Thursday, October 02, 2014

Did Electronic Medical Record-mediated problems contribute to or cause the current Dallas Ebola scare?

[10/3/2014 Note:  my suspicions were correct - see addenda below.]

This story merits special scrutiny in relation to EHR dangers (see for instance http://hcrenewal.blogspot.com/2014/04/fda-on-health-it-risk-reckless-or.html):

http://bigstory.ap.org/article/44a12c35649f4f6782fcb3c9f476da09/ebola-case-stokes-concerns-liberians-dallas

Oct. 1, 2014:  Ebola-infected passenger was sent home from ER

DALLAS (AP) — The airline passenger who brought Ebola into the U.S. initially went to a Dallas emergency room last week but was sent home, despite telling a nurse that he had been in disease-ravaged West Africa, the hospital said Wednesday in a disclosure that showed how easily an infection could be missed.

The decision by Texas Health Presbyterian Hospital to release the patient, who had recently arrived from Liberia, could have put others at risk of exposure to Ebola before the man went back to the ER a couple of days later, when his condition worsened.

... The patient explained to a nurse last Thursday that he was visiting the U.S. from Africa, but that information was not widely shared, said Dr. Mark Lester, who works for the hospital's parent company.

"Regretfully, that information was not fully communicated" throughout the medical team, Lester said. Instead, the man was diagnosed with a low-risk infection and sent home.

How could a "failure to fully communicate" information about the man's travels have occurred, at a time when the Ebola issue has been prominent in the press and is causing a worldwide scare?

The Texas Health System hospitals, including Texas Health Presbyterian Hospital Dallas, are EHR users and have been recipients of millions of dollars of federal incentives:

http://www.bizjournals.com/dallas/news/2011/05/24/texas-health-gets-19m-for-ehr-upgrades.html?page=all

May 24, 2011: Texas Health Resources has received more than $19.5 million in Medicare incentive payments for meeting “meaningful use” criteria for installation and use of its electronic health records.

It appears these hospitals are using EPIC. From a July 6, 2009 HIStalk blog interview with the CMIO:

http://histalk2.com/2009/07/06/histalk-interviews-ferdinand-velasco-md-chief-medical-information-officer-texas-health-resources/

What are the most important projects you are working on at Texas Health Resources?
The project is our EHR deployment. We’re an Epic customer. We’re in the middle of deploying the EHR. We’re live now with probably close to three-quarters of our beds. We’re a 14-hospital health system.

EHR's are known to disrupt normal, even mundane channels of medical communication (my mother is dead thanks, in part, to this problem).

Per the Agency for Healthcare Research and Quality (AHRQ) of HHS, possibly relevant modes of disruption in this scenario include:

  • "information hard to find", 
  • "suboptimal support of teamwork (situational awareness)", 
  • "confusing information display", 
  • "design contributed to entry into wrong patient's record", 
  • "lost data", 
  • "excessive workload (including cognitive)" ...

... and other potentially relevant factors (including system outage).

Below is a checklist of such failure modes from the May 2012 AHRQ Health IT Hazard Manager Report (http://healthit.ahrq.gov/sites/default/files/docs/citation/HealthITHazardManagerFinalReport.pdf):


 
AHRQ Health IT Hazard Manager Report - Hazard Modes of Health IT (click to enlarge)


I've worked in ED's and find this matter puzzling.  While I have no evidence as to any role of EHRs in this seemingly strange, cavalier and incomprehensible medical decision to send this man home, resulting in potential exposure of numerous other individuals to Ebola (and I am certainly not in a position to have such evidence), I believe this possibility needs to be investigated fully.

Yet there does not appear to be a rush for investigation of what transpired in the ED.  See "Health regulators not rushing to probe Dallas hospital’s handling of Ebola patient", Miles Moffeit, Dallas News, Oct. 1, 2014 at http://watchdogblog.dallasnews.com/2014/10/health-regulators-not-rushing-into-probe-of-dallas-hospitals-handling-of-ebola-patient.html/.

Finally, by the way, after all the tens of millions of dollars spent by this organization on EHR's at taxpayer expense and with the awards and accolades heaped on them by the likes of HIMSS, Leapfrog etc., at the very least one might have expected a blatantly obvious case like this to have been recognized as a serious matter. Yet it was not. (One can only imagine what happens with more subtle issues.)

