Monday, June 27, 2011

Babies' deaths spotlight safety risks linked to computerized systems

Two new tragic cases. (A third, earlier case of infant death was posted in March 2011 here).

Case #1. See Baby's death spotlights safety risks linked to computerized systems by Judith Graham and Cynthia Dizikes in the Chicago Tribune.

A baby died at Advocate Lutheran General Hospital due to an intravenous solution containing a massive overdose of sodium chloride — more than 60 times the amount ordered by the physician. The authors write:

Although a series of other errors contributed to the tragedy, its origin — a piece of data entered inaccurately into a computer program — throws a spotlight on safety risks associated with medicine's advance into the information age, a trend being pushed aggressively under health reform.

One wonders if any type of alert was generated, and if not, why not.

Case #2. There's also been a baby death locally to me that is allegedly due to HIT (PACS) interference in otherwise simple care processes.

See "Blake vs. Abington Memorial Hospital" at this link. The Complaint in the latter case, a public document, is available as a .tif image at this link (1 MB; you may need to use IE to download beyond the first page).

(July 8 addendum: since this blog post was published, I've noted multiple "hits" on it from the IP of the hospital involved, 65.216.195.x , at different times of the day, as well as on my Drexel website on HIT failure. Clearly they are aware of the issues noted herein.)
(Sept. 2011 addendum: the hospital has now filed a Motion to Compel Settlement that can be seen as item #67 and downloaded from the aforementioned courts dockets page here.)

A critical x-ray of a PICC (peripherally inserted central catheter) line placement allegedly was never read due to misdating. See for example counts 17 - 25, and 39 - 45. It was dated several days earlier than it was taken.

Counts 39-49 (emphases mine):

39. Defendants also described that the February 27, 2010, x-ray was digitally transmitted to radiology, and the technician attempted to match the image to doctors’ orders, and then put all the identifying information on the film, including the patient’s name, date of birth and date of study.

40. Defendants told Plaintiffs that the radiology technician picked the wrong name and put it on the x-ray, and then later realized it was the wrong name and then somehow changed the information on the image.

41. Defendants told Plaintiffs when the technician changed the information on the image, she typed the date of birth where she should have typed the date of study. [Puzzling; my $99 Canon camera embeds date information in the image itself, as do PACS systems - ed.]

42. Defendants told Plaintiffs that the incorrect date on the image caused the x-ray not to be read by a radiologist in the night the image was taken, and that if the image had been correctly dated February 27, 2010, it would have been reviewed by a radiologist immediately, who would have alerted Defendants that the PICC line was improperly positioned and had to be moved back to a safe position. However, since the film was incorrectly dated February 23, 2010, the film was classified as “old” and was not put in line for immediate review, and was not reviewed in time to prevent injury to Plaintiffs Decedent.

43. The Defendants told Plaintiffs that an older x-ray has less significance than a newer image and is, therefore, not as critical to review. [This is exactly the reverse of good radiological practice. 'Old' - but never officially read - means 'potential danger' such as missed findings of importance, subtle or gross findings that could alter the interpretation of later films, etc. - ed.]

44. The Defendants told Plaintiffs that the x-ray was reviewed by Dr. Griffith on the morning of February 28, 2010. At that time, Dr. Griffith, thinking that the image was from the 23rd, compared it to the next most recent image, which was from the 25th In reviewing these x-rays, Dr. Griffith reportedly concluded that the PICC line shown in the image she believed was from the 23rd had been removed by the 25th. [Gevalt - what a tragic error. In healthcare, critical attention to detail at all times is required, or your patient's dead. There is no room for computer fritter. - ed.]

45. Defendants told Plaintiff that the x-ray from the 27th, which was mislabeled as being from the 23rd, would have been reviewed by the neonatologist on duty on the morning of the 28th, but this did not happen because the computer which places films in line to be reviewed would have ignored such an “old” film.

[Note the accusation of the computer making a decision on what clinicians should and should not look at, thereby acting as cybernetic 'governor' or 'conductor' of care. The machine or, more properly, the algorithms created by its programmers, decided what the humans should or should not look at. This is not uncommon in HIT. Yet, who qualified these machines and their code to play this role? More on this critical issue
at this link. - ed.]

46. The Defendants told Plaintiffs that the next 24 hours were uneventful, but late in the day on Monday May 1, 2010, Destinee’s condition began to deteriorate rapidly.

47. The Defendants told the Plaintiffs that the during the efforts to reverse Destinee’s deterioration, they discovered that the PICC line was in the wrong position, and that they confirmed this by sampling the fluid from around the heart, which was found to be full of hyperalimentation fluid, and then pulled the line back from its incorrect and dangerous position in the heart.

48. The Defendants told Plaintiffs that they did not know if the PICC line caused a hole in the heart, but that just having the line in the heart was enough for it to push the hyperalimentation fluid into the pericardium.

49. As a direct and proximate result of Defendants’ negligence, Plaintiff’s Decedent’s suffered injuries, including but not limited to: [numerous catastrophic complications followed by death - ed.]

If these counts are accurate, I ask the same question as in case #1: if an x-ray was accidentally dated five days or so before it was taken (how can that even happen?), why was an alert not triggered that an "old" (and unread) x-ray needed an official radiologist reading ASAP?

What are hospitals getting for their tens to hundreds of millions of dollars they are throwing at clinical IT?

This latter story really disappoints me if true. I was a Resident in Internal Medicine in this hospital in the mid 1980's and used my knowledge of radiology, in which I spent a year of formal postdoctoral education, and of computing to fix a CT scanner in the middle of the night and save a young man's life (as documented in a letter from my former Residency Director, now the Chief of Staff here and here , pgs. 1 and 2 respectively).

