Wednesday, July 25, 2012

A simple mistake?

If you pay attention at all to news about medicine, you must have heard about the tragic death of a 12-year-old boy who was seen and discharged from the emergency department of the prestigious New York University Medical Center. He had developed sepsis from what appears to have been a small cut and the diagnosis was missed. When he eventually returned to the ED and was admitted, it was too late and he died. Read more details here.

Many have commented on this sad story. The good people at ProPublica, some of whom I follow on Twitter, wrote about this and other types of medical errors in a piece entitled “Why Can’t Medicine Seem to Fix Simple Mistakes?” I will grant that some of the errors they mentioned, such as wrong-site surgery and reusing syringes, are indeed simple mistakes and should be 100% preventable.

But the death of young Rory Staunton, admittedly caused by medical errors, was not the result of “a simple mistake.” The ProPublica story says, “The hospital's emergency room sent Rory Staunton home in March and then failed to notify his doctor or family of lab results showing he was suffering from a raging infection.” Specifically, he had a number (said to have been five times the normal value) of immature white blood cells called “bands,” which is a sign of the body reacting to an infection.

While human errors were made, this event also fits James Reason’s “Swiss cheese model” of a complex series of occurrences that will take more than a new policy about notifying ED MDs about abnormal lab results to fix.

Since others who have written about this case also did not have all of the facts about it at their disposal, I won’t let that stop me either. Based on what I have read, I will point out the many areas where things may have gone wrong and why the outcome was not the result of a simple mistake. Disclaimer: I am not in any way trying to defend the actions of any of the doctors or hospital staff.

There is great pressure brought to bear on ED MDs to increase “throughput.” EDs are very crowded these days and beds do not even have chance to cool down between patients. I imagine the “move ‘em out” mentality played a role in what at first appeared to be a simple flu-like illness in a young boy.

Some have attempted to defend the ED MD by pointed out that a child’s heart rate is much more rapid at baseline that that of an adult or that fever increases heart rate by 10 beats for every degree of elevation, which is not a consistent finding. The problem with that line of reasoning is that this 12-year-old boy was 5 feet 9 inches tall and 169 pounds. A heart rate of 140 is not normal. The ED was seeing some cases of the flu. This was a child with a fever, an upset stomach and dehydration. But in addition to his abnormal vital signs, he also could not walk and had leg pain, symptoms which are not usually associated with the flu. It seems that the history of mottled skin was not picked up by the ED MD.

The private pediatrician laments the fact that she was not given the results of the boy’s lab tests. The original article quotes her, “I never knew that testing was done.” She also said “I sent him to a major medical center.” She is on the staff of the hospital. Did she call the ED and tell anyone she was sending in a sick child with “mottled skin”? Does she have any responsibility to follow up on the referral of a patient of hers to the ED? Could she have accessed his ED chart and/or lab work via the electronic medical record?

Apparently the lab results were not printed until 3 hours after the child was discharged from the ED. 3 hours? It takes an automated CBC machine 5 minutes to run the test. The differential would have taken maybe 20 minutes. The result should have been available within 30 minutes. And printed? Who is still printing lab results? Pieces of paper get lost. Is it possible that NYU is still using a paper chart?

My rule has always been “If you order a test, it’s your responsibility to find out the result.” I guess that is not the way it went at NYU that day. I wonder if a shift change and sign out was involved. Of course, the patient would have been home already, but most EDs have the capacity to call a patient back if an overlooked or unexpected finding occurs.

After the discharge from the ED order was written, vital signs that were still ominous were again taken. No nurse or patient care technician thought to report those findings to the ED MD.

I feel very sorry for the family of the patient but I don’t blame them. They had faith in the doctors but it was misplaced. One must aggressively advocate for one’s loved ones. If the advice doesn’t seem right, ask questions. Get a second opinion.

A physician from another hospital suggested what may have happened. He said, “The big questions are about how to integrate new information that doesn’t fit with the perception you have formed. How to listen to the patient when they are telling you something that doesn’t fit with your internal narrative of the case. These are the hardest things to do in medicine and yet the most important.”

