This is probably a record length post for me, but I thought it was important to respond to Mr. Dwyer’s comments to a post written on this blog regarding the article he wrote that appears in the NY Times.
I had planned to leave my comments after his, but they became too long and involved and I also wanted to paste a couple of pictures from Mr. Dwyer’s article, so I instead decided to answer his criticisms in a post.
If any of you were wondering, I was not the anonymous physician who authored the previous post on Mr. Dwyer’s article. I spent most of my afternoon creating this response because Mr. Dwyer’s original article was somewhat frustrating to me, but I found his justifications and explanations for what was contained in his article to be misleading.
Dear Mr. Dwyer,
When re-reading your article, I absolutely agree with Rory’s wish that no other child – and no other family for that matter – should have to go through what Rory went through. He sounded like a great kid and he obviously had a close family and a bright future. As you also mentioned, Rory’s uncle was a friend of yours, so I can imagine that this incident affected you more than most other investigations you have performed. This topic hit home for me as well. My daughter nearly died from an invasive pneumococcal infection when she was younger. She was hospitalized for a week in a university medical center on triple antibiotics. Very scary times and we thank God that things turned out well.
So let’s go through your article and responses you made to the criticism about your article so that we can determine how to prevent kids from dying from sepsis due to invasive organisms.
JIM DWYER COMMENT:
1. You say that the stop sepsis campaign is for tracking severe sepsis. That misstates both the nature of the campaign and my citation of it in the article. The campaign’s goal is to aggressively identify sepsis and begin treatment within an hour. (The tracking of cases you cite is secondary.) To begin the process of identification, the initiative created a triage screening tool which gives a list of 8 signs and calls for additional investigation if a patient has three of them. As I wrote, Rory Staunton had two when he came into the ER. He had three when he was leaving. (BTW — his heart rate over a period of two hours ranged from 131 to 143. That’s in the article, too.) In the distribution literature with the screening tool, there is no distinction between pediatric and adult patients. Whether or not you think the values are relevant to a 12 year old, 5’9″, 169 lb boy, Rory was assessed for possible sepsis in triage.
Let’s look at the sepsis criteria according to the checklist that you posted. Then let’s apply them to children.
1. Pulse greater than 90. In children up to 2 years of age, a pulse rate less than 90 is considered too slow. In other words, ALL children up to 2 years of age should have a pulse rate greater than 90.
2. Respiratory rate greater than 20. In children up to 5 years of age, a respiratory rate less than 20 is considered too slow. In other words, ALL children up to 5 years of age should have a respiratory rate greater than 20.
So now in children who have entirely normal vital signs for their age, right away you have two of the three “danger signs” your article repeatedly emphasizes.
Add a temperature of 100.5 degrees which is essentially not a temperature at all.
Let’s give the child a runny nose which causes us to suspect a viral URI – the suspicion of an “infection” required by the criteria.
This two year old child, running around the room and laughing with his parents, with essentially normal vital signs for age, now has 4 of the criteria on the screening tool you cite.
According to the premise of your article, we must rigidly follow the criteria on the “screening tool,” which means that on every such child, doctors should get a mandatory serum lactate level, order immediate IV antibiotics, and hospitalize the patient. Heck, we should probably throw in a central line and urinary catheter as well to monitor central venous pressure and input/output.
Can you even begin to imagine all of the unnecessary added expense and adverse reactions from the antibiotics/invasive monitoring that would occur if every medical center in the country adopted Jim Dwyer’s rules of pediatric management? Every influenza season, there would be no hospital beds available for months as hospitals were forced to overtreat healthy well-appearing children while delays for care of other emergent patients precipitously increased.
The problem with your article, and something that you conveniently hid from your readers, was the disclaimer at the bottom of your so colorfully highlighted checklist
Doesn’t the disclaimer at the bottom of the checklist say something to the effect that it “should not be used as a substitute for clinical judgment”? Can’t really see the whole sentence because your placement of Rory’s labs just happens to obscure the rest of the wording. But I’ve read enough checklists and disclaimers to know that the disclaimer most likely states that the checklist should not be substituted for a physician’s clinical judgment.
Yet, despite the checklist specifically telling you NOT to do so, that’s exactly what you did, isn’t it, Mr. Dwyer? You published an article asserting that regardless of the clinical judgment of a physician who has many years of training in medicine and who is described in your article as being “hyper-conscientious”, this protocol must be rigidly followed. You misused this guideline in order to inappropriately attack the qualifications of physicians you never met and to whom you never even spoke.
You state that the guideline and its literature made “no distinction between pediatric and adult patients,” yet you didn’t even know enough or didn’t care enough to ask what patient populations the guidelines were created for.
