Wednesday, July 13, 2016

Two recent stories of EHR un-exceptionalism, and a connection to prior HC Renewal posts including my own experiences

The following stories recently appeared in the press about the "un-exceptionalism" (or, I might more accurately state, the negative exceptionalism) of today's EHR technology.

Regarding each organization mentioned, I have had personal experience that, in retrospect, aligns with views I expressed about the organizations years ago.


EHR safety goes to court
By Lisa Schencker
June 25, 2016

One patient's blood pressure plummeted dangerously after he was allegedly discharged with the wrong medications. In another instance, a physician couldn't place a pharmacy order for a newborn to receive vitamin K, which is given to babies to prevent serious bleeding.

On several other occasions, patients weren't accurately tracked, creating potential problems getting drugs to them.

Each of these alleged mishaps occurred at PinnacleHealth, a three-hospital system based in Harrisburg, Pa. PinnacleHealth blames each of the mishaps on its electronic health records vendor, Siemens; Cerner Corp. purchased Siemens' health IT business in February 2015.

The relationship between healthcare system and vendor devolved into this dueling lawsuit:

The incidents came to light as part of a breach-of-contract lawsuit Cerner filed against PinnacleHealth last year after the system, which had used Siemens as a vendor for 20 years, sharply curtailed its relationship and entered into a contract with a competing EHR vendor, Epic Systems Corp. PinnacleHealth related the incidents in its counterclaim; the counterclaim was filed in March of this year in state court in Pennsylvania, where it is seeking damages for Cerner's alleged fraud and breach of contract.

(I am cited in the article with respect to hold harmless clause-related issues not relevant to this blog posting.)

Cerner Corp. purchased Siemens' HIT business in Feb. 2015.  Prior to that, Siemens was an independent vendor of an EHR suite known as Soarian, deployed at PinnacleHealth as early as 2008.

See for instance

Soarian Plan of Care Assists Clinicians in Developing and Deploying Goal-Oriented Plans of Care
Monday, February 25, 2008

Siemens ( today announced that PinnacleHealth, a 750-bed, non-profit hospital and healthcare system serving Central Pennsylvania, and CentraState Healthcare System, a 272-bed, non-profit medical facility located in Freehold, N.J., have each signed on to serve as beta sites for Soarian® Plan of Care, a solution designed to support inter-disciplinary care teams in defining and helping to manage patient problems, interventions and expected outcomes, to provide care in order to achieve desired outcomes and to generate new knowledge at the point of care.

Notwithstanding my warnings on this blog and elsewhere that hospitals should NEVER be "beta testing sites" for experimental HIT software used on live patients without informed consent, that the relationship devolved to dueling lawsuits involving safety is not a surprise to me.

Relating to my own experiences as posted previously on this blog:

On August 20, 2009 I posted "Why Siemens Healthcare Fails" (  In that post I noted regarding a job posting by Siemens for a "physician consultant" on HIT implementation:

... This [Siemens] job description might better be described as "glorified salesperson." It might be a good exit route for a "techie doc" (usually, someone who knows just enough about HIT to be destructive) who hates the current practice environment. It might also be good for managers who don't want knowledgeable experts pointing out their bad decisions and mismanagement, but I think a global company like Siemens should be setting its sights higher in such a crucial area as electronic medical records and clinical IT consultants.

I would not want such a physician advising or supporting complex HIT projects at my organization.

I also recalled my own Siemens face time:

... I spent time at Siemens Healthcare headquarters in Erlangen in 2000, and was offered gracious hospitality and a position overseeing the Soarian cardiology suite [as a result of my work in development of an extensive cardiology information system as CMIO at Christiana Care in Delaware from 1996-8]. The people I met in Erlangen then seemed extremely competent and informatics-savvy, but I turned the offer down through no fault of Siemens. I'd received a near-simultaneous offer (FAXed to my hotel in Erlangen, in fact) from pharma that involved a much stronger management role.

I understand through conversations over the past few years with current and ex-Siemens personnel that most of the Siemens personnel I'd met in Germany in 2000 are no longer with the company. I was told they'd performed suboptimally after the acquisition/merger with Shared Medical Systems (SMS) in Malvern, PA. (I do not find that credible, and would find it far easier to accept that the problems were on the American side, but that is a personal opinion.)

