EHRs and Ebola in the Texas Health Presbyterian Hospital ED: the ED physician finally speaks out

At my Oct. 2, 2014 post "Did Electronic Medical Record-mediated problems contribute to or cause the current Dallas Ebola scare?" (http://hcrenewal.blogspot.com/2014/10/did-electronic-medical-record-mediated.html) I had written:

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 [that is, an EHR-related information snafu - ed.] needs to be investigated fully.

 
I then did an update:

10/3/2014 Update:

My suspicions were apparently correct.  [The hospital admitted an EHR role - ed.]

Then, the hospital retracted its admission, blur and obfuscation broke loose in the press, and the situation became foggy.  See posts by Roy Poses and myself at query link http://hcrenewal.blogspot.com/search/label/Ebola%20virus, including Dr. Poses' Nov. 24, 2014 post "Public Relations and the Obfuscation of Management Errors - Texas Health Resources Dodges its Ebola Questions" at http://hcrenewal.blogspot.com/2014/11/public-relations-and-obfuscation-of.html.

Finally, the primary clinician involved speaks.  Do read the whole article, as it delves into behind-the-scenes issues:

ER doctor discusses role in Ebola patient’s initial misdiagnosis
By REESE DUNKLIN and STEVE THOMPSON
Dallas Morning News
Dec. 6, 2014
http://www.dallasnews.com/ebola/headlines/20141206-er-doctor-discusses-role-in-ebola-patients-initial-misdiagnosis.ece

... "[ED physician Joseph Howard Meier's] notes in the medical records say he had reviewed the nursing notes. Hospital officials told Congress that the ER physician several times accessed portions of the electronic records where the nurse had recorded Duncan’s arrival from Africa. It wasn’t clear, though, “which information the physician read,” hospital officials told Congress. 

Meier told The News he hadn’t seen the Africa notation in Duncan’s records. The physician said the hospital’s electronic medical records system contained a lot of information, which, like patients, “must also be triaged.” 

Clinicians in an ED have to "triage" information from their records systems, just like patients need to be triaged?  That is a candid and astonishing (to anyone with common sense) admission.

Paper charts never had those problems in my own time working in the ED.

Further, ED charts used to be relatively brief, which is why as a Chief Medical Informatics Officer I recommended document imaging systems only in ED's, to make charts available 24/7/anywhere, and data transcriptionists to capture important data into computers later, not full EHR systems where clinicians enter data which I felt (and still feel) are inappropriate in faced-paced, high-risk settings.

(Put another way, the experiments of direct data entry by busy clinicians, and clinicians attempting to drink information from a tangled cybernetic EHR firehose, are proving a failure.)

... The “travel information was not easily visible in my standard workflow,” he said. “This has now been modified very effectively.”

Modified only after near-catastrophe.  How many other "modifications" (i.e., experimental software changes) will be needed over time in this and other EHRs, I ask?  (Perhaps 10,000 such as here: http://hcrenewal.blogspot.com/2014/06/in-fixing-those-9553-ehr-issues.html?)

... The News asked Meier whether knowing Duncan’s travel history would have changed his evaluation. 

“If he told me he came from Liberia, this would have prompted me to contact the CDC and begin an evaluation for Ebola,” Meier said, “but the likelihood would have still been low since Mr. Duncan denied any sick contacts.”

Over the next few hours, Meier ordered tests, along with an IV for saline. He prescribed extra-strength Tylenol, which the nurse gave Duncan at 1:24 a.m. Meier reviewed Duncan’s vital signs. CT scans of Duncan’s head were “unremarkable,” the medical records say, showing no sign of sinusitis.

Doctors typically order CT scans to rule out more serious possibilities, such as a hemorrhage or meningitis. In his responses to The News, Meier said he ordered the CT scan because of Duncan’s headache.

Meier did not say whether the CT scan’s lack of an indication of sinusitis factored into his diagnosis. “Sinusitis is mostly a clinical diagnosis,” he said.

At 3:02 a.m., Duncan’s temperature was 103 degrees, his medical records say. Sixteen minutes later, however, Meier entered a note saying: “Patient is feeling better and comfortable with going home.” Meier told The News he hadn’t seen the higher temperature in Duncan’s chart.

Duncan was discharged at 3:37 a.m. with the diagnosis of sinusitis. His last recorded fever, at 3:32 a.m., was 101.2 degrees. Meier prescribed Duncan the antibiotic Zithromax, 250-milligram tablets, to be taken twice the first day and once daily for four more days.

I note two things:

1.  If an EHR company has hiring practices allegedly such as described via Histalk blog at my Aug. 15, 2010 post "EPIC's outrageous recommendations on healthcare IT project staffing" (http://hcrenewal.blogspot.com/2010/08/epics-outrageous-recommendations-on.html), where rank-novice recent college graduates suddenly become EHR experts afters some transfusion of wisdom at corporate HQ (perhaps via this alien neural interface device that imparts the Knowledge of the Ancients: http://stargate.wikia.com/wiki/Repository_of_knowledge?), then what can one expect?

The Stargate neural interface device that imparts the Knowledge of the Ancients via direct brain download.  Presto - instant EHR expert!

and

2.   I note what I am going to somewhat satirically going to call the "Silverstein EHR principle", that states:

• When bizarre and otherwise inexplicable information-related snafus occur in hospitals, especially in fast-paced, high-risk areas, suspect bad health IT as causative or contributory as #1 in your differential diagnosis (or post-mortem, as the case may be).

-- SS