Wednesday, May 06, 2015

Politico: Electronic record errors growing issue in lawsuits

The indefatigable Arthur Allen of has authored a nice piece on the issue of EHRs being a cause of medical malpractice, with resultant litigation.  I was a contributor:

Electronic record errors growing issue in lawsuits
By Arthur Allen
5/4/15 6:40 AM EDT

Medical errors that can be traced to the automation of the U.S. health care system are increasingly an issue in medical malpractice lawsuits.

Some of the doctors, attorneys and health IT experts involved in the litigation fear that safety and data integrity problems could undercut the benefits of electronic health records unless HHS and Congress address them aggressively.

I already believe the benefits of EHR technology have been severely undercut - if not destroyed - by the unbelievably poor quality, user experience and incompetent implementations presented by most commercial health IT software today.

I even have a dead mother to offer as evidence, due to an ED EHR's lack of fundamental and crucial confirmation dialogs and notification messages to team members.  These computer science-101 level deficiencies permitted a triage nurse's failure to successfully click a heart medication for continuation to propagate through several days of ICU/floor hospitalization unnoticed.  Gross overconfidence in computer output and cavalier attitudes in the ICU about med reconciliation sealed the deal, where, recognizing something seemed amiss with the meds list vs. the history of arrhythmia, the ICU doctor did nothing except leave a question in the chart about it, resulting in catastrophe.

“This is kind of like the car industry in Detroit in 1965,” says physician Michael Victoroff, a liability expert and a critic of the federal program encouraging providers to adopt EHRs. “We’re making gigantic, horrendous, unsafe machines with no seat belts, and they are selling like hot cakes. But there’s no Ralph Nader saying, ‘Really?’”

There are, actually, but an "Unsafe at any MHz" has not yet been written and taken seriously by the public about EHRs.  The industry has been too in control of the narrative for that to happen.

According to a review by The Doctors Company, the largest physician-owned U.S. medical malpractice insurer, EHR issues were involved in only 1 percent of a sample of lawsuits concluded from 2007 through 2013. But that finding could be deceptive since it takes five or six years to close a suit, and during that period the numbers of such cases grew rapidly as electronic health records become more pervasive in hospitals and physician offices. The pace of these cases doubled from 2013 to early 2014.

At the linked report at the med mal insurer "The Doctor's Company" noted:

... Shortly after electronic health records (EHRs) began to be widely adopted, The Doctors Company and other medical professional liability insurers became aware of their potential liability risks. We anticipated that EHRs would become a contributing factor in medical liability claims. Due to the three- to four-year lag time between an adverse event and a claim being filed, however, EHR-related claims have only recently begun to appear. In 2013, we began coding closed claims using 15 EHR contributing factor codes (eight for system factors and seven for user factors) developed by CRICO Strategies for its Comprehensive Risk Intelligence Tool (CRIT).

In 2013, The Doctors Company closed 28 claims in which the EHR was a contributing factor, and we closed another 26 claims in the first two quarters of 2014. During a pilot study to evaluate CRICO’s EHR codes, 43 additional claims closed by The Doctors Company were identified (22 from 2012, 19 from 2011, and 2 from 2007–2010).

What is not stated is the fact that many EHR-related harms are not recognized as such; many injured patients do not sue, and many who want to cannot do so due to the expense and time involved for plaintiff's attorneys (I have heard the figure that perhaps 5% do make it to suit).  Along with the time lag noted, these figures are another Red Flag, as are the ECRI Deep Dive harm figures, representing what is likely just the "tip of the iceberg."

(See "Peering Underneath the Iceberg's Water Level" at and "FDA on Health IT Adverse Consequences: 44 Reported Injuries And 6 Deaths, Probably Just 'Tip of Iceberg'" at

Back to Politico:

The lawsuits allege a broad range of mistakes and information gaps — typos that lead to medication errors; voice-recognition software that drops key words; doctors’ reliance on old or incorrect records; and nurses’ misinterpretation of drop-down menus, with errors inserted as a result in reports on patient status.

In addition, discrepancies between what doctors and nurses see on their computer screens and the printouts of electronic records that plaintiffs bring to court are leading some judges and juries to discredit provider testimony and hand out big awards. In one case, a patient in septic shock had suffered gangrene and a severe skin rash, but computer records read “skin normal.” They also showed repeated physician interviews with the patient — when she was comatose.