You keep asserting that Rory was the size of an adult. Fine. I agree. But he was still 12 years old. Unless you have evidence that the criteria have been validated in children – even adult-sized children – don’t assert that the criteria are valid in children. You know darn well that if the situation was different, the medical treatment involved medications not approved in “children,” and Rory died after receiving the medications, you’d be the first one writing about “warnings ignored” in giving the medication. 20/20 hindsight is just crystal clear.
JIM DWYER COMMENT:
2. He didn’t have lactate levels done when he came in because, apparently, he only had two signs on the triage screening tool. Of course, it didn’t matter, since even the labs that were done were not used in the assessment of him.
No, actually Rory probably didn’t have lactate levels done because they weren’t indicated. This is part of the problem with your article and with your argument. You have little or no idea how medicine is practiced, then you create fact patterns which make little prospective clinical sense in an attempt to justify your position. Again, knowing that a child died from sepsis, of course the conclusion (completely tainted by hindsight bias) is that a lactate level should have been ordered.
What if the lactate level was normal? Would you have discounted that like you did the normal strep swab and his essentially normal serum chemistries?
JIM DWYER COMMENT:
3. Blood differentials showed the 53% bands noted in the blog post and the article. They were not part of his assessment, even if your anonymous author believes they could appropriately be folded into the belly-ache diagnosis, without any indication that the treating doctor actually saw them.
Here, we agree. A band count of 53% is concerning and may very well have affected Rory’s care. The problem with your article is that it doesn’t focus on the real problem with the elevated band count. According to your article, labs were drawn on Rory when he arrived in the hospital. We don’t know whether the doctors even knew they had been ordered and I’m sure you didn’t ask. But a CBC which takes less than 15 minutes to run took more than 5 hours to be reported. That’s a system problem that needs to be addressed. Who is at fault? We don’t know and you can’t say. Follow up of abnormal labs is another system function that needs to be addressed. I’m sure that NYU had some system in place for following up abnormal labs. The question becomes where the system broke down. No one can say whether Rory would have survived if these breakdowns were not present. His chances of survival most likely would have been greater, but no one could predict the outcome.
JIM DWYER COMMENT:
3. [sic] Your anonymous doctor cites literature stating that a temperature up to 106.7 is not harmful, and takes me to task for reporting that the boy’s temperature at home, 104, was his highest. I gather this should not have been mentioned because it would unnecessarily alarm people, in the view of Anonymous. In this case, the parents were correct that his fever was a sign of something dangerous happening. (Later on, after his temperature had dropped, it rose two degrees following his receiving fluids and tylenol in the ER.)
Regarding the 106 citation, here is just one article that studied more than 130,000 pediatric patient visits. 103 of those 130,000 had fever >106 (hyperpyrexia) and of those 103, 20 had a serious bacterial illness, and only 11 had bacteremia – similar to but likely not as severe as the “sepsis” described in your article. If you look through the types of patients who had bacteremia, almost all of them had urinary tract infections as the source of their infection.
It would have been helpful if your article cited literature stating that a temperature of 104 degrees that drops and rises two degrees after “receiving fluids and tylenol [sic]” is related to “something dangerous happening.” Unfortunately, you won’t find such an article.
However, here are a few “Myths” and “Facts” about fever “in the view of” not Anonymous, but the American Academy of Pediatrics:
MYTH: If the fever doesn’t come down (if you can’t “break the fever”), the cause is serious.
FACT: Fevers that don’t respond to fever medicine can be caused by viruses or bacteria. It doesn’t relate to the seriousness of the infection.
MYTH: Once the fever comes down with medicines, it should stay down.
FACT: The fever will normally last for 2 or 3 days with most viral infections. Therefore, when the fever medicine wears off, the fever will return and need to be treated again. The fever will go away and not return once your child’s body overpowers the virus (usually by the fourth day).
MYTH: If the fever is high, the cause is serious.
FACT: If the fever is high, the cause may or may not be serious. If your child looks very sick, the cause is more likely to be serious.
MYTH: The exact number of the temperature is very important.
FACT: How your child looks is what’s important, not the exact temperature.
Medical recommendations for evaluation of a febrile immunized child more than 2 years old stop at “obtaining a history and performing a physical examination,” while “specific workup and/or treatment is based on the clinical findings and suspicion of disease.” This same article also notes that, in evaluation of febrile children, “several factors indicate an increased risk of bacteremia and/or sepsis, including age less than 2 months, an immunocompromised state (eg, neutropenic or underlying malignancy), being unvaccinated or undervaccinated, hypothermia (core temperature < 36.8°C, or 98°F), and hyperthermia (core temperature >40.5°C, or 105°F). Rory had none of these – at least according to your article.
Your article perpetuates the myth that “fever is the enemy” when, in fact, very few children with fevers have any serious bacterial illness, much less the sepsis from which Rory died.