From a follow-up article in the NY Times: In a statement, the hospital said that emergency physicians and nurses would be “immediately notified of certain lab results suggestive of serious infection, such as elevated band counts.” The hospital has developed a new checklist to ensure that a doctor and nurse have conducted “a final review of all critical lab results and patient vital signs” before a patient leaves.

My concerns about the supposed remedy for this are what level of elevated bands, a very common finding in any type of infection, is going to prompt a call? The normal value for bands in a differential blood count is 3-5%. What will be the threshold? Will the ED be overloaded with false alarms?

NYU’s ED sees over 48,000 patient visits per year and its major affiliate and next door neighbor Bellevue Hospitals sees 89,000. Does anyone really think that a checklist is going to be filled out for every single ED visit at these two facilities?

Bottom line: The death of Rory Staunton was not the result of a simple mistake. A new policy and a checklist cannot guarantee that a similar tragedy will not occur. Although some system issues appear to have been in play here, human errors were made and cannot be totally obviated by any policy.


RobertL39 said...

In addition to the 'multiple holes in the cheese' that you mention, maybe one of the holes is the failure of the pediatrician to do her job. Surely she knew this patient better than the ED docs and was hopefully FAR more likely to see the seriousness of the problem. If she didn't see the patient, she should have. If she did see the patient, she should have been more aggressive in following up, knowing that he was clearly not well. As you have mentioned before, the hand-off is a dangerous time. Just ask any quarterback. And I hate to mention 'incentives'(aka financial issues), but I will anyway; if she were penalized by her ACO for sending one of her patients to the ED during office hours, would this still have happened? Probably not.

Todd J. Scarbrough, M.D. said...

In ~2000, at the age of 28, I got a kidney stone. I was a radiation oncology resident at the time. In March of 2001, still a resident, I felt the familiar pain and problems again. I traveled 200 miles from my parents' (I was on vacation) to go back to my home institution to get cared for there (I knew the docs and I had extreme faith in their expertise).

On a Sunday night, I showed up in the academic institution ER in a classic kidney stone sort of presentation. I was given fluids, pain meds, and discharged home. I spent all Monday feeling miserable. Not knowing what else to do, I went back the ER that Monday night. Labs were drawn, fluids were given, I had a CT abdomen done this time, and I was discharged home. Tuesday... I felt rotten. It felt like my whole right side of my body was heavy. I just felt "sick." I *again* went back to the ER, seeing the same ER doctor, given fluids, pain meds, and sent home. Tuesday night, at midnight, feeling horrible still... I logged in to the hospital system from home (I was a resident, I had access obviously). I was being told everything was normal and I just had to wait for the stone to pass.

As I checked my labs, my 'lytes were pretty screwy and I had a white count of about 18K, neutrophils very high too. "Wow" I thought. "I'm pretty sick and feel it--wonder if I should take antibiotics?" No one mentioned my abnormal labs to me. So then I thought--I wonder what my CT showed. So I pulled up my films. I had a HUGE right kidney! It was pretty significant hydropnephrosis and hydroureter (my ureter was as a thick as a ballpark frank on that side). I was told in the ER that my CT was "normal." "Hmmm... kind of septic... blown up kidney... I should call the ER doc."

So I got the ER doc on the phone. "Hey, regarding my normal labs and normal CT: I have a big hydronephrosis on the right, and my white count is pretty elevated. Can you look at my CT scans again?"

Long story short, I was taken to the OR first thing Wednesday AM for laser lithotripsy. The ER doc later told me that the resident reading CTs that night had told him my CT was normal; the ER doc himself had not looked at the films (nor would I necessarily expect him to).

So in medicine there are oft times no "simple mistakes." Even the simplest require a complex series of errors to occur sometimes. Self-advocacy and family advocacy is invaluable too. I don't know how things would've turned out if I hadn't had the capability to see my own labs and scans and understand them--AND offer corrective guidance to my caregivers.

Skeptical Scalpel said...

Robert, I agree with you comment. Thanks.

Todd, that is some story. You were fortunate to have been able to help yourself. I think ED MDs should look at films they order. Your case is proof that radiologists do make mistakes. But shouldn't the resident's reading have been checked by an attending radiologist?

Carolyn Thomas said...

