world can be sent out to hospitals, but that won't fix the problem.
The hospital was, thankfully, trying to do more than that, taking measures including:
- Safety checks at each step of the process
- Double-checking the dose
- Labeling IV tubes
- Monitoring the patient closely afterward
But, the methods didn't prevent this error and this fatality.
Monitoring the patient AFTER the wrong amount of drug is given is hardly an error proofing step. Inspecting the drugs isn't always effective, the same way you can't “inspect quality into” a car. Quality means it has to be done right the first time. It means true error-PROOFING, not error-detection. In the Bryant case, they discovered the overdose had been given, but the antidote only helped save the baby, not the mother.
More prevention attempts:
- Tampa General Hospital uses a smart pump that warns if too much magnesium sulfate is dispensed.
- Bayfront Medical Center requires two nurses check the drug and label the IV tubes.
This “smart pump” (at another hospital) is a nice (if not expensive) error proofing tool. It would probably be cheaper to put a true error-proofing method in place before this pump could detect the error.
Having TWO nurses inspect something causes me to worry. It's well known (Deming taught us this) that when TWO people are responsible for inspecting something, it's far too easy for each person to relax, knowing the OTHER person will catch the problem. I've heard people say that having TWO people inspect something is just as good as NOBODY inspecting something.
Some factors raise questions about hospital procedures. Bryant's lawyer has said she was given a 40-gram solution, while the safety group recommends a 20-gram solution.
To be fair, are multiple potential “root causes” as far as their invesigation goes. I don't mean to jump to a conclusion without data, but let's explore this hypothetical.
I am just guessing, but I'm wondering if the nurse (or the pharmacy) was responsible for making a bagged (or injectable) solution just for this patient? Sometimes, hospitals “save money” by buying drugs or chemicals in bulk and then they re-package them into smaller quantities for use. Many hospitals are moving away from this because of A) the extra labor involved and B) the risk for errors or mistakes.
See, hospital purchasing might be as silo-ed and as disfunctional as purchasing in a manufacturing company. Hooray, you got a slightly cheaper price, but increased total cost (and increased the risk of death).
Again, I'm just guessing, because I've been in hospitals where the drug re-packaging “production” process scared me to death. There were too many opportunities for error.
If a 20 mg solution was the recommended dose, either they need to purchase it that way or do something to prevent a stronger solution from being formulated.
Or, prevent using this drug altogether. That would be error-proofing, if there are safer alternatives:
Magnesium sulfate is one of several drugs commonly used to slow preterm labor contractions. All the drugs have only limited impact, and many have potentially life-threatening side effects, says the American College of Obstetricians and Gynecologists. As a result, the group says there's no clear choice of which drug to use.
The hospital's solution is to “increase supervision.” The right approach is stated by the Institute for Healthcare Improvement:
“What we understand more often is where people like this nurse (make a mistake), they're not supported appropriately, they don't have the right systems in place,” he said.
Wow, this is sad. It's horribly frustrating to see preventable errors occur in healthcare. Don't let this happen to a loved one.