e_moon60 (e_moon60) wrote,

Failure Analysis and Ebola

When I was a child and knocked a glass off the table, my mother the engineer led me through one of my first failure analyses:  a consideration of why and how the glass ended up on the floor along with the milk in it, a model of how, as she said, "Accidents don't happen: they're caused."   Not always by a human (though usually) and certainly not always by the person who's hurt.  A friend of mine was killed by a tire that came off a large truck going the other way, into oncoming traffic, where it smashed into the windshield of her car and crushed her face into her brain.   Her teenage daughter, a front-seat passenger, grabbed the steering wheel and steered the car safely to the side of the road.  That was certainly not Sally's fault.   But in any situation, my mother's response was to analyse what went wrong so that a similar unwanted event would not happen again.   I had set the glass of milk on the edge of the table (easier for me)  where it could easily be knocked off the table.  I remember, in that and other occasions when I was led through the process of analysis, resenting it.   It was an accident...I hadn't meant the glass or plate to fall, I hadn't intended to drop something breakable, I hadn't planned to flunk that test.   Those responses, so natural to a child and to many adults, met firm correction from my mother:  it wasn't about what I meant to do, what I planned to do, but what I actually did.   I learned---not as well as she wished--to think ahead at least two steps, to imagine alternate possibilities, to look for ways things might fail, rather than assume they would not.   I was encouraged to consider what to do if they did--to have a Plan B, and also a Plan C, D, E.  Did I always do it?  No.  Did life itself teach me why it was a good idea?  Yes.

Which leads us to the curious affair of the Dallas hospital letting an Ebola patient leave the hospital because they didn't realize he might have Ebola, even though he told them he had come from one of the countries known to be suffering an Ebola epidemic.   This is, in epidemiological terms, a big-F Failure.  How could a major hospital in a major US city that has a very large, international hub airport, make that mistake, a mistake with potentially deadly consequences?
That's the topic here--my own analysis, based on my experiences with medical facilities in Texas (albeit not in Dallas, but down along the Border, in San Antonio, Houston, Austin, and smaller communities here and there.  I have been reading medical journals for my own enjoyment for over 40 years; I have a degree and some graduate work in biology, so I understand what I'm reading, and I"m familiar with current issues in hospital management.

The situation has been well-reported.  An apparently well individual, who passed inspection at both the departing airport and subsequent transfer airports, arrived in Dallas still apparently well...but actually incubating the disease.  This is a scenario that has been widely discussed in the medical literature (and in politics, for that matter) for years--that modern air travel makes it possible for someone incubating a dangerous disease to travel by air and arrive in a distant country before showing any signs of illness.   Diseases have already been transported that way--measles, for instance, and influenza.   In the medical journals and conferences, the possibility is known, and protocols for dealing with it have been discussed and refined.

With regard to Ebola, commentary in the main medical journals, including the possibility--even the likelihood--that a person incubating the disease might arrive in the US by air was mentioned months ago, and as the situation in West Africa got worse, as it became clear that the situation was not at all contained and the case count increased into the hundreds and then thousands, the probability rose ti near certainty.   The Centers for Disease Control (CDC) and the National Institutes of Health, due to some procedural failures in their organizations, found themselves unable to respond as quickly to the growing threat as usual, because Congress forced the shutdown of CDC's only laboratory capable of rapid testing and characterization of Ebola virus in clinical samples, and then called in staff for Congressional hearings.  Those politicians who distrust science and scientists, CDC's influence in epidemiology and NIH's in biomedical research, and who have insisted on cuts in budgets related to infectious disease, were having a fine time sneering at the only people who knew damn-all about Ebola, but...it's an election year.  And yes, I'm both realistic and cynical.   So it was almost August before CDC could get untangled from Congressional red tape and back to their real work of informing physicians and hospitals about how to respond to the Ebola situation.

Their commentary was clear; the protocol understandable to me, the non-doctor.  The initial symptoms of Ebola were nonspecific--easy to confuse with other common and less dangerous disease.  Fever--but not a spectacularly high fever.  Headache in some, abdominal pain in others.   Ebola had been mistaken in the past for upper respiratory illnesses, flu, malaria, food poisoning, a stomach virus.   Therefore, in the case of travelers from the area of the outbreak, anyone with fever and any other symptoms should be immediately isolated and tested for Ebola; all personnel in contact with the patient should wear full protective gear, and any contact with bodily fluids meant isolating the contact with surveillance, including testing, for the full incubation period.   Waiting for the characteristic symptoms would be a mistake, and if the person did not have Ebola, no harm was done by a day or two of isolation until the test results came back.

A couple of days after arrival in Dallas, the man from Liberia felt sick--feverish, headachy, abdominal pain--and went to the emergency room at Texas Health Presbyterian Hospital.   There he told the staff at intake that he had traveled recently, from Liberia.   That information was entered into a computer records.   He was not admitted; he was given antibiotics and sent back to where he was staying, because (said the hospital spokesperson--not medical personnel who had contact with him in the ER)  his condition did not require admittance and his symptoms were not specific to Ebola.  Two days later, much sicker, he returned to the same hospital, where he was admitted and placed in isolation; tests showed that he did indeed have Ebola.