One wonders how much in the real world, as opposed to in the world of EHR marketing and hype, these systems really do help in critical decision making and safety.

-- SS

10/3/2014 Update:

My suspicions were apparently correct.

See:

 "Travel Information Wasn't Communicated In Dallas Ebola Case Due To Electronic Health Record Flaw" (Huffington Post),
http://www.huffingtonpost.com/2014/10/02/dallas-ebola-patient-hospital-error-electronic-health-record-flaw_n_5924698.html

and:

"Dallas hospital blames ‘flaw’ in ‘workflow’ for release of Ebola patient as a more complete picture of his travels emerges" (Washington Post), http://www.washingtonpost.com/news/morning-mix/wp/2014/10/02/dallas-hospital-blames-flaw-in-electronic-record-keeping-system-for-release-of-ebola-patient/

-- SS

More:
  
http://www.wptz.com/health/urgent-ebola-texas-hospital-flaw/28381038

(CNN) -- The Texas hospital treating the first person diagnosed with Ebola on American soil says a "flaw" in its electronic health records prevented doctors from seeing the patient's travel history. Patient Thomas Eric Duncan told the nurse he'd been in Africa, but that information was entered into a document that isn't automatically visible to physicians [apparently even after being filled out with positive information, I note - ed.], Texas Health Presbyterian Hospital Dallas said in a statement Thursday. After discovering this, the hospital says it has changed the system so doctors and nurses will see travel history documentation. "We feel that this change will improve the early identification of patients who may be at risk for communicable diseases, including Ebola," the hospital said.  

Well, yes it will.  Perhaps that could have been thought of sooner?

Additional thought:  this situation might end up being the "Cybernetic Libby Zion case" I've been predicting - where some major debacle leads to serious attention to EHR safety issues.  (On Libby Zion, see http://en.wikipedia.org/wiki/Libby_Zion_Law.)

-- SS

10/4/2014 addendum:

Now...gee whiz...the hospital changes its tune. "No, it wasn't the EMR after all!" 

See "Hospital reverses explanation for fumbling Ebola case" at http://www.dallasnews.com/news/metro/20141003-hospital-reverses-explanation-for-fumbling-ebola-case.ece. The reversal strains credibility and sounds like redirection, to my ear possibly due to inside attorney and/or EMR company attorney pressure. 

The "new explanation" itself per the new article is that:

... A written statement Thursday said hospital officials identified and corrected “a flaw in the way the physician and nursing portions of our electronic health records (EHR) interacted in this specific case.” That statement implied, without directly saying it, that the flaw left the doctor uninformed about Duncan’s travel history. In Friday’s statement, though, the hospital said, “The patient’s travel history was documented and available to the full care team in the electronic health record.” “There was no flaw in the EHR in the way the physician and nursing portions interacted related to this event,” the statement said.

Again, sounds like redirection and making the doctor (and perhaps the ED doctor's group, if they were contractors) the sole scapegoat.   

"Available to the full care team?"  "Available" in a complex computer system with myriad screens is a very relative term.  The issue seems not "how the physician and nursing portion interacted", it is "how the physician portion made the information readily apparent to the physicians and other team members, or not."
 

The problem here, I believe, still likely amounts to "information hard to find" and "suboptimal support of teamwork (situational awareness)", among others, per the AHRQ hazards taxonomy. 

See, for instance, this.  Either it is true, or not, regarding the travel history:



 (click to enlarge)


I think an impartial investigation is needed to get to the truth.  

What we have now is likely healthcare defense attorney and/or risk management "fog", a phenomenon I have both professional and (sadly) personal experience with.

One also wonders if the EHR vendor had a contractual defects non-disclosure ("gag") clause with the organization, and is now threatening suit, leading to the retraction.  (See http://cci.drexel.edu/faculty/ssilverstein/cases/?loc=cases&sloc=koppel_kreda for more on that issue, and of corporate "hold harmless" clauses).   

Unfortunately, a comprehensive investigation would be likely to only occur in a courtroom via Discovery if others become infected.

If any reader has knowledge of details, my email address is scotsilv AT aol DOT com.

-- SS

16 comments:

Anonymous said...

In general, the nurses' reports of findings require search to find in these complex user unfriendly systems. When time os of the essence, if I am not told of key findings by a nurse, it will never be addressed unless my interview and exam of the patient uncovers it.

I have said for years that these systems are impediments to team work, despite the nonsense propaganda spouted by the EHR vendors to Congress and HHS.