I'd informed my German hosts in 2000 that they'd better be very, very careful about acquisition of an American HIT company due to my experiences with a number of such companies, which I found highly political and highly ignorant of EHR quality issues.

Then, in my Feb. 2016 post "Plaintiff's Lawyers Are The Cause of EHR Problems?" ( I recounted further experience with Siemens:

... an anecdote regarding the health IT Industry in the U.S.: the cardiology information system I developed, linked to above in the 2nd bullet point, was seen in 2000 by German engineers at Siemens Healthcare Erlangen as exemplary, and they offered me a position to further develop it, that I declined due to a simultaneous superior offer from Merck Research Labs.  However, in 2007 when I again spoke to Siemens, this time to Americans at the former Shared Medical Systems in Malvern, PA that had been acquired by Siemens, they found a system that actually produced clear, detailed outputs in a critical care area and was in use at the time in a major healthcare system in the region "impractical" - and never followed up with me.  Pearls before....)

 Finally, Siemens seems to have ignored a well-qualified informatics internal expert on explicit EHR safety warnings, as in my Feb. 8, 2010 post "A Lawsuit Over Healthcare IT Whistleblowing and Wrongful Discharge: Malin v. Siemens Healthcare" (

  • Siemens Medical's physician internal consultants (skilled anesthesiologists/informatics specialists), apparently in writing and including a remediation plan, opined that an IT system developed for critical care had numerous severe flaws, of sufficient seriousness that in their opinion the systems could harm or even kill patients if deployed;
  • These were warnings of defects and flaws in IT in the worst possible medical environment, critical care with the sickest and most vulnerable of patients, not some ambulatory clinic or primary care office;
  • Allegations were raised that the warnings were ignored, with at least one of the physicians, Dr. Malin, who was strongly concerned and vocal about the risks then being told his position was being eliminated. [It is not a surprise to note this rings alarm bells about the possibility of wrongful discharge based on retaliation and/or "greasing the skids" to eliminate potential whistleblowers or "non-team players" who could delay release of the software and affect revenue - ed.];
  • The company possibly ignored the remediation plans of their own clinician/informatics experts;
  • There were allegations of company misrepresentations about the new system to the FDA;
  • There were allegations of decision making on these issues by non-clinician IT managers lacking healthcare or healthcare informatics expertise.
  • A wrongful discharge complaint and then lawsuit were filed by Dr. Malin on the basis of violation of the whistleblower protection provisions of the Sarbanes-Oxley Act of 2002 (SOX), 18 U.S.C. § 1514A ("Civil action to protect against retaliation in fraud cases - Whistleblower Protection for Employees of Publicly Traded Companies");
  • The suit was dismissed on the apparent technicality that Siemens Medical in the U.S. is not subject to the provision in the SOX Act as is the publicly-traded corporate parent, Siemens AG; from document #4, pg 15:
... Health Services is a wholly-owned subsidiary of Siemens Med ... Siemens Med is a wholly-owned subsidiary of Siemens Corporation, which is indirectly owned, through two intervening layers, by Siemens AG, a German company that is publicly traded as defined by § 1514A of SOX ... Of these entities, only Siemens AG is a publicly-traded company. While both Health Services and Siemens Med are incorporated in Delaware and located in Malvern, Pennsylvania, they are separately incorporated entities.
  • Siemens Medical in their response to the suit denied the most severe allegations regarding the IT defects, but this issue was not followed up upon due to the lawsuit's dismissal on SOX issues despite the obvious potential public hazards the allegations of IT defects could represent;
  • In fact a US District Judge in the case, Peter J. Messitte, in the period after allowing Malin to prove the validity of SOX towards his case, opined that "No other discovery will be permitted, including but not limited to the alleged safety problems of Defendants’ product", document 4 above, Judge Messitte opinion, US District Court Maryland, pg. 25. [While perhaps understandable from a legal perspective, injured or dead patients don't really care about what legal precedents got them into the injured or dead state - ed.]
  • Siemens Medical admitted in their responses to the suit that some of the software in question, actually put in use in hospitals, was in fact "beta" software, i.e., experimental (per Item 38 in 'Siemens Answer to Complaint' document).