Mere trickles of information about what happened at the hospital on that first visit, that missed opportunity,  have emerged: the computer record apparently edited patient travel data in such a way that the medical staff in the ER did not know he had come from Liberia.    (The patient's own statement was that he told people at the hospital several times that he had, but it's not clear if he told one person repeatedly or several persons, or whether that was in triage or was said to a medical staffer.) [ [EDITED 10/4 afternoon:  The hospital has now changed its story and says the information about travel was available to both doctors and nursing staff.  This is not what they said yesterday and the day before.  Their new statement was taken from the hospital's own website.]]  The identities of the  individual who released him, on the first visit, the person who admitted him, on the second, and the person who recognized the possibility of Ebola have not been released, so it is not possible at this time to know their qualifications and experience,  or their reasons for the decisions they made, though I'm sure this has been part of the investigation.   Nor do we know exactly how or why the computer record was modified so that it did not deliver the travel data to the medical personnel in the ER.  This leaves a lot of room for speculation, and only the hospital to discuss in some detail.

Texas Health Presbyterian Hospital, where the patient arrived a few days after his arrival in the United States, is an 898 bed acute-care religious-affiliated nonprofit hospital belonging to a regional association, Texas Health Resources, of such hospitals.   When ranked according to the typical criteria for hospitals (ER visits, ER discharges-readmissions,  outcome measures for various specific conditions, etc.)  THPH ranks about midway in its own association, and a little below the middle for all Dallas area hospitals.   It is not a teaching hospital, and ranks well below the large teaching hospitals in Dallas.   It's a middling hospital for Texas in general--many are better, many are worse.  But it's not where you'd want to go if you had a serious unusual problem, a critical illness that was being seen in that facility for the first time.

This was obvious to me even before I  looked up the hospital and its rankings.  A first-rank hospital's senior staff, especially the chief of infectious disease, would have been well aware of the progress of the Ebola outbreaks in West Africa, well aware of the hospital's chances of seeing persons from that area, and would have kept on top of current recommendations for handling a sick person who had been in those countries.  The possibility would have been discussed with other doctors, with any medical staff who might come into contact with such a person, and the ER's ability to recognize a potential Ebola case would have been assessed and improved.  Not just the ER doctors, but the nurses, the aides, and particularly those who did intake--who gathered information from the patients as they arrived.   Staff would be reminded repeatedly that anyone arriving from anywhere in West Africa with a fever should be considered a potential Ebola patient.  That travel history was vital, and the full information must get to the medical personnel evaluating the patient.  They might not have been able to get the entire nursing staff trained in the elaborate safety measures needed to care for an ward full of Ebola patients, but they would almost certainly have given extra training to the team they planned to assign to do it for one patient.

None of that happened at THPH.   Ebola was not even suspected at the first visit, despite this person's having typical early symptoms of Ebola and being from one of the outbreak countries.

But there's another clue to this failure, and it's also from the first visit ER record.  The patient had fever and reported abdominal pain, and he was given antibiotics without testing for any of the bacterial causes of abdominal pain.  The routine administration of antibiotics for abdominal pain and moderate fever is...no longer best practice. Hasn't been for decades.  Most stomach/gut upsets are viral in origin.   Unless the pattern of the illness suggests one that is bacterial--and research has shown that a specific antibiotic works against that particular bacteria, antibiotics are not given.  (Among other things, antibiotics disturb the intestinal flora, resulting in more problems.)    Overuse of antibiotics has other consequences and first-rank hospitals and clinicians know that.

So here's a hospital that is not keeping up to date on common things--the person with a fever and abdominal pain--is it any surprise that it's not keeping up on something new, exotic, unexpected.     And that brings up another question.  Why would a large hospital in a large city, a city with excellent teaching hospitals, not want to catch up, to be better, to be, in fact, excellent?  A variety of answers are possible.   Perhaps they believed their own website and thought they already were excellent.   Perhaps they felt constrained by financial concerns and skimped on education and training because it would cost more.  Perhaps they did not believe anyone with Ebola would ever show up in their ER--surely such a person would go to one of the teaching hospitals or a public hospital. Perhaps they felt politically motivated resistance to instructions coming from CDC.   It's important to find out what their motivation for lack of preparedness was.

And when the analysis is finished--when it's known exactly who made which decisions that led to the failure to recognize the possibility of Ebola on that first visit--it's then time to figure out how to prevent another such failure.   How to motivate all hospitals to make the changes necessary, to take time and money to train nurses and other staff.  How to compensate hospitals for the costs involved in treatment.   It is doable.   It's been done elsewhere, with fewer resources than we have.   But to accomplish it--to make every hospital in the nation capable of recognizing and safely caring for an Ebola patient--we must pursue every failure even if it bruises some feelings.

Tags: ebola, failure analysis
  • Post a new comment


    default userpic

    Your reply will be screened

    Your IP address will be recorded 

    When you submit the form an invisible reCAPTCHA check will be performed.
    You must follow the Privacy Policy and Google Terms of use.