Tuesday, October 21, 2014

Judy Faulkner and EPIC, Show us your EHR screens

At "Congressional committee releases timeline detailing how Presbyterian treated Ebola patient Thomas Eric Duncan", Dallas News, Oct. 17, 2014 there is a link that provides acccess to documents released by the U.S. House of Representatives' Energy and Commerce Committee.

These documents address the EHR issues in the care of Ebola patient Thomas Duncan I wrote of 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) and others:

Congressional committee releases timeline detailing how Presbyterian treated Ebola patient Thomas Eric Duncan

By Robert Wilonsky
Dallas News
Oct. 17, 2014 

http://thescoopblog.dallasnews.com/2014/10/congressional-committee-releases-timeline-detailing-how-presbyterian-treated-ebola-patient-thomas-eric-duncan.html/

According to a timeline released moments ago by the U.S. House Energy and Commerce Committee, Texas Health Resources Presbyterian Hospital Dallas released Thomas Eric Duncan at 3:37 a.m. on Sept. 26 — just 35 minutes after his temperate jumped to 103 degrees.

The timeline, provided by Presbyterian officials, also shows that “obtaining the patient’s travel history was not part of the triage nurses’ process on September 25, 2014,” when Duncan initially went to the hospital. He arrived in Dallas from Liberia five days earlier. A nurse noted that he’d just come from Africa but “attached no further significance to this travel history,” according to the timeline.

Another document shows how Presbyterian prepared to deal with Ebola dating back to Aug. 1 when officials were told that all Emergency Health Records should include a travel history for every patient. In Duncan’s case, it’s not clear whether a doctor read his emergency health records.

The record does not show which information the physician read, only which information was available,” according to the timeline.

The documents are available at https://www.scribd.com/doc/243373964/Thomas-Duncan-Presbyterian-Treatment-Time-Line

The key phrase to parse is the one also quoted in the newspaper article above:  “The record does not show which information the physician read, only which information was available.”

From the timeline itself, in pertinent part:

12:33 – 12:44 a.m. RN assessment
- The primary ED nurse continues the assessment.
- She identifies his complaints as “sharp, intermittent epigastric/upper abdominal pain;
sharp, frontal headache; dizziness; lack of appetite”
- She asks about Mr. Duncan’s travel history.
- The nurse documents that Mr. Duncan “came from Africa 9/20/14"
- RN states she recalls the discussion because of how long the plane flight was. (She had personal experience with very long plane fights). Attached no further significance to this travel history.
- This information was not verbally communicated to the physician, as prompted by the EHR.

12:52 – 1:10 a.m. ED physician begins evaluation of Mr. Duncan
The ED physician accesses the EHR again. A review of the EHR shows that the physician, on several occasions, accessed portions of the EHR where the travel history was now available including:
ED Lab Results Screen
ED Triage [twice]
ED Rad Results

The record does not show which information the physician read, only which information was available.

Again, the statements that the physicians "accessed portions of the EHR where the travel history was now available" after the RN recorded it, and that "the record does not show which information the physician read, only which information was available" sound like lawyers writing to obfuscate EHR realities from our Congresspeople.

Let's examine these statements:

  • ED physicians "accessed portions of the EHR where the travel history was now available" after the RN recorded it, and
  • The record does not show which information the physician read, only which information was available

These is a fundamental semantic problem here with the word "available."  In an EHR, "available" has a far different meaning than in a paper record.

The question is:

What is the precise meaning of the word "available" as stated here?

