Friday, March 24, 2017

An eloquently expressed lesson from Nanaimo (Canada) on electronic medical records failure

Unfortunately, this eloquent piece on EHR failure expresses precisely the major problems with this experimental technology that generic medical managers and other medical bureaucrats are unwilling to hear, and/or unable to fully comprehend.

At my May 31, 2016 post "HIT Mayhem, Canadian Style: Nanaimo (Vancouver Island) doctors say electronic health record system unsafe, should be shut down, non-medical PR hacks say it's perfectly safe" at, I wrote:

... To hell with doctors and nurses and their concerns about horrible health IT.   

That seems the international standard in 2016 regarding their concerns.  There's just too much money to be made in this business to worry about such piddling annoyances as maimed and dead patients.

Doctors, after all, don't know anything about computers, and cybernetic medical experiments on unconsenting human subjects are
 just good fun.

This new example from Canada: 

Nanaimo doctors say electronic health record system unsafe, should be shut down

By Cindy E. Harnett
Victoria Times Colonist
May 27, 2016

Implementation of a $174-million Vancouver Island-wide electronic health record system in Nanaimo Regional General Hospital — set to expand to Victoria by late 2017 — is a huge failure, say senior physicians.

Who cares what they say?  They're just doctors, so sayeth the imperial hospital executives.. 

Here is an update:

A lesson from Nanaimo on the human costs of electronic health records


Dr. David Forrest (Dr. Forrest is a Nanaimo internist and the president of the Nanaimo Medical Staff Association.)

For the last 12 months, staff at Nanaimo Regional General Hospital in B.C. have been using a Cerner-based electronic health record (EHR). This includes computer physician order entry, linked to computerized order management, and fully electronic documentation.

Since shortly after the system was activated, physicians have consistently and increasingly expressed concerns about the safety of the system (with, for example, orders being mistranslated by the system or disappearing) and its inefficiency, which reduces patient access to care. As a consequence, with the support of the B.C. Ministry of Health, the Island Health Board has directed the Health Authority to suspend CPOE and related processes—though Island Health is delaying this.

While the primary issues with iHealth have focused on safety concerns, little attention has been paid to the human costs of the EHR and its implementation.

"Little attention has been paid to the human costs" due to the fact that most of that attention has gone to the entity that seems to have usurped the rights of patients and clinicians, namely, computers.   "Cybnernetics over all" seems the continuing saga of healthcare, as I'd written in the past.

I am increasingly appalled when I read common accounts like this, now ongoing since my entry into medical informatics 25 years ago.  The term "learning organization" that I've been hearing since the 1990's extolled as what healthcare managers claim they aspire to seems a sad joke.

Currently designed EHRs significantly alter processes of care. Computerised order entry is a laborious process, requiring multiple steps to perform simple tasks. As demonstrated in other settings (such as the airline industry), such complex processes are inherently error-prone. Moreover, the inputting of more intricate orders is even more difficult, resulting in “work-arounds” or inaccurate enactment of physician management plans, with additional safety risks.

I've written in past posts that the "atomization" of order entry and other functions into multiple subcomponents with different user interface widgets, rules and "gotcha's" is a monstrous tool created by those without a knowledge of what it's like in chaotic patient care environments.  Those designers also seem, for example, to lack the ability to design tools that parse freetext orders into those atomic parts with little user intervention.  The syntax of the bulk of medical ordering is not all that complicated, for example.

“User error” has become a buzz phrase for system-based processes that result in inability to enact orders as intended by physicians—and physician users are the ones responsible, thus morally if not legally liable not just for order entry but for their management downstream. It is little wonder physicians in Nanaimo since the introduction of the EHR feel uncertain, anxious, frustrated and exhausted.

Two points, the first of which I expressed to the author via email. 

First, "user error" is a buzz phrase for those ignorant of human-computer interaction best practices to lay blame for faulty designs on victimized (and usually compelled) users.

The more correct term in many cases is "use error" - not "user error."  Use error, as defined by the US National Institute of Standards and Technology (NIST) in their study of health IT, is as follows (see also my Oct. 2011 post at

“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.

The second point I will make regarding clinicians feeling "uncertain, anxious, frustrated and exhausted" is that this health IT fits the definition of bad health IT in terms of its creating stressors:

Bad health IT is ill-suited to purpose, hard to use, unreliable, loses data or provides incorrect data, is difficult and/or prohibitively expensive to customize to the needs of different medical specialists and subspecialists, causes cognitive overload, slows rather than facilitates users, lacks appropriate alerts, creates the need for hypervigilance (i.e., towards avoiding IT-related mishaps) that increases stress, is lacking in security, lacks evidentiary soundness, compromises patient privacy or otherwise demonstrates suboptimal design and/or implementation. 

That definition was the result of joint thinking by myself and a colleague, Australian informaticist Jon Patrick, PhD.

This experience is not singular, as a number of studies in the U.S. suggest that EHRs are now the primary cause of physician burnout. This further jeopardises patient care since the presence of an EHR has become a barrier to recruitment and retention of healthcare workers—and not just of physicians.

