Health IT Pundits and Perhaps the Most Logically Fallacious (And Even Cold-Hearted) Statement I've Seen About Health IT to Date

The KevinMD blog has reposted George Lundberg's MedPage Today post "Health IT: Garbage In, Garbage Out", retitled as "Health IT has problems, but is worth the price." I covered Dr. Lundberg's original post at my Nov. 16, 2011 essay "George Lundberg, MD: The Promise of Health IT, and a Caveat."

As the KevinMD blog is exceptionally well-read, I expected the HIT pundits to come in with "see no evil, hear no evil, speak no evil" accolades for the technology.

I was right, even early on.

Keep in mind that Dr. Lundberg specifically quotes me in his article:

... However, there is another harsh critic worth listening to.

His name is Dr. Scot Silverstein, and he seems to have made it his life's work to call attention to really bad problems that he discovers in this mass move to automation.

Heed his cautions. They are real.

My writings and opinions are known of by the following pundit, including the fact that my relative was injured as a result of the technology, who commented on KevinMD's reposting of the Lundberg essay. The comment is here (I do not personally know the commenter, only having exchanged numerous back-and-forth comments on a few health IT blogs in the past):

Margalit Gur-Arie [a partner at EHR pathway, LLC and Gross Technologies, Inc. - ed.]

Wow! There's something to be said for extreme statements, whether right or wrong.

... do EHRs kill people? Probably, but every single item used in medicine can be shown to have killed people at one time or another, depending on how you define "killed" [1]. Do more people get harmed where EHRs are present, compared to where they are not? There are no conclusive studies to that effect and there are no conclusive studies showing the opposite either. There are not very good studies at all, but if mass murder was occurring, we would have probably known by now.

The appeal to ridicule and/or argumentum ad ignorantiam-like statement "if mass murder was occurring, we would have probably known by now" is both fallacious and egregious. Is that a criteria medicine uses, in the explicitly admitted situation where conclusive studies are lacking, to promote diffusion of some new treatment or tool? That is, since we don't note catastrophic levels of toxicity, the toxicity is of minor import?

On other logical fallacies, the statement that "every medical intervention can kill", implying that any morbidity and mortality due to EHRs is just a foregone conclusion, is doubly fallacious.

One fallacy is the absolute nature of the statement itself. It isn't true that 'all medical interventions can kill.' Another fallacy is the cavalier lack of distinction between a small vs. large risk of injury or death.

That said, even without considering 1) the literature aggregated here, 2) the context of the IOM Committee on Patient Safety and Health Information Technology's report that states the technology has risks, and worse, that impediments to information diffusion prevent the magnitude of the risks from being known (PDF available here), and 3) the context of my relative's travails, this is perhaps the most wishy-washy, ethically unsatisfying, cold-hearted excuse for health IT's problems -- and for reneging on fixing those problems before national rollouts -- that I have ever seen.

The argument is so bad, it's difficult to parse out the precise nature of all the logical fallacies contained within.

COI disclosure: I note that I have no associations with, receive no payments or royalties from, or have any other relationships with healthcare IT vendors, consultants or customers. I decided to offer my services as an expert witness for attorneys on health IT-related injuries and records tampering as a result of my relative's travails, however.

Note:

[1] "Depending on how you define "killed"? Let me take a stab at that (it should be easier than defining what the meaning of the word 'is' is). How about "resulted directly or indirectly in the termination of all biological functions, as in, the patient's dead?"

-- SS

BUY IT NOW!

Two Opposing Views of EHR: InformaticMD vs. NextGen's Holder of "American Medical Informatics Certification for Health Information Technology"

The AMA's publication American Medical News recently quoted me following comments from IOM EHR Safety committee member Richard Cook in the Nov. 21, 2011 article "IOM calls for monitoring and probe of health IT hazards" by Kevin O'Reilly:

... Not everyone on the panel agreed with delaying FDA regulation. [Per the IOM report on health IT safety released Nov. 10, 2011, see here - ed.]

Committee member Richard I. Cook, MD, filed a dissent in the report in which he recommended that health IT systems be regulated as class III medical devices.

"It is quite remarkable that we're in this situation," said Dr. Cook, associate professor of anesthesia and critical care at the University of Chicago Pritzker School of Medicine. [Also, an expert in Medical Informatics - ed.] "It's not surprising that such adverse events are being found related to health IT, and it's not surprising that those promoting these systems have neither looked for them nor anticipated them. To make large-scale investments in these systems and only now be looking at the impact on patient safety borders on recklessness."