JIM DWYER COMMENT:
4. Your blogger writes, “The article implies that the white blood cell count of 14.7 should have triggered a more aggressive workup. There is a mountain of evidence to say that a high WBC count does not rule in or rule out an infection or severe sepsis.”
I wrote no such thing about the WBC, nor did I imply it. Many experts I consulted told me about the non-specific nature of an elevated WBC. In actual fact, I wrote that the lab found highly abnormal levels of bands and neutrophils.
Here is what I wrote concerning his labs:
“About three hours later, Rory’s lab results were printed. He was producing neutrophils and bands, white blood cells, at rates that were ‘very abnormal and would suggest a serious bacterial infection,’ Dr. Edmond said.”
Here, you’re again showing your unfamiliarity with the topics you are writing about. Neutrophils and bands are white blood cells. Quoting another person who stated that the rate of production of these WBCs was “very abnormal and would suggest a serious bacterial infection” is absolutely implying what the other blogger stated.
JIM DWYER COMMENT:
5. This is a narrative about important information that was not used — whether due to oversight, unavailability, or lack of understanding — when it might have made a difference in the life of the patient. I don’t blame any one or any place for this, because I don’t know how things went wrong — just that they did, somewhere along the line as Rory Staunton moved from pediatrician to emergency room to emergency room doctor to discharge. The names of the doctors and the institution, as well as the name of the patient, are germane to the narrative of his treatment for viral gastroenteritis, when he actually had a Group A strep infection in his blood.
No, this SHOULD have been a narrative about important information not used. You could have even made it educational by informing your readers about the signs that a fever may represent a more serious disease. Instead, it came across to me (and to a good proportion of the 1600+ people who commented on your article at the NY Times web site) as a story about how some negligent doctors and a negligent hospital missed all this information clearly demonstrating that Rory was septic. These aspersions were cast based on the story of your friend – Rory’s uncle – and your friend’s family – Rory’s parents. While creating your story you knew or should have known that there was no way that the doctors would be able to respond with their side of the story to defend themselves.
How are the names of the doctors germane to Rory’s treatment? Did you name the nurses that treated Rory? Are they less “germane”? How about the lab personnel who analyzed his blood? Those people would seem really germane to me. What about the administrators who hadn’t instituted the policies that you think every hospital should have? You were out for revenge on these doctors and you got it.
JIM DWYER COMMENT:
6. NYU has announced changes to address the issue of vital signs being reviewed before the patient leaves the ER, as well as clinically relevant lab results. In the event a patient is gone when they come back, they are creating a procedure to contact the patient or the referring doctor. These things clearly went wrong in the Staunton case. That’s what the article is about. It’s journalism. I’ll sign my name to it.
NYU has “announced changes”. Congratulations. When an inflammatory article about their emergency department has been published to millions of people, what do you expect any hospital to do? Of course they’re going to “announce changes.” They want to get out of the spotlight. Show me an example of a hospital that has received national attention for a bad patient outcome where the hospital didn’t “announce changes.” Did you expect NYU to do something different?
It’s also kind of underhanded that you repeatedly note how the doctors did not respond to a request for comments. Your article stated that one of the doctors “could not discuss the case,” and that another “could not be reached for comment through the hospital,” making it appear that they were trying to hide something. You know or should know that discussing a patient’s medical care with a reporter under these circumstances without consent of patient or guardian is a violation of federal law. Here is a summary of the “permissible disclosures” under HIPAA laws. Here are the full regs if you want to read them.
Instead of saying “I as a ‘journalist’ didn’t get written consent from Rory’s parents so that their doctors could speak with me and all other journalists about the case,” you make more insinuations to portray the doctors as trying to hide their involvement in the case. Did you even ask NYU about hospital policies for speaking to media? Of course not. Stating in your article that NYU has a policy that no employees may speak to the media wouldn’t villainize the physicians as much.
To me, you inappropriately used your article as a weapon to attack the medical providers in this case whom you believe caused Rory’s death. If that’s your definition of good journalism, go ahead and sign your name to it. Anyone who does an internet search for your name will be able to review your “signed” article (contained here as a .pdf just in case the NY Times decides to remove it or make it unavailable to the public), this discussion, and then make their own decision about your motives and your appropriateness as a journalist.
At least you got a chance to respond to the issues that were raised here about your professionalism – a courtesy that wasn’t available to the doctors that you flamed.
We will never know if Rory’s life could have been saved had his underlying sepsis been discovered earlier. Sepsis is a horrible disease and many patients who develop sepsis ultimately die. Despite my disagreements with the way you presented your article, my thoughts and prayers go out to Rory’s family. I hope that God gives them the strength to heal through what I know have been, and will continue to be, difficult times.
P.S. Please provide us with the name of the editor that approved your article. I think that would be very germane to this issue.