Dr. Scalpel - your observations do indeed seem to confirm that this sad outcome was not merely the result of a simple mistake.

Rather, if things indeed unfolded as you indicate here, the outcome looks like the inevitable result of MANY simple mistakes, one after another after another, from miscommunication to poor hand-off protocol, misdiagnosis of troubling symptoms that were clearly not flu-related (inability to walk, etc), a 3-hour delay in lab results, and on and on.

Throw in the E.R.'s "throughput" pressures and you have a perfect storm of catastrophic proportions.

I suspect your doctor friend from another hospital may indeed have been closest to the truth: the diagnostic issue of confirmation bias and other forms of "thinking errors" are well-documented. I've read that some med schools are in fact now beginning to teach something called "metacognition" = a reflective approach to problem-solving that involves stepping back from the immediate problem to examine and reflect on the thinking process.

Skeptical Scalpel said...

Carolyn, I agree with you, especially your last paragraph. I wish more med schools would teach problem-solving. I've blogged about this before.

K_Chie said...

Dr. Scalpel, I agree with you. Too many system errors. Others that I explored in my own blog. It truly is a shame.

K_Chie said...

Wow, Dr. Scarbrough, that is truly something else.

Anonymous said...

About Dr. Scarbrough's case, I think at some places attending reads are done in the morning, so perhaps the usual rigors weren't applied by then. I agree docs should recheck their reads, I've seen things missed a thousand times. With the sheer volume of reads that the radiologists have to do, I'm sure a few out of the thousands may slip through no matter how careful someone is.

I wonder if there is a better way around it. For instance, are there computer systems that can detect and stratify reads according to suspicion? Perhaps out of naiveté, I imagine a hospital 20 years from now with an integrated EMR that can communicate intake vitals and labs to such a computer to create a clinical rating just like this. But as my girlfriend (who is a lawyer) points out, if somethings goes wrong, then who would get sued? The onus still may still end up on the physician.

Skeptical Scalpel said...

The onus is always on the physician. People have tried to create algorithms for things like this. Medicine is a little too complex so far. Maybe some day.

Anonymous said...

Hipaa now prevents you from accessing yor own record in that manner. With MU you can get it... But there's an administrative window. Would not have assisted you much to have seen your results 48, 72, hours or even days later now would it?

Skeptical Scalpel said...

Anon, I'm with you. The HIPAA law was over the top. It is one of the most misunderstood and annoying laws ever passed.

Anonymous said...

This an example of the failure to assure that electronic medical records devices' silos are not silent when new results arrive. The default is to blame users, but, the workflow is terribly disrupted as a result of EMR devices' design flaws, rendering even the most competent doctors and nurses error prone.

Skeptical Scalpel said...

I agree that the EMR may be a factor. But one of the articles about this incident mentioned the printing of the lab results. This suggests that maybe the NYU Medical Center is not up to speed with their EMR. Some EMRs flag all unread lab results, which is quite helpful unless the doctor ignores the flags.

Skin & Laser Center said...

I found your website the other day and after reading a handful of posts, thought I would say thank you for all the great content. Keep it coming! I will try to stop by here more often.

Anonymous said...

I went into an ER with a temp,high blood pressure, hr,and I was complaining of the red spots in my groin, pain in my legs,inflammation that was spreading. I knew I had a spreading infection and asked for antibiotics to clear things up but I was refused the antibiotics. The ER doctor had listed what I had as"Probably Viral". Then I see purple spots by my chest and left armpit..Well by ER visits 4 I also have Meningitis..It traveled up my spine into my head I can feel my brain swelling in my head as a response to the burning going from the right side, back of my head, left side and then the top..What happens next is the kind of thing that most people would not want to believe..An extensive cover up ensued with a doctor getting me out of the ER immediately after the spinal tap and also a CT of my head confirmed the site of infection. I become drowsy and cannot feel my limbs when I am brought home and then returned back to an ER after the spinal tap and I am denied medical help..The condition I am left in made to look like a psychiatric problem..The rest of what happened is horrific.

Skeptical Scalpel said...

Sorry. It sounds like you suffered a lot. I hope you are feeling better now.

Post a Comment