Scot M. Silverstein, MD, agreed.

"The bone I have to pick with the IOM report is that the action agenda is weak," said Dr. Silverstein, a consultant in medical informatics at the Drexel University College of Information Science and Technology in Pennsylvania.

It is unethical to expand health IT so dramatically without understanding the precise nature of the risks it poses to patients, Dr. Silverstein said.

Ironically, right below my statement was the following from HIT industry figure Charles Jarvis, blaming the user:

Users faulted

Leaders in the health IT industry also had their share of objections to some of the IOM panel's conclusions.

"We don't think there's a great deal of data to substantiate that there are major safety problems with the majority of electronic health records systems in use today," said Charlie Jarvis, executive committee vice chair of the EHR Assn., a trade group that represents 46 organizations that supply most of the EMR systems implemented in medical practices. "These products are safe, dependable, time-tested and display a lot of the safety features we think are necessary to prevent problems going forward."

Jarvis, also a vice president at the health IT firm NextGen, said vendors and the government should work to help physicians and other health professional users understand systems, take advantage of their safety features and avoid errors.

[Charitable translation: computers are infallible, so medical errors due to HIT are the user's fault, the Sept. 2011 National Institute of Standards and Technology (NIST) report on usability be damned. Clinicians should spend their valuable time learning to compensate for and then actually wading through mission hostile user experiences. If only those stupid doctors and nurses would use our cybernetic miracle tools the way we want, the members of the EHR Association could be making even more money. Oh, and by the way, the NIST's concept of "use error" [1] is nonsense. - ed.]

I presume the "EHR Assn." is the HIMSS EHR Association, with HIMSS itself being a gargantuan "cause-based, not-for-profit organization exclusively focused on providing global leadership for the optimal use of information technology and management systems for the betterment of healthcare":

The HIMSS Electronic Health Record (EHR) Association is a trade association of Electronic Health Record (EHR) companies, addressing national efforts to create interoperable EHRs in hospital and ambulatory care settings. The EHR Association operates on the premise that the rapid, widespread adoption of EHRs will help improve the quality of patient care as well as the productivity and sustainability of the healthcare system.

I observe that there are no conflicts of interest here that could cause Mr. Jarvis' stated opinions to be skewed towards the rights of computers and away from the rights of patients ... right?

First, Mr. Jarvis makes a logical error related to the error illustrated in my earlier post today "Magical Thinking on Health IT from ModernMedicine.com." His error is that of "proof by lack of evidence" [2]. No need to actually study the issue rigorously, despite repeated risk management-relevant incident reports (as opposed to the industry's preferred and highly erroneous term "anecdotes").

Just one recent, highly alarming example of an "anecdote" affecting probably tens of thousands of patients due to programming malpractice and grossly negligent quality assurance, at both vendor and end user hospitals, is illustrated here. Since it's an "anecdote", perhaps Mr. Jarvis would agree there's nothing to see there, so we should all move along.

(See the Aug. 2011 post "From a Senior Clinician Down Under: Anecdotes and Medicine, We are Actually Talking About Two Different Things" which puts the misuse of the "anecdotes" label in its proper place - the garbage can.)

Not only is "proof by lack of evidence" in the face of hair-raising incident and defects reports (e.g., as in FDA's MAUDE database) a prima facie logical fallacy unfitting in medicine, and in fact alien to medical ethics, but the IOM report specifically stated in no uncertain terms that nobody really knows the magnitude of the risks. This is due in part to numerous inhibitory factors in evidence diffusion. From the IOM report:

... Several reasons health IT–related safety data are lacking include the absence of measures and a central repository (or linkages among decentralized repositories) to collect, analyze, and act on information related to safety of this technology. Another impediment to gathering safety data is contractual barriers (e.g., nondisclosure, confidentiality clauses) that can prevent users from sharing information about health IT–related adverse events. These barriers limit users’ abilities to share knowledge of risk-prone user interfaces, for instance through screenshots and descriptions of potentially unsafe processes. In addition, some vendors include language in their sales contracts and escape responsibility for errors or defects in their software (i.e., “hold harmless clauses”). The committee believes these types of contractual restrictions limit transparency, which significantly contributes to the gaps in knowledge of health IT–related patient safety risks. These barriers to generating evidence pose unacceptable risks to safety.

Imagine such a situation in, say, the pharmaceutical, automotive, aviation, or nuclear power industries. The responsible individuals would likely be hauled off to jail.