I can verify both of these issues personally, from my legal work - not to mention outright electronic record tampering.

... While the percentage of EHR-related cases is still low, “this is going to become a bigger and bigger issue,” said David Troxel, medical director of The Doctors Company. “I get more calls from frustrated, angry doctors about their EHRs than any other subject.”

Doctors may be following my advice (see end of my post at where I wrote:

... When a physician or other clinician observes health IT problems, defects, malfunctions, mission hostility (e.g., poor user interfaces), significant downtimes, lost data, erroneous data, misidentified data, and so forth ... and most certainly, patient 'close calls' or actual injuries ... they should (anonymously if necessary if in a hostile management setting):

(DISCLAIMER:  I am not responsible for any adverse outcomes if any organizational policies or existing laws are broken in doing any of the following.)

  • ... Inform their personal and organizational insurance carriers, in writing. Insurance carriers do not enjoy paying out for preventable IT-related medical mistakes. They have begun to become aware of HIT risks. See, for example, the essay on Norcal Mutual Insurance Company's newsletter on HIT risks at this link. (Note - many medical malpractice insurance policies can be interpreted as requiring this reporting, observed occasional guest blogger Dr. Scott Monteith in a comment to me about this post.)
Back to Politico:

The industry “takes very seriously the need to enhance the well-documented ability of EHRs to increase patient safety,” an association spokesperson said. “It also recognizes the importance of looking for opportunities to identify and reduce any potential risks associated with development and use of EHRs. All these efforts are essential to the goal of learning more about the ways in which technology, training and configuration can be rolled out in the safest possible ways.”

This is pure B.S. and spin.  There is no "well-documented ability of EHRs to increase patient safety", just a number of methodologically flawed/biased studies (like this one,, a lot of pro-HIT rhetoric, and a lot of harms data that the industry ignored for many years.

The statement also ignores what the domain expert end users - physicians and nurses - themselves are saying, see "Accenture - Fewer U.S. Doctors Believe IT Improves Health Outcomes (April 2015)" at

Why, one should ask as well, should an industry that's been around for decades only now be "learning more about the ways in which technology, training and configuration can be rolled out in the safest possible ways?"  

I see that statement as an industry self-condemnation of years of cavalier IT practice. 

... But providers and health care systems are eventually going to start suing vendors, analysts said, in part because software companies are viewed as having deep pockets. “It’s only a matter of time before a company like athenahealth or Allscripts or Epic or Cerner gets sued,” said Klein.

Plaintiffs’ attorneys are already eyeing such cases, according to Scot Silverstein, a Drexel University health IT expert and internist who is suing a hospital over a lapse in care of his mother that Silverstein claims was caused by poor EHR implementation. Silverstein and two plaintiff’s attorneys met with Rep. Matthew Cartwright (D-Pa.) and other lawmakers in November to plead for more government regulation of EHRs.

I was actually more direct with the author of the article, Mr. Allen.  I said that the sellers of these systems deserved to be sued - that they had earned it through slovenly practices in thought and application enabled by the extraordinary regulatory accommodations afforded to and protected by the industry since its inception (i.e., no regulation) - when system flaws result in patient harm. 

... Some recent studies show that EHRs do make hospitals safer. But the data isn’t conclusive, said William Marella, executive director of the ECRI Institute Patient Safety Organization. Last year, ECRI convened a partnership of EHR vendors, safety experts, academics and medical groups to share and analyze health IT problems ... EHR safety issues are frequently misdiagnosed — and thus under-diagnosed — by providers, according to ECRI’s Marella. “They say, ‘wrong site surgery,’ or ‘drug error,’ which can make it hard to ferret out the cases where IT is responsible.”

It is absurd and disingenuous to speak of 'safety improvements' when the true harms rates are admittedly unknown (see "FDA on health IT risk:  "We don't know the magnitude of the risk, and what we do know is the tip of the iceberg, but health IT is of 'sufficiently low risk' that we don't need to regulate it" ( for more on that issue.

... In about 200 EHR-related legal cases that the liability firm CRICO analyzed, the glitches rarely led directly to patient harm, said Dana Siegal, the company’s director of patient safety services. But she added, “We’re seeing failures to communicate or providers acting on inaccurate information that was driven in part by an EHR issue.”