If this is true, and I believe it largely is, any claimed benefits of EHR technology must be seriously questioned as to the benefit/cost (downsides) ratio.  This is also NOT what the pioneers who taught me during my Medical Informatics fellowship a quarter-century ago intended, and this deviation from their intentions can be accounted for only by numerous social pathologies I will not get to in more depth in this post.

The following is quite eloquent, expressing the problems in a way I rarely recall seeing so lucidly stated:

Documentation changes have also affected patient care. A patient’s journey through illness and during their stay in hospital is a story or narrative—hence our documentation of that narrative as a history. Our understanding of it in this manner is critical not only to diagnosis and management, but to contextualizing it for the individual patient.

“Progress notes” in the EHR no longer describe progression of disease, but document it rather as an episodic and disjointed accounting of the patient’s condition. And the nursing narrative has been eliminated in favour of checklists of patient experience as data points. As a result, we have lost our holistic knowledge of the patient and his/her illness and are unable to transmit this understanding clearly from one provider to another.  This is detrimental to providing high-quality patient care.

Understanding the crucial differences between narrative of a patient's journey and an episodic, disjointed "data dump" as EHRs now foster is perhaps a capability that differentiates those who were able to get the grades necessary for entry into medical school and succeed there through school, internship, residency and perhaps specialty fellowships, from those with lesser abilities to think abstractly and outside narrow mental confines of datapoints as descriptors of the messy real world.

Focus on the collection and inputting of clinical data or struggling with order entry and documentation further dehumanizes patient care. The interaction of healthcare providers and patients is perhaps the most intimate of relationships outside personal or familial ones. This therapeutic relationship is crucial to providing care, whether for cure or for comfort.

To those whose most intimate relationships are with metal boxes with flashing lights, this issue may also be hard to comprehend fully.

An episode of patient care is now primarily defined by interaction with the computer, which detracts from the provision of care. This experience has been corroborated in a recent study from Calgary, documenting that healthcare workers operating in an EHR environment spend up to 90% of their time during clinical work on the computer rather than with the patient.

I have seen this myself while visiting friends in hospital - banks of nurses lined up against a wall of computer workstations, typing away for the bulk of my visits.    It is - I'm sorry to use this term - nuts, and those who are behind such a system of medical work, managerially speaking, lack both competence and common sense.

... Reliance on order sets developed by experienced practitioners is potentially detrimental for trainees, who do not then have the opportunity to learn the clinical thinking processes that underpin them. Given the reliance on order sets to address inefficiency and safety concerns inherent to the ordering processes in an EHR, physicians at the most crucial stages of learning are not able to develop these critical thinking processes that form the basis for the practice of medicine.

The term of art for this is "deskilling."  I am highly concerned about the quality of young physicians training in cybernetic hospitals, who become automatons without the skills to act appropriately when "off-script" (a rather common occurrence in medicine).

Moreover, given the inefficiency of EHR processes, there is reduced time available for clinical teaching. This has been a universal experience (and complaint) of trainees in Nanaimo. In other jurisdictions, trainees have become effective scribes to unburden staff physicians—an activity I do not believe contributes to the educational experience. Additionally, trainees focus on learning to navigate and use the EHR, rather than attending to patient care or clinical education. We are at risk of producing a generation of physicians with poor clinical skills and who are disconnected from patients.

Medicine is becoming outright enslaved to its cybernetic masters.

The loss of bond with patients is mirrored by disconnection within the healthcare team. EHR processes serve to isolate rather than enhance personal interactions between physicians and nurses, pharmacists and allied health professionals. Healthcare requires collaboration and coordination between many types of providers, a process that is not just crucial to optimal patient care but necessary to support and enhance the performance of individual team members. Disruption of the team approach impairs patient care and has demoralized the healthcare community in Nanaimo.

The social bonds and interactions that hold the complex endeavor of medical care together, in reality, have been usurped by (as I've written before) a grand human subject experiment, without consent of the experimental subjects, namely, clinicians and patients.  Both utopian idiocy and old fashioned opportunism are at work (see "Background On The 'Ecosystem' of Commercial Healthcare IT" at

This has been made worse by the adversarial relationship created when a majority of medical staff who have found continued use of the computerized order management processes unacceptable are opposed by physicians, some allied health professionals and administrators who do not. Some of our most experienced physicians and allied health professionals have retired or changed practice. To say the effects on the health care community in Nanaimo have been destructive is not an understatement. This also means that collaborative work on care needs for our community beyond iHealth has largely stalled.

I believe that the physicians who truly like today's EHRs are in the minority.  As to administrators, if they are not clinicians, they have no business overruling their clinicians on such matters.

What is being eloquently described in this essay is a mass failure of bad health IT that is disrupting medical documentation, medical care (likely resulting in numerous bad outcomes), medical communications, medical education, medical morale, medical esprit de corps, medical recruting and retention, and the health of the community - among others.

Other than those minor drawbacks, everything is fine.

-- SS

1 comment:

Live IT or live with IT said...

This then is no different than the managerialism the infects all of Healthcare.
Docs need to stand up and resist.