However, this all may be irrelevant. After all, who can argue with the expert personal opinion of someone who holds "the American Medical Informatics Certification for Health Information Technology?" That astonishing credential could conceivably elevate Mr. Jarvis' opinion over all others - even mine, with my meager background in the domain.

I don't know if he still claims that credential, but he did as I described in my post about prior interactions with Mr. Jarvis and NextGen (dating back to 2004) in my Feb. 2009 post "NextGen and Vendor/Doctor Dialog: Yet Another Patronizing EHR Company of Certified HIT Experts?"

I guess the fact I'd never heard of such a qualification represents my dearth of familiarity with the field of Medical Informatics and healthcare information technology.

-- SS

Notes:

[1] “Use error” is a term used very specifically by NIST 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, misuse of color-coding conventions, omission of information, etc. From "NISTIR 7804: Technical Evaluation, Testing and Validation of the Usability of Electronic Health Records." It is available at http://www.nist.gov/healthcare/usability/upload/Draft_EUP_09_28_11.pdf (PDF).

[2] Example of proof by lack of evidence, courtesy Scott Adams: "I've never seen you drunk, so you must be one of those Amish people. "

More on these issues is at the site "Contemporary Issues in Medical Informatics: Common Examples of Healthcare Information Technology Difficulties."

BUY IT NOW!

How Academic and Government "Anecdotes Are Not Data" Ideologues Kill People

I'm already receiving comments that, regarding Prof. Jon Patrick's detailed exposé of the dangers of ill-suited-for-purpose ED EHR's, Patrick's observations are:

... not really valid because they're not peer reviewed; they're just anecdotal.

Only an egghead could pen such words.

I always get hives immediately after eating strawberries. But without a scientifically controlled experiment with all the right peer review, it's not reliable data. So I continue to eat strawberries every day, since I can't tell if they cause hives.

I'd already written about anecdotalist refrains at my Mar. 7, 2011 post "Australian ED EHR Study: Putting the Lie to the Line "Your Evidence Is Anecdotal, Thus Worthless" Used by Eggheads, Fools and Gonifs." In that essay I cite Dr. Patrick himself on "anecdotal evidence", regarding which he hit the ball out of the Southern Hemisphere in an editorial in "Applied Clinical Informatics" entitled "The Validity of Personal Experiences in Evaluating HIT."

Aside from the fact that eggheads also don't seem to care about the issues of faulty peer review, especially in profitable biomedical sectors, such as at "The Lancet Emphasizes the Threats to the Academic Medical Mission" with its embedded links, and "Has Ghostwriting Infected The Experts With Tainted Knowledge, Creating Vectors for Further Spread and Mutation of the Scientific Knowledge Base?", there's this simple fact:

Public health catastrophe warnings from responsible sources don't need peer review, they need investigation.

Yes, there were those pesky, off-narrative journalistic reports that the Japanese nuclear reactors were not entirely safe, that Bernie Madoff was a fraud, that mortgages for everyone was not a good idea, that the O-rings couldn't stand sub-freezing temperatures, that the the foam that broke off the Columbia launch tank caused a danger, and that the Titanic didn't have enough lifeboats, but they weren't peer reviewed...so we ignored them. Saved us a lot of money, too.

-- SS

Addendum:

At my post "Real" Medical Informatics: What Does a Problem List of Typical Health IT Look Like, Part 2", I opined:

If the purpose of Medical Informatics is the improvement of healthcare (as opposed to career advancement of a small number of academics through publishing obscure articles about HIT benefits while ignoring downsides in rarified, echo-chamber peer reviewed journals), then:

• Who are the "real" medical informatics specialists, and;
  

• Who are the poseurs?

... researchers like Jon Patrick who address real-world issues of great import to patients on HIT risks, and further go public on the web with their work without the full blessings of some dusty journal (and those like Ross Koppel who also directly address the downsides, and others who make available to the public material such as on blogs like this and this, papers like this and sites like this) are the former.

Those who deem only "peer reviewed" articles worthy of daylight, and everything else - especially and particularly reports of downsides - "anecdotal" (the anecdotalists) are the latter.

I stand by this assertion.

Finally, I ask: at what point does ignoring work such as Prof. Patrick's, if patient harm is caused by the system he reviewed, constitute reckless endangerment and perhaps criminal negligence by hospital and government officials?

-- SS

Addendum Mar. 23:

As if on cue, this story appeared in the WSJ:

March 23, 2011

Japan Ignored Warning of Nuclear Vulnerability

TOKYO—Japanese regulators discussed in recent months the use of new cooling technologies at nuclear plants that could have lessened or prevented the disaster that struck this month when a tsunami wiped out the electricity at the stricken Fukushima Daiichi power facility.