There's that "beta-software" issue again.

So, the acquirer (Cerner) of a product of an American Siemens HIT subsidiary that appeared to have what I consider defective HIT talent management practices (as exemplified by the aforementioned JD); bizarre views about successful, high-quality EHR software (seen by their German counterparts a few years prior as exceptional); and that appears to have not been very happy about its own employee warning explicitly of beta ICU software being so bad it could kill patients is now embroiled in litigation regarding safety by a former customer.

None of this is surprising to me at all.  Incompetence can only go so far in healthcare before things blow up.

An unanswered question, of course, is whether patients were harmed or killed as a result of Siemens software, including its beta tests on unsuspecting patients.


Here is a second story of EHR un-exceptionalism with a personal twist, this article in the Journal of the American Medical Informatics Association (JAMIA):
Comparison of accuracy of physical examination findings in initial progress notes between paper charts and a newly implemented electronic health record, June 28, 2016

Introduction There have been several concerns about the quality of documentation in electronic health records (EHRs) when compared to paper charts. This study compares the accuracy of physical examination findings documentation between the two in initial progress notes.

Methodology Initial progress notes from patients with 5 specific diagnoses with invariable physical findings admitted to Beaumont Hospital, Royal Oak [Michigan, near Detroit -ed.], between August 2011 and July 2013 were randomly selected for this study. A total of 500 progress notes were retrospectively reviewed. The paper chart arm consisted of progress notes completed prior to the transition to an EHR on July 1, 2012. The remaining charts were placed in the EHR arm. The primary endpoints were accuracy, inaccuracy, and omission of information. Secondary endpoints were time of initiation of progress note, word count, number of systems documented, and accuracy based on level of training.

Results The rate of inaccurate documentation was significantly higher in the EHRs compared to the paper charts (24.4% vs 4.4%). However, expected physical examination findings were more likely to be omitted in the paper notes compared to EHRs (41.2% vs 17.6%). Resident physicians had a smaller number of inaccuracies (5.3% vs 17.3%) and omissions (16.8% vs 33.9%) compared to attending physicians.

Conclusions During the initial phase of implementation of an EHR, inaccuracies were more common in progress notes in the EHR compared to the paper charts. Residents had a lower rate of inaccuracies and omissions compared to attending physicians. Further research is needed to identify training methods and incentives that can reduce inaccuracies in EHRs during initial implementation.

Apparently not covered in the article was the issue of patient harms that might have occurred (and could still occur) due to the "significantly higher rate of inaccurate documentation" in the EHR.   

"Beaumont Hospital" caught my eye, as I had interviewed for the CMIO role there in 2007 or early 2008.  My experiences are memorialized in an anonymized post at my Drexel informatics site which I reproduce in part here (the full essay is at
Sure, the experts think you shouldn't ride a bicycle into the eye of a hurricane, but we have our own theory

A medical informaticist who formerly held a “Director of Informatics” role a number of years ago in a very large hospital system, and who left the role due to a toxic management environment and lack of authority commensurate with responsibility, was seeking applied Chief Medical Informatics Officer (CMIO) positions once again.  ... He makes the following observations after completing two full rounds of interviews at a prestigious hospital system similar in size to his erstwhile employer, in a very competitive environment, that recently experienced a decline in its clinical quality stats. The organization feels the quality stats themselves were inaccurate, in part due to lack of good healthcare IT.

From what the informaticist was able to gather, their leadership was displeased. Board members were seasoned executives from a heavy-manufacturing industry that is extremely dependent on information technology and concurrent supply chain data. These executives apparently recommended that the organization move quickly on implementing EHRs.

The organization is thus planning to implement EHRs for thousands of physicians, most of whom are not employed by the hospital but are independent private practitioners, and likely to be skeptical or concerned about time impact and “grading” that could affect their livelihoods.  The hospital leaders also wanted to create integrated systems drawing on EHR data to automate quality reporting to regulatory agencies, as well as to support ongoing, funded clinical drug and device trials.