Ironically, even after all these years, I probably could have read that PICC line placement x-ray.

(Note:  more on the x-ray foul up in the Phila. Inquirer here.  "Use error" - per NIST, bad IT design that promotes mistakes, as opposed to "user error" i.e., simple carelessness - seems a factor.)

Technology can be a great blessing in medicine. I am alive and healthy in 2011 thanks to it. The medical condition referred to in the aforementioned letter was in fact cured as a result of advances in cardiac electrophysiology science, as here and here, pgs. 1 and 2).

However, technology cannot be applied cavalierly, lawlessly (per attorney Sharona Hoffman), experimentally, without regulation, and without informed patient consent as is often done in the world of HIT.

-- SS

Feb. 29, 2012 Addendum: the Allen-Blake case was settled 2/29/12 for $1.5 million:

2/29/2012OrderOF 2/28/12 OTT, J PROPOSED SETTLEMENT OF $1,500,000.00 APPROVED CC

This will not, of course, bring the baby back.  Also, I certainly wonder if this is looked at by some as a cost of doing 'EHR business.'


June 27, 2011 addendum:

In further review of the dockets in case #2, I find there seems to be a wee problem getting the hospital's Chief of Staff Jack Kelly to a deposition.

In a motion filed June 10, 2011 by the plaintiffs, "BY ADMINISTRATOR ERICA L ALLEN-BLAKE; RICARDO BLAKE; ESTATE OF DESTINEE LOTOYA BLAKE DECEASED MOTION TO COMPEL DEPOSITION OF DR. JACK KELLY" (.tif file at this link), in Exhibit "J" the plaintiff's attorney writes to the defendants' attorney (emphases mine):

I received your faxed letter of May 2, 2011. I don’t agree with any of your sophistry, and want to depose [Chief of Staff] Dr. Kelly. I remind you that you objected to the Complaint, claiming the meeting with the Plaintiffs had a specific purpose under MCARE (which never made any sense, but you wrote it, not me). Your objection raised issues of fact. If you don’t want to back them up, or can’t, then withdraw the objections.

If you’d like to ask the Plaintiffs one question at a deposition taken pursuant to your pending objections - i.e., whether they were told by your clients that MCARE written notice was purpose of the meeting or not - I’d agree to let you do that. I can represent to you that their answer will be "no.”

Be advised that I am tired of Abington’s dilatory tactics in this clear liability case in which an infant was killed by, among other things, some of the most egregious professional malpractice imaginable. On top of that, Abington refused to provide the name of the radiology technician until suit was filed. Now, you’re playing ticky-tack games about depositions, and my clients are understandably not impressed.

Under cover of separate letter you will find a Notice of Deposition for Dr. Kelly for this Friday, May 6, at 2 pm in my offices.

Wow.

A motion to compel was filed on 6/22/2011 so apparently the May 6 deposition did not take place.

I add the the MCARE Act of 2002 ("MEDICAL CARE AVAILABILITY AND REDUCTION OF ERROR ACT") in Pennsylvania (PDF at this link) states:

MCARE Section 308. Reporting and notification.

... B. Duty to notify patient.--A medical facility through an appropriate designee shall provide written notification ['written' - this word would appear to mean "in writing" - ed.] to a patient affected by a serious event or, with the consent of the patient, to an available family member or designee within seven days of the occurrence or discovery of a serious event. If the patient is unable to give consent, the notification shall be given to an adult member of the immediate family. If an adult member of the immediate family cannot be identified or located, notification shall be given to the closest adult family member.

Allegedly from the June 10 Plaintiff's Motion, as it appears to me, the hospital seems to be trying to claim a verbal meeting with the family satisfied the MCARE Act's notification requirements even though nothing in writing was apparently given, but that the information exchanged in that family meeting is not admissible in court.

It seems hospitals (perhaps pushed by their insurers) attempt to play legal games to try to avoid culpability for their "computer-anecdote" patients, some who end up as roadkill on the experimental racetrack of health IT.

Again, this is doubly disappointing if accurate, as I knew some of the principals on the defendants' side two and a half decades ago.

Also see a third case at my aforementioned March 2011 post "Thanks to the wonders of EHR, this premature baby went to the grave, prematurely" at this link, from an Oct. 2010 story in Virginia Lawyers Weekly ("Failures in care alleged after premature birth - $1,000,000 Settlement"). A PICC line was also involved.

-- SS

4 comments:

Live it or live with IT said...

The Supreme Court has stated that materiality is a fact-based inquiry; “[t]his is not to say that statistical significance (or the lack thereof) is irrelevant – only that it is not dispositive of every case.”

Anecdotes, especially those that portend failures, are very important. Every malpractice claim is at its core an anecdote.

InformaticsMD said...

"Anecdotes" are often roadkill on the experimental racetrack of health IT.

-- SS

Anonymous said...

When the child of an inside the beltway lawmaker is overdosed and killed from an EHR or electronic ordering gismo, there may be an interest in Congress to understand the deception and fraud playing out in its chambers.

InformaticsMD said...

Anonymous June 27, 2011 9:18:00 PM EDT wrote:

When the child of an inside the beltway lawmaker is overdosed and killed from an EHR or electronic ordering gismo, there may be an interest in Congress to understand the deception and fraud playing out in its chambers.

That would be the "cybernetic Libby Zion case" I have written about, that is probably inevitable.

-- SS