However, they chose to ignore the vulnerability at existing reactors and instead focused on fixing the issue in future ones, government and corporate documents show. There was no serious discussion of retrofitting older plants with the alternative technology

I guess the "vulnerability reports" just weren't peer reviewed, therefore meaningless - or not reviewed by the "right" peers.

This sounds like our own FDA, ONC office and Institute of Medicine (via the Committee on Patient Safety and Health Information Technology), "choosing to ignore" health IT "vulnerabilities" (such as the aforementioned) and focusing on future issues such as comparative effectiveness research, "the common good", etc. instead.

I call this attitude "reckless endangerment" and hope plaintiff attorneys are paying close attention.

-- SS

BUY IT NOW!

IOM Committee on Patient Safety and Health IT, Meeting Two: Institute of Medicine, or of Mediocrity?

In my Jan. 2011 post "Institute of Medicine Committee on Patient Safety and Health Information Technology, and Thoughts on Social Aspects of Health IT Evaluation" I wrote that:

The U.S. National Research Council of the National Academy of Sciences issued a report in early 2009 on the state of health IT. That study's report, led in part by pioneers in Medical Informatics G. Octo Barnett and William Stead, was entitled "Computational Technology for Effective Health Care: Immediate Steps and Strategic Directions" (pre-publication PDF available free at this link). The report was announced under the following header:

CURRENT APPROACHES TO U.S. HEALTH CARE INFORMATION TECHNOLOGY ARE INSUFFICIENT

The insufficiencies were largely in the areas of difficulties with data sharing and integration, deployment of new IT capabilities, large-scale data management, and lack of cognitive support by health IT for busy clinicians.

One might reasonably conclude such deficits could affect patient safety.

Recently the Institute of Medicine (the health arm of the National Academy of Sciences) formed a Committee to study health IT safety. It held its first meeting on Dec. 14, 2010 (quite a few years late in my opinion, and only after tens of billions of dollars have been earmarked for health IT, but better late than never):

The Institute of Medicine Committee on Patient Safety and Health Information Technology is holding its first meeting on December 14-15, 2010. The first day, December 14, 2010 beginning at 10:30 am, is open to the public to observe the committee proceedings. The committee will hear presentations by the Office of the National Coordinator and other invited guests. There will also be an opportunity for members of the public and representatives of interested organizations to make a brief statement before the committee. Prior registration is requested for attendees and required for those wishing to make a statement.

Here are links to the PPT presentations from Meeting 2 of the Committee on Patient Safety and Health IT that took place Feb. 24, 2011:

http://www.iom.edu/~/media/Files/Activity%20Files/Quality/Patient%20Safety%20and%20HIT/Meeting%202/Dwork.pdf

http://www.iom.edu/~/media/Files/Activity%20Files/Quality/Patient%20Safety%20and%20HIT/Meeting%202/WoodsNormanFeb2011.pdf

http://www.iom.edu/~/media/Files/Activity%20Files/Quality/Patient%20Safety%20and%20HIT/Meeting%202/Harper%20IOM%20HIT%20Patient%20Safety.pdf

http://www.iom.edu/~/media/Files/Activity%20Files/Quality/Patient%20Safety%20and%20HIT/Meeting%202/Chrisman-.pdf

http://www.iom.edu/~/media/Files/Activity%20Files/Quality/Patient%20Safety%20and%20HIT/Meeting%202/Palmer.pdf

The PPT's can be downloaded directly from these links.

I note several observations:

• The overall quality of these presentations appears mediocre;

• Issues of healthcare IT risks - as they exist on the ground in 2011 - are addressed poorly if at all;
  

• Proposed "solutions" are really nothing novel or new compared to existing literature or recommendations made in earlier studies, including that of the US NRC;

• That these presentations come from the highest scientific body in the United States is, in my opinion, a disappointment and, indeed, an embarrassment.

The IOM's rules of engagement, according to the Study Director, preclude my testifying, as a Medical Informatics specialist and former CMIO, about a relative's nearly being killed by poorly designed and implemented health IT. Instead, the linked presentations above are presented.

Here's an example of what I consider a somewhat rigorous and critical thinking-based presentation on health IT risks:

http://www.ischool.drexel.edu/faculty/ssilverstein/Clinical_IT_benefits_risks.ppt

I think the IOM should be able to do better than a mere small-university medical informatics adjunct professor.

-- SS

BUY IT NOW!