Anonymous said...

Too much information in grids that confuses the user by overloading the senses such that no one looks at it, and no one has time to tell the narrative to the decsion makers.

Is this any way to run a hospital? It remknds me of the kid who was sent home from a hospital in NYC and died due to similar communication failure.

Anonymous said...

So another person shows up in an ER expecting likely free medical care. I expect the ER is overloaded and mistakes are made in triage in those circumstances. Maybe the underlying issues of emergency care need to be addressed.

Anonymous said...

EHR's have *plenty* of problems, but to blame this gross error and oversight on a computer is just ludicrous. A man is transported to a hospital with Ebola symptoms and *tells the nurse* that he has recently come from West Africa, and it's the *computer's* fault that they sent him home? Oh, please. That nurse should have verbally relayed that information to anyone who needed to know, and those people should have then continued to verbally relay that information to the proper personnel until the man was placed in isolation, which, by the way, should have taken very little time ("Hi, I have fever and flu symptoms, and I just came from West Africa." "Really? Let's get you isolated right away until we determine whether you're infected with Ebola.").

This is not a computer problem. This is a human problem. I work in the medical field, and I'm not a fan of EHR's, especially as faulty as they are, but hospital personnel blaming the computer for this oversight is like a three-year-old blaming his imaginary friend for spilling his milk.

InformaticsMD said...

To the last two anonymous posters:

Your anonymity suggests you have conflicts of interest you wish to conceal.

You also jump to the now-ludicrous "it was all the user's fault" excuse, clearly lacking any expertise in Medical Informatics, human-computer interaction or any other domain-specific and relevant field.

That said:

I would first argue that while a travel history might not be present on ED physician viewing screens even if irrelevant or blank (although even still, ideally it should be), certainly a positive travel history should be, or at least a notification message of some sort should alert physicians to such a history as captured by another clinician (e.g., triage nurse).

I would agree that the overall causes of this are complex, also probably involving clinician de-skilling due to EHRs (a known phenomenon), over-reliance on the EHR, incompetence, and other human factors, as in any sociotechnical setting.

At best, I would opine that this debacle falls into a "use error" scenario.

Of note, from the NIST usability report previously covered in this blog (search "use error"). This passage, from page 10:

... The EUP (EHR usability protocol) emphasis should be on ensuring that necessary and sufficient usability validation and remediation has been conducted so that use error [3] is minimized.

[3] “Use error” is a term used very specifically to refer to user interface designs that will engender users to make errors of commission or omission. It is true that users do make errors, but many errors are due not to user error per se but due to designs that are flawed, e.g., poorly written messaging [or lack of messaging, e.g., no warnings of potentially dangerous actions - ed.], misuse of color-coding conventions, omission of information, etc.


At worst, this scenario falls into the category of the disruption to mundane care communication and practice caused by just plain bad health IT - e.g., poorly conceived, designed, implemented, maintained, etc.

I would also mention that this type of debacle was what the billions of dollars spent on clinical IT were supposed to prevent, and that with paper perhaps an ED physician would have seen the travel history.

Finally, ironically and in a reverse situation from the usual - who are you to second guess the hospital, who's already admitted to a world wide audience that the EHR was at fault...and even more importantly, they are changing it?

-- SS

Live IT or live with IT said...

If it were a paper chart, I imagine the nurse would have circled it in red. Simple, effective communication in a busy environment.

What was so wrong with the prior method of communications between doc and nurse that it had to be changed? I can see lots of benefits in EHRs, but not everything can be handled with a one-size-fits all situations at all times approach.

EHRs are bed at separating the important alarms from the less important ones.

EHRs don't watch CNN, they have no judgment capability even close to what a human does. They are cognitive assists not cognitive replacement.

Well I sincerely hope this doesn't cause a pandemic in the US. EHRs aren't saving medicine, they are making it harder.

Anonymous said...

But since it was a flaw, like all EHR "flaws", no one was "harmed." Never mind the eventual fatalities from this infection - no one was "harmed."

Anonymous said...

There is something about these devices that cause perfectly competent and seasoned health care professionals to make errors. From screens and grids of legible gibberish, to mind numbing tiny fonts, to the where's waldo phenom, these devices promote errors.

Anonymous said...