The informaticist was interviewed by the usual mix of clinician eager adopters, clinician skeptics, knowledgeable executives, skeptical executives who knew little about clinical IT, IT personnel who seemed overconfident given the enormity of the tasks at hand, and those who were clearly frightened by the prospect of being held accountable for a project of this magnitude. In the end, the informaticist did not get the position due to the organizational leaders being adamant the incumbent CMIO needed to also practice medicine.  The informaticist had explained on the first round that he believed a leadership role in a project of such magnitude and challenge called for the highest levels of executive presence and freedom from distraction, thus he did not intend to practice medicine (he had not practiced in his former Director of Informatics role for the same reasons).

He’d thought this issue had been settled after the first round of interviews, leading to the invitation for round two. This line of questioning was revisited, however, in round two in a group interview setting. The group interview was attended by a number of people with whom he’d already discussed this issue via individual meetings in round one. This suggested his time was being wasted and was rather annoying, especially considering that the informaticist had flown cross-country not once but twice to an organization not in consensus about a very basic hiring requirement
... It is not as if the organization had a doctor shortage, or that such a role would have ample free time where the incumbent would be idly sitting at their desk unless this time was absorbed seeing patients in the clinic. 

The stated reason for the organization’s wanting the CMIO to practice medicine was “to have credibility with the doctors.”  The informaticist explained that he’d found this not to be the case, that physicians being put “under the gun” of using EHR’s were generally more concerned that the CMIO had the executive authority to best represent their interests and understood medicine from training and practice at some point, not necessarily concurrently with the CMIO role.  Interestingly, there is no empirical research on this point, so the issue was the informaticist’s experience vs. the hospital’s ‘second-guessing’ a seasoned expert.

Ironically, the informaticist was told during his interviews that a CMIO they'd hired a few years ago had left, in part due to being overextended.  He was also told that some of the clinical IT problems he'd solved as a CMIO in the past Director of Informatics position were problems this organization had not been able to solve during the same time frame.

... Regarding underestimation, this organization appeared to have little idea of the difficulties they were getting into

... The informaticist had observed another indication that this organization ‘didn't know what they didn't know.’  He was informed that the organization had selected their EHR vendor prior to seeking a medical informatics expert. This implies they really did not understand what a medical informatics specialist does and can do, which is far more than being a tactical "EHR implementation assistant." ... The informaticist had been there, and had done that in his past role. He found it unrewarding then and actually had decided not to take the risk again, rejecting the new position, even before the organization decided they wanted an (effectively) part-time CMIO who also saw patients.

It was clear this organization, who the informaticist tried to “take to school” based on hard-earned expertise and extensive references on social issues in health informatics (e.g. on this website), felt they were the experts on what was best regarding CMIO background.   This may have been a dysfunctionality satirist Scott Adams once described like this:

Ignoring the Advice of Experts Without Good Reason
Example: Sure, the experts think you shouldn't ride a bicycle into the eye of a hurricane, but I have my own theory.

As I recall, an informatics physician much younger than I and with less experience was hired.  I do not know his fate.  However, I believe that my expertise at that time in EHR issues, and my no-nonsense stance on EHR safety and excellent management, might have averted or minimized the "inaccuracy" problems.

Therefore, it is also no surprise to me that the organization now relates the EHRs were causing significant charting inaccuracies.  It would also not surprise me if close calls and harm occurred. They apparently rode their bicycle into that hurricane, and doing so will usually produce a less than optimal outcome.

In summary, two recent articles on EHR-related lawsuits and operational problems, in my own personal experience, have potentially relevant "backstories" that may help provide context.  Those backstories relate primarily to what I've called "HIT amateurism" (e.g., faulty expertise evaluation) as well as HIT mismanagement (e.g., ignoring explicit internal warnings of dangers.)  I believe my experiences, and that of other such as Dr. Malin, are symptomatic of widespread health IT industry dysfunction.

I also believe that if this industry is ever to learn from its mistakes, it will only occur in the courtroom.  Being that the defense side is usually fierce, however, I am not sanguine needed learning will occur any time in the near future.

-- SS

1 comment:

Anonymous said...

The defense side is not only "fierce", it s shameless.