Sunday, May 30, 2010

Did EPIC CEO Judy Faulkner of Epic declare that ‘healthcare IT usability would be part of certification over her dead body?'

At the HisTALK blog 5/31/10 update, a site with thousands of readers involved in all aspects of health IT, the following anonymous (at this point) report appeared:

From Tabula Rosa: “Re: EMR usability. At one of the ONC Policy Committee meetings, [founder and CEO] Judy Faulkner of Epic supposedly declared that ‘usability would be part of certification over her dead body.’ I wonder if she has similar sentiments about making software accessible to people with disabilities?” Unverified. This inspired my new poll question – keep reading below.

Epic Systems Corporation is one of the largest health IT vendors in the U.S.

If this report is true, it would have very, very serious implications towards the healthcare IT industry's attitudes about the usability - and ultimately the safety - of its products.

The essential nature of usability and the relationship to safety is not an issue for debate - period. It was settled long ago in relation not only to IT, but also to devices and machinery that use IT. For instance, see the mid 1980's wisdom written for the U.S. Air Force on user interfaces at my eight part series on mission hostile health IT user experiences at Microsoft and other responsible IT vendors spend billions on usability research.

I think it incumbent of the U.S. Office of the National Coordinator for Health IT, and its leader Dr. David Blumenthal, to confirm or deny that such a statement was made, on or off-record, at an ONC policy committee meeting, and if it was made, its exact context.

If it was made, it would raise the following questions:

  • Would its utterer be saying this out of concern that their products have usability deficiencies that competitors do not, and that are not correctable in a short time frame?
  • How much concern would such a statement indicate for the complex jobs of physicians, nurses and other clinicians whose safe conduct of their medical obligations depends on health IT?
  • How much concern would such a statement imply towards the injured and dead bodies of patients affected by health IT of deficient usability?
  • Since the CEO sets the tone for their company and has the ultimate responsibility for its conduct (and its Board of Directors as well), would such a statement reflect willful negligence and perhaps criminal conduct?

The last question is pertinent, for if, say, the CEO of Boeing were to relate that "usability of the control systems of its aircraft would be an FAA certification requirement over his or her dead body", or the CEO of oil drilling companies said same about their instrumentation that helps avoid underwater blowouts, I believe it would rise to the level of criminally negligent conduct.

This is all the more pertinent to me, as my relative was just seriously injured by a medication error that appears to have started with the med list in an EMR.

I am seriously wondering, in fact, if the average hospital has the gravitas and political and technologic maturity to truly implement health IT so that the technology is uniformly beneficial, instead of spontaneously harmful, to meet the needs of the grandiose plans for national health IT.

The lack of seriousness, petty political fights, and technological dyscompetence and incompetence I have noted in my 20 years in this field do not seem surmountable in any time frame I consider "soon."

Here seems to be the health IT industry and hospital MIS department view of healthcare IT, as some sort of game to be played for profit, power and political advantage:

-- SS


Anonymous said...

Despicable conduct by the industry. The CEOs ought be investigated with criminal charges in mind.

taotechuck said...

I've worked with Epic. Their user interface is a bit better than most, but still awful.

Eclipsys is apparently working with some design firm to re-engineer their UI. Curious to see what (if anything) comes from that, particularly since the code behind Eclipsys is garbage.

With that said, it's easy to point fingers at health IT and let the rest of the healthcare industry off the hook. Clinicians and administrators have (rightly) taken to heart the philosophy of "don't computerize a broken process." As such, in addition to moving paper-based processes to computers, we're redesigning these processes to be "better." The problem is, even a perfect UI is not going to be able to document a patient encounter in a meaningful and efficient way that is visible in one screenview and provides the level of documentation to meet legal and regulatory requirements (let alone ward off potential lawsuits).

IT is profit-driven and delivers sub-par products, and clinicians/administrators have not yet figured out a meaningful baseline level of what data needs to be captured. Until these two issues are resolved, there will be no change.

Scot, I read your posts all the time, and one of my biggest beefs is the way you've shifted the blame to IT. I agree that IT has completely screwed the pooch with existing EMR/CPOE systems, but clinicians and administrators are right there, holding the pooch while the IT team screws it.