Anonymous who states "I work in the medical field":

What field exactly? Let me tell you something bro, that these systems of devices are drop dead dangerous, causing errors, injuries, and deaths that never would have occurred under the systems of care that were destroyed in one fell swoop at the flip of the switch turning on the EHR go live.

Bizarre and absurd adverse events are common.

In fact, as soon as I read that the pt was sent home, I surmised that it was an error of the EHR.

Why not simply have the FDA recall these products and vet them for safety, efficacy, and usability, as should have been done in the first place?

Anonymous said...

Since EHR have contaminated care in the hospitals, when I round on a patient and ask the nurses what is going on with the case, they hand me the mouse to click and find.

There is an attitudinal indifference to the case cause it is "all on the computer". Thoughts are blunted and controlled by the computer.

The EHR is an emerging disease.

Anonymous said...

This case is similar to the Rory Staunton case in NYC, also a collosal communication failure due to EHR:http://www.nytimes.com/2012/10/26/nyregion/tale-of-rory-stauntons-death-prompts-new-medical-efforts-nationwide.html?pagewanted=all&_r=0

The kid was sent home from an ER and died due to data in the EHR but not in the brain.

Unknown said...

It was the doctor.
It was the nurse.
It was the EHR.
It was the hospital.
It was the patient.

So simple to blame one or the other, even though this is a network of interactions among these actors/vectors. More realistic to attribute percentage blame across the five, though that gets complex.

Eg, if the nurse was depending on verbal communication and had directly told the doctor "He just came from Liberia 3 days ago," then the EHR's % of blame would be small.

On the other hand, if one is depending on the EHR as the primary avenue of communication, the failure of this critical nugget to get to the doctor's awareness places the EHR in a much higher blame category.

So, like Heisenberg's principle, the blame is attributed to different actors, depending on how one observes the interactional system. How to change this blame game?

Anonymous said...

It may be available to the doctors if they spent ten or minutes silo searching!

The EHR devices are toxic to safe and tmely care.

Barbara said...

I'm not a medical person and (I don't think) this comment is related to the EHR, but can anybody tell me why he was given antibiotics? I've read that he had a number of blood tests done (before they tested for Ebola). Did he have an infection? Thanks in advance for any info the medical types can provide

InformaticsMD said...

Barbara said...

I'm not a medical person and (I don't think) this comment is related to the EHR, but can anybody tell me why he was given antibiotics?

I can only surmise that the clinicians thought he might have a bacterial infection. Antibiotics are useless in treating viruses.

-- SS

Anonymous said...

Why to blame the nurses are unjustified. The EHR are to many notes to read in 1 minute. It will take 15-20 minutes to treat each patient. First the triage nurse is the one assigned to do triage at the US healthcare institutions to keep the WR patient's flow saved. Triage is defined by ESI, 2012 " The purpose of triage in the emergency department (ED) is to prioritize incoming patients and to identify those who cannot wait to be seen. "The triage nurse performs a brief, focused assessment and assigns the patient a triage acuity level, which isa proxy measure of how long an individual patient can safely wait for a medical screening examination and treatment " It takes just 5 min assessment and the bed will already be assigned from inside by the charge nurse. Do the charge nurse read the triage assessment? Do the primary nurse did her own assessment and review the chart? Do the physician review the nurses notes in the chart prior to proceed to do his full assessment, document intervention, treatment and evaluation before discharging the patient home. Nurses don't discharge patients home without a physician order. Was the Ebola simulation training done at the ER before this event happens? Was the infectious control aware of this event or on duty at the ER that day? Was risk management notify by the charge nurse or physician?. Was the administration on call notified? Why to blame nurses and not to accept it was a complete EHR failure commonly noted by evidenced base practice articles published in the last 5 years as IOM stated"ERR is Human NO to blame one individual just look for the root cause to avoid more EHR hospitals system flaws, learn from it and implement the new policies to avoid more flaws/mistakes. Avoid scaring nurses by punitive actions instead of hospital complete failure. I will love to see more nurses being supported by nurses and STOP blaming nurses. We do care for our patients and respect our profesionas other members of the healthcare team. As a healthcare system in the 21 s century we need to adjust and learn from new diseases arriving to our doors every day. We were not ready for The Ebola diseases to hit us so rapidly. We need nurses to care for patients at bedside not to get rid of them. Ebola patients need nurses as other patients too .
You need a nurse to care for you in case any diseases may hit YOU 24/7. Fix the computer system and train the staff well on quartely bases.