This brings up another issue.  Computers don't pull triggers and the mayhem they cause doesn't usually immediately kill people.  My mother, for example, survived the initial EHR-led assault on her life, though her survival required emergency reversal of anticoagulation in the face of critical carotid stenosis that had already caused a TIA, emergency craniotomy (brain surgery), and other risky interventions.  She died a year later of complications of her injuries.  Her case would not be likely to be counted as an "EHR-related death" in any statistics (if it was reported at all).  This "time delay" would likely cause any statistics on EHR-related deaths to be understated on their face.

... While the effect of EHRs on malpractice suits is still modest, many analysts worry about the overall uncertainty concerning information in such systems. Confusing or inaccurate records, if they proliferate, not only cast doubt on a doctor in court but could taint clinical research that draws on these large pools of data.

Bravo to Arthur Allen for pointing out that clinical research that draws on garbage, uncontrolled data will turn out garbage.  See my paper "The Syndrome of Inappropriate Overconfidence in Computing" at for more on that issue where I observed:

... This increasing confidence in EHR data to perform far more complex tasks than postmarketing surveillance of a single drug is of great concern. Prompt detection of adverse drug events (ADEs) from single drugs, using aggregated EHR data, is within the realm of possibility. Detection of relatively more nebulous (i.e., compared to major ADE) “outcomes differences” between two or more drugs or treatments via EHR data—such as, did treatment A lower blood pressure more than drug B, or did drug C lessen depression more than drug D—rises to the level of “grand overconfidence in computing” and perhaps “grand folly.”

To accomplish this task with reasonable scientific certainty from aggregated EHR data originating from different vendor systems, input by myriad people of different backgrounds, with differing interpretations of terminologies (students/residents/attending
MDs/RNs etc), under different pressures and motivators (time limits, cognitive overload from poor HIT user interfaces, reimbursement maximization, and so forth), seems improbable.

What levels of statistical validity could arise from such studies? Could they even approach the level deemed “acceptable” in good science?We do not know, although I suspect a “garbage in, garbage out” (GIGO) phenomenon, leading to studies whose results are more likely related to chance than to solid reality.

Back to Politico:
A recent report by the HHS Office of Inspector General said the department has failed to assure that EHR data are secure and accurate. Many hospitals have unsecured audit trails—meaning that information in the record could be altered without detection, it said.

This ability - to alter records - has been admitted under oath by EHR systems administrators in cases in which I have been an advisor to the Plaintiffs' attorney - and had the questions asked.  Hospitals have possession and control of easily-alterable information (far easier than paper) that is the only evidence of potential misadventures - a major conflict of interest.

... “There’s really no one with a vested interest in the integrity of the record besides you, the patient,” said Reed Gelzer, a physician and health IT expert.

Yes, but getting it can cost thousands of dollars.  I paid about one thousand dollars for a few reams of printed EHRs, and this is not uncommon.

... Concerns led the Institute of Medicine in 2011 to propose the creation of a dedicated IT safety center with the power to investigate EHR risks. ONC has since settled on a center that would have no investigatory power but would provide a safe environment in which real-life problems could be analyzed and solutions developed.

In other words, ONC has settled on a joke, on tension management.

The safety center is a “critical priority right now for ensuring the transformation from a world of paper to a world of electronics and connectedness,” said Patricia McGaffigan, COO and a senior vice president at the National Patient Safety Foundation.

Yes, a powerless, industry-friendly "safety center" is a priority for good reason.  Not to sound mean-spirited, but if the wife, child, parent of one of the proponents of a milquetoast "Safety Center" suffers EHR-related injury and dies a painful and lingering death as a result, as did my mother, the proponent will get little sympathy from me.   They've certainly tempted the Gods.

I will leave the final word to physician Michael Victoroff:

“The vendors are very right that if they had true product liability they wouldn’t make these things,” he said.

To which I say to the Information Technology vendors, if you can't take the heat (of accountability in medicine), then get the hell out of our medical kitchen.

-- SS


Anonymous said...

When you use an EHR, rest assured that you are not practicing at the standard of care that existed a decade ago.

Anonymous said...

Amen anonymous ^^^^^

And thank you scot. Keep it up.

Anonymous said...

The Ralph Nader approach probably wouldn't work today. Maybe a reality TV show would.

Live IT or live with IT said...

Once the US Govt gets its head out of the sand maybe they can claw back some of them money to help fix this mess. I'm willing to help.