Personally, I think a not-for-profit health IT firm is one possible solution, assuming it's run by ethical and intelligent people (who, frankly, are a rare breed in health IT). It would completely disrupt the field of healthcare IT, as the business model would enable the organization to place quality above profits (something no existing health IT firm can say).

Given the amount of ARRA money that's available, now is an unprecedented time to do this. Any takers out there?

InformaticsMD said...

I dole out blame where it is due.

Anonymous said...

Administators, in bed with the vendors, pay for loyalty, establishing a barnyard of physician do IT yourselfers "champions" who smother their clinicians with propaganda and threaten them with "disruptive" conduct sham peer review if they complain about the "pooch screw". There is not any recourse for the clinical staff who are struggling to come up with workarounds to protect their patients from the crap devices being sold as CPOE and EMRs.

If the devices at Harlem hospital were any good, there would have been an indicator that heart studies central to the care of sick patients were lost on the computers:

This happens all of the time with radiographs. 2 weeks after a test, the radiologist calls and asks if I still want him to read the study since it was "lost" for two weeks.

InformaticsMD said...

Bev md, for starters google 'microsoft usability.'

Anonymous said...

Ask the VM ware person to disclose the number of patients who were injured and killed when 8 hospitals went down in New Mexico as published on HIStalk last week:

“Re: Presbyterian Albuquerque. VMware installation caused two major network outages within 48 hours. The clinical system (McKesson) is still trying to recover: interfaces and prod down, pharmacy handhelds all needed hard booting, runaway processes triggered, unable to pull PACS studies, etc. Affected all eight facilities across the state.”

Does any one know if this was reported in newspapers or TV news and if so, where?

Anonymous said...

OK, thanks; so I did google it. So this means the Epic CEO does NOT want to allow her product to be tested on real people to see if it works??!! Or am I missing something?
My mind is kind of blown here....

bev MD

InformaticsMD said...

bev MD wrote:

"OK, thanks; so I did google it. So this means the Epic CEO does NOT want to allow her product to be tested on real people to see if it works??!! Or am I missing something?"

I report, you decide.

I do know that it took this industry decades to admit a problem and to produce this document (PDF) just last year.

As far as testing on "real live people to see if it works", that is already being done all over the country -- and without patient consent.

See my partially tongue-in-cheek post on that issue at "Draft Patient Rights Statement and Informed Consent on Use of HIT."

But heck, just like the Tuskegee experiments, it's all in the name of progress, so no harm done...

-- SS

Unknown said...

"Tuskegee experiments"? You were doing well until that cheap shot.

InformaticsMD said...

It's not a "cheap shot" at all. The literature on HIT dangers is growing and incontrovertible, yet HIT is used w/o patient consent as a grand experiment, with industry profits growing and suppression of accountability and gag clauses contractual (per Koppel and Kreda and my own JAMA letter of July 2009). So, how exactly is the HIT experiment different?

I have used that metaphor with reference to the HIT experiment in the past at this blog:

I find it remarkable how weak-kneed people have become in terms of language, perhaps conditioned by years of PC and "let's go along to get along" pervasive in today's schools.

It being just past Memorial day, thank God we have real men and women fighting our wars.

Anonymous said...

To the extent that the HIT industry in partnership with hospitals have profit goals aligned, the downtrodden and desperately sick patients, are serving as non-consented subjects in the HIT experiments. The only thing different between this and The Tuskegee Project is the race factor. Patients do not know that their care is being altered by devices that have not been approved for safety and efficacy, a violation of Federal Law.

InformaticsMD said...

Those involved in Tuskegee also felt they were working for the 'greater good.'

Anonymous said...

I remain puzzled as to why the feds don't seem to realize that at least parts of EMR's are medical devices and should be regulated as such. There is precedent for this in that the FDA must validate and approve blood bank/transfusion service software and any changes to it, before it can be used in a hospital or blood bank. This is due to the critical nature of its function in assuring the right blood goes to the right patient. So what is different about a pharmacy system, or a CPOE system, or one of numerous other examples?