  • (1)  Does "available" mean "present on the actual screen(s) the physicians had up at one time or another on the monitor, that made up the "portions" of the EHR they "accessed"?"  
  • In other words, was the positive travel history from Africa "illuminating the phosphor", or illuminating the LED arrays for a modern computer monitor, of an actual screen in actual eyesight of the physicians that was a subset of the "portions of the EHR" they accessed?
 Or (and I believe this quite possible):

  • (2)  Does "available" mean that the travel history was available as data on disk or on RAM, and thus potentially on a screen for a physician to see, but that the specific screen never actually illuminated the LED arrays on the physicians' monitors? (E.g., such screen(s) were a component or subcomponent of the EHR "portions" they accessed, but the specific screen(s) in those "portions" had to be navigated to in order to see the travel data.)
  • In other words, was the case that the travel information taken by the nurse never appeared visually to the physicians, but only resided in the computer as data where it was invisible as intangible bytes on a disk or in RAM?  (This does not happen with a paper chart - the paper is tangible.)
  • Further, was there a meaningful alert drawing the physician to a screen that did then present the travel data to them?

There is no way to know by parsing the words, but based on their semantic blur I suspect the second scenario.

Unfortunately, what  really is essential to understand the EHR interaction are screenshots of precisely the screens viewed by the physicians, not "available" to the physicians.

Note that, for example, my Windows System Event log is "available" to me at all times in "portions" of Windows I may look at - by right-clicking "My Computer" and clicking the "manage" menu item that appears -  and only then if I actually then navigate to find it.  

Of course, EPIC and the other EHR sellers do not make the actual EHR screens available to the public - they are considered "protected IP."

Perhaps it's time for EPIC and the Texas Health Presbyterian Hospital to show Congress their screens.

Assuming they even know what screens to show.  EHR audit trails of user activity are notoriously imprecise. 

-- SS

10/21/14 Addendum:

At Health Data Management (http://www.healthdatamanagement.com/news/Epic-Stands-By-Integrity-of-EHR-System-at-Dallas-Hospital-49039-1.html),  Carl Dvorak, president of Epic Systems Corporation, is quoted as saying "... obviously it [the travel history - ed.] was on the opening screen of the physician’s workflow.”

I say:  prove it.  And as above, prove the doctors actually "put the data up into the screen LEDs."

Show the screens (before the hospital changed them, I add).

Show the audit trail.

This EPIC statement makes no sense, considering the hospital's initial claims as I wrote about earlier:

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, Texas Health Presbyterian Hospital Dallas said in a statement Thursday.

However, the screens and the audit trail are the only way to authenticate the EPIC claims.

-- SS

7 comments:

Anonymous said...

Epic and the hospital client are making up the real truth as they go along.

$$$ billions at risk for the vendor, and untold costs for society because the patient was sent home.

It is hard to believe that any doctor would send Mr. Duncan home, with low platelets, acute renal failure, and a fever of 103 F, if they had that data in hand. They obviously did not, consistent with silent silos and the need to search silos for any and all data; and data may have come back to the EHR after they exited from that case.

Of course, Epic will say that travel history was on the home screen, after they revised the device to emphasize travel hisotry.

As you know, once a change is made, all screens and presentation of data are retrofitted showing the update.
Thus, the truth is altered since the screens that the docotrs saw in real time no longer exist.

Anonymous said...

lol @ as if they would prove it.

Anonymous said...

Dammit. This makes me so mad. Retrofitted data. Wrong wrong wrong. We may never see the truth from this Sentinal event and all future patients are at risk.

Anonymous said...

Yep, these are the right questions. I wish we could get the answers.

Anonymous said...

Retrofitted data.

We may never see the truth from this Sentinal event and all future patients are at risk.


Ask yourself: "Bug or Feature?"

Anonymous said...

After what happened in Texas, Epic has added additional admission questions about travel history, specific to Ebola where I work. There were never any questions pertaining to travel before. One would have to write a progress note if there were a concern, of course someone would then have to read it.

Anonymous said...

Paul Rowan sez: This just came out a week ago, end of Oct:

Divvy K. Upadhyay, Dean F. Sittig, Hardeep Singh. Ebola US Patient Zero: lessons on misdiagnosis and effective use of electronic health records. Diagnosis, 2014, online ahead of print.