A Message for Xerox: Americans Not 'Resistant to Change'; They Are Resistant to Reckless Change That Endangers Them

A press release from Xerox Healthcare Provider Solutions:

Only 26 Percent of Americans Want Electronic Medical Records, Says Xerox Survey

The subtitle is a rhetorical question:

When it comes to healthcare, are Americans resistant to change?

ROCHESTER, N.Y. – Americans routinely use electronic files to manage their finances, communicate with friends and family and even take college courses – but when it comes to medical records – only 26 percent want them digital. The findings come from the third annual Electronic Health Records (EHR) online survey of 2,147 U.S. adults, conducted for Xerox (NYSE: XRX) by Harris Interactive in May 2012. 

According to the survey, only 40 percent of respondents believe digital records will deliver better, more efficient care. That response fell two percent from last year’s survey, and matches the response reported in 2010. Overall, 85 percent of respondents this year expressed concern about digital medical records.

Americans have a healthy skepticism of putting their private information online to be hacked, of the disruptive effects of today's commercial health IT on their clinicians, and the costs of doing so.  This likely comes from common sense, reading and observation.

As one reader of this blog wrote:

Xerox kindly shared all three years of their annual Electronic Health Records (EHR) online surveys by Harris Interactive. The media, industry and government unrelentingly promote health technology as the latest, greatest best stuff. But the public ain’t buying it. They want smart phones, but they don’t want EHRs.

The Xerox article paternalistically continues:

“We continue to see a resistance to change from consumers – meaning providers need to continue to educate Americans on the value of EHRs,” said Chad Harris, group president, Xerox Healthcare Provider Solutions.

(Note the use of the 'EHR' acronym.  As I've written, the acronyms 'EHR' and 'EMR' are anachronisms used to describe what are no longer innocuous filing systems, but greatly intrusive enterprise clinical resource and workflow control systems.  I think the public increasingly understands that.)

This patriarchal statement by Chad Harris about "resistance to change" by "healthcare consumers" needing re-education is, in a word, depraved, because I think the person uttering the sentence knows better.  Let's define depraved:

Depraved (adj.):  morally bad or debased; corrupt; perverted

This statement implies is that health IT is a perfected technology without significant flaws, whose benefits are well-proven and whose drawbacks and risks are well understood.

Unfortunately, none of those are true.  From my May 2012 post on ONC's embarrassing "Health Data Palooza", worth repeating here, with hyperlinks:

  • There is a markedly unscientific "irrational exuberance" pushing clinical IT into wide use at a dangerously rapid pace. This exuberance is contradicted by a growing body of literature that shows the benefits are likely far less than stated, e.g., by way of example, the ad-hoc set at http://www.ischool.drexel.edu/faculty/ssilverstein/cases/?loc=cases&sloc=readinglist;
  • The technology remains experimental, its rollout is a human subjects experiment on a massive scale lacking nearly all the protections of other human subjects experimentation and for IT in mission critical settings (e.g., informed consent, formal quality control/validation/regulation, formal postmarket surveillance and reporting) due to extraordinary legal and regulatory special accommodations afforded the technology and its purveyors;
  • Defects of in-use systems are rampant, inappropriately turning patients and clinicians into software alpha and beta testers (e.g., as in the voluntary FDA MAUDE database, http://hcrenewal.blogspot.com/2011/01/maude-and-hit-risk-mother-mary-what-in.html which contains information for just one HIT vendor, Cerner, who voluntarily reports such issues);
  • The technology is unsupportive of clinician cognitive needs (2009 National Research Council study, which also stated that accelerating interdisciplinary research in biomedical informatics, computer science, social science, and health care engineering will be essential to perfect this technology);
  • The roles of scientific discovery and anecdote have been turned on their heads. RCT's of clinical IT are nearly non-existent and lower-level evidence (e.g., weak observational, pre-post, qualitative, and other study types) are cited as "scientific proof" of efficacy and safety justifying hundreds of billions of dollars of taxpayer (or is it Chinese loan?) expenditures.  Yet, risk management-relevant case reports of harmful events and near misses, crucial to help organizations and regulatory agencies understand risks are dismissed as "anecdotal" (e.g., Blumenthal: "The [ONC] committee [investigating FDA reports of HIT endangement] said that nothing it had found would give them any pause that a policy of introducing EMR's could impede patient safety," he said, while ONC issued an article based on questionable research methods entitled "The Benefits Of Health Information Technology: A Review Of The Recent Literature Shows Predominantly Positive Results" extolling the virtues of HIT, written about at http://hcrenewal.blogspot.com/2011/03/benefits-of-health-information.html).
  • Risks are definite, with known patient injury and death, but the magnitude is admittedly unknown as admitted by JC (2008 Sentinel Event Alert), FDA (2010 Internal memo on HIT risks and statements of Jeffrey Shuren MD JD about known harms likely being "the tip of the iceberg"), IOM (2011 report on HIT risk), ECRI Institute (Top ten healthcare technology hazards for 2011 and 2012), NORCAL Mutual Insurance Company 2009 report on EHR risks, others;
  • Existence of severe impediments to information diffusion about risks explicitly admitted by FDA (2010 memo), IOM (2011 report), others;
  • Usability of commercial products in real world settings is often poor (e.g., NIST 2011 study on usability), promoting "use error" (user interface designs that engender users to make errors of commission or omission, where 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.)
  • These systems promote capture and display of clinically irrelevant information in the interest of charge capture, and result in reams of "legible gibberish" with many negative characteristics that make it difficult for other clinicians and reviewers to establish a cohesive, definitive narrative of clinical events and timelines.

The article continues:

Despite consumers’ misgivings of the value of EHRs, caregivers [largely hospitals and healthcare systems who force it on their staffs and owned physician practices - ed.] are quick to adopt digital technology.  [Thanks to incentives and looming penalties - ed.] When asked how their healthcare provider recorded medical information during their last visit to a doctor or hospital, 60 percent of respondents – who have visited a doctor or hospital – reported that the information was entered directly into a tablet, laptop or in-room computer station versus 28 percent who reported the information was taken via handwritten notes. 

As the numbers don't justify the practice, even if the numbers as stated are valid, my response is:

So what?

To help caregivers do more with this patient information, Xerox is working with researchers at PARC, A Xerox Company, to explore EHRs as a gateway to a variety of healthcare innovation possibilities. The resulting technology tools will simplify back-office and front-line processes, reduce errors, and free up caregivers to spend more time and attention on day-to-day patient care.

The "possibilities" will accomplish all these things?  

Not only does that not follow logically, but where's the data, or is this simply wishful thinking? 

The latter "possibility that will come true" - "free up caregivers to spend more time and attention on day-to-day patient care" - is the most laughable.   These systems do anything but, as for example here regarding the the time costs of data acquisition and the time costs of data input.  I have yet to see serious studies that consistently demonstrate any time savings at all for clinicians.  Quite the reverse, actually.

Of course, the mandatory marketing puffery:

“A big part of PARC’s healthcare work for Xerox is using ethnography and other social science methods to observe and analyze actual work practices – not just what people say they do,” said Steve Hoover, CEO, PARC, A Xerox Company. “If there’s one thing that this survey tells us, coupled with our own experiences, it’s that you should never develop or deploy technology outside of the human context.”

Precisely what is being done now, and on a national scale in the 'National Program for IT in the HHS.'

Xerox, either you're part of the solution or part of the problem.

Which is it?

-- SS