I keep making this point on different blogs but no one with any actual position on any of these committees seems to acknowledge it - just us poor dumb slobs out in the trenches.....

bev M.D.

Anonymous said...

Bev, MD

Well said, excellent points.

Have you written to Margaret Hamburg, Josh Sharfstein, and Jeff Shuren with your question? And for good measure, send a note to Secretary Sebelius with the same question.

Dose of Reality said...

You say that the CEO sets the tone for their company and I agree.

Judy Faulkner once asked her staff "Who do you work for? Epic, customers or patients? Patients. Patient care always come first."

InformaticsMD said...

Dose of Reality,

I have little interest in, or confidence in, corporate slogans and CEO rah-rah sessions. These slogans, mission statements, vision statements, etc., etc. are just that: words.

Every corporation that has conducted misdeeds always has slogans with wonderful-sounding words.

BP Oil, as just one example.

What matters are actions.

-- SS

Anonymous said...

Dear Bev, MD who wote

"...but I wonder if part of it is a result of automating a bad process - and that is, the decades-old, traditional attitude that a patient transferring from ER (or OR, or ICU, etc.) to floor is essentially "readmitted"..."

You should write to the FDA today and state your opinion. They put their skirts and pants on just as you and Dr. Silverstein do. Follow it up with a phone call to the CDRH, asking for the Office of the Director, and then the Office of the Commissioner.

There is this misleading, and probably erroneous, dogma which defends the crap HIT, that IT causes death because they are automating a bad process.

That could not be further from the truth. If there was a list of medications that was printed for this patient and carried to the floor with her, the life threatening omissions are unlikely to have occurred.

To the extent that doctors and nurses have their cognition disturbed by the pages of irrelevant information roaring at them on a small terminal, they are unable to focus on the salient issues, and trust the computers to be "right" when they can never be trusted.

InformaticsMD said...

I must agree with "Anonymous 10:23:00 AM EDT."

"Automating a bad process" is not the root problem. "Automating a process, whether good or bad, with suboptimal IT" is the problem.

Anonymous said...

I am sorry I inadvertantly stumbled into a code phrase used as an excuse - e.g., automating a bad process. I did not mean that it was an excuse for bad IT, just a complicating factor. I still believe that having to do over all the orders, etc. when the patient is admitted is outdated and inefficient, with or without IT (and we used to do it without IT, remember?)

Having said that, one cannot untangle IT and process. It's the attempts to simultaneously improve process and superimpose complicated and (usually) poorly designed IT on that struggle, that is leading to despair in many hospitals.

bev M.D.

InformaticsMD said...

Re: I am sorry I inadvertently stumbled into a code phrase used as an excuse - e.g., automating a bad process.

It is an excuse, as it tends towards implying that automation is inherently benign and beneficial in its present form.

Correcting "bad process" is fine, but I think first the experimental automation technology's flaws need to be better understood - and fixed.

Anonymous said...

Did you write to the FDA, Bev MD?

There is a fallacy to your statement "I still believe that having to do over all the orders, etc. when the patient is admitted is outdated and inefficient, with or without IT (and we used to do it without IT, remember?)"

Bev MD, this cut and paste will perpetuate mistakes. Because it is in the computer it must be right, right? The key to good medical care is skepticism and critical thinking. The processes being promoted as efficient cost innocent lives by promoting errors. Do you know who entered the orders? Was it a medical student who had the orders cosigned by an over burdened sleep deprived resident???

InformaticsMD said...

Re: Anonymous April 12, 2011 2:51:00 PM EDT

I removed your comment as it contains details that I cannot vouch for or verify, though they sound typical of problems I've been hearing about regarding health IT from varied vendors, from trusted sources.

These stories should be reported to FDA's Manufacturer and User Facility Device Experience (MAUDE) database via a form 3500 at the very least (link), and perhaps emailed to FDA's Center for Device and Radiological Health (CDRH) director Jeff Shuren MD, JD, as well as local and national political representatives, for further investigation.

These issues could represent a risk to the public if true.


Anonymous said...

And this is the woman that Obama wants to put onto his healthcare board?