A large part of the savings projected from “healthcare reform” is supposed to come from wider use of information technology. The federal government is expected to “invest” some $45 billion in encouraging (or compelling) doctors and hospitals to use electronic records systems.

“Information is the lifeblood of modern medicine. Health information technology (HIT) is destined to be its circulatory system,” writes David Blumenthal, M.D., M.P.P., of the Office of that National Coordinator for Health Information Technology (New England Journal of Medicine 12/30/09). "Physicians and institutions trying to practice highest-quality medicine without HIT are like Olympians trying to perform with a failing heart,” he states. (Note that Blumenthal’s numerous financial disclosures are in a separate document.)

In the real world, there are many problems in implementation. The University of California San Francisco Medical Center is one institution that is quietly writing off about a third of the $50 million it has poured into electronic medical records over the past 5 years. The system is still not fully up and running. UCSF terminated its contractor and is prepared to start part of the project from scratch (Huffington Post 11/23/09).

According to one study, between 50% and 80% of electronic health records systems fail. The larger the EHR project, the higher the risk of failure (IEEE Spectrum 1/1/10).

Instructional materials from real institutions include such eye-openers as a complicated 90-page guide for simply entering orders and a 30-page House-Wide Discharge (Depart Process) Training Manual. It is no longer possible to discharge a patient by writing prescriptions and a “discharge today” order in the record. “It’s a wonder clinicians can get any clinical work done at all any more,” writes Scot Silverstein, M.D. (Health Care Renewal 1/3/10).

For more than a decade, Silverstein has been making the case that “health IT is very, very much harder than it looks, especially to those in IT lacking healthcare expertise.” Health IT is still largely a social experiment, and hospitals are a highly risky environment for implementing it.

Paper is far from being technologically obsolete, he notes, citing a December 2009 article in the Milbank Quarterly, “Tensions and Paradoxes in Electronic Patient Record Research: a Systemic Literature Review Using the Meta-narrative Method,” by Greenbaugh et al. of the University College London.

“Paper records, being flexible, portable and tolerant of ambiguity, support the complex work of clinical practice remarkably well…. [H]igh-tech healthcare environments such as intensive care units often make extensive use of paper charts, white boards, sticky notes, and oral communication” (Health Care Renewal 12/15/09).

HIT has become intensely political, note Greenbaugh et al. Publishers need to “invite studies that ‘tell it like it is,’ perhaps using the critical fiction technique to ensure anonymity.”

Silverstein calls the idea that “investment of tens of billions of dollars on a frenetic timeframe” will create massive quality improvements and cost savings “the height of magical thinking and political hubris.”

Specializing in medical informatics, Silverstein is not opposed to HIT, he in fact supports it and dedicated his career to informatics. He is only opposed to HIT that is badly done. He observes that local projects built by experts are far more likely to provide major benefits than extant “shrink-wrapped” and massively expensive HIT.

Numerous serious problems have been reported with HIT in operation. Some prompted an Oct 16, 2009, letter from Senator Charles Grassley (R-IA) to Cerner Chief Executive Officer Neal Patterson.

Sen Grassley wrote: “Over the past year, I have received numerous complaints from patients, medical practitioners and technologies engineers regarding difficulties…with HIT and CPOE devices…. These complaints include faulty software that miscalculated intracranial pressures and interchanged kilograms and pounds, resulting in incorrect medication dosages.”

Sen Grassley also referred to “gag orders” that prohibit disclosure of defects, and lack of a system to monitor performance of these devices.

Experienced systems professionals are increasingly raising concerns about the poor design of electronic medical records (EMRs), which frequently require workarounds and patches. The process is “unsustainable” and could lead to “data breakdowns” (Design Dialogues 11/12/09).

Some physicians like their EMR system, but one senior internist at a major hospital, who feared losing his job if he spoke on the record, reported on one 2006 system that crashed soon after it went online. He struggled to keep patients alive while vendor employees “ran around with no idea how to work their own equipment” (Washington Post 10/25/09).

One study showed that more than one in five hospital medication errors were caused at least in part by computers (ibid.).

Emergency physicians in 200 hospitals in Australia were affected by a system credited with decreasing by 50% the number of patients seen within 20 minutes of arrival. Descriptors included “user hostile,” “dangerous,” and “slow at any task I tried.” Vendors offered “more support.” Clinicians said that was like “giving us a defective car and then sending out someone to show us how to drive it” (Health Care Renewal 10/20/09).

HIT raises serious liability concerns, note Sharona Hoffman and Andy Podgurski of Case Western Reserve University. “EHR [electronic health records] systems cannot remain unregulated and largely unscrutinized. It is only with appropriate interventions that they will become a much-hoped for blessing rather than a curse for health care professionals and patients.”

In an earlier report, these authors concluded that “the advantages of EHR systems will outweigh their risks only if these systems are developed and maintained with rigorous adherence to best software engineering.” Unlike other life-critical medical devices subjected to FDA oversight, EHR systems have not been comprehensively assessed.

The Veterans Administration system of EHRs has been in use since the mid-1990s. While reportedly very successful, a software problem that led to major treatment errors in 2008 is still under review. Though no evidence of harm to any patient was found, “the potential for serious injury was staggering” (Ann Intern Med 2009;151:293-296).

After a harrowing hospital experience featuring many staff members pushing around “laptops on wheeled sticks,” his life having been saved by a heroic ICU nurse who worked around the system, and his wife who sneaked his inhaler into his room, a very intelligent patient concluded that “electronic health information systems are mostly broken.”

“The national health information network envisioned by President Barack Obama is a pipedream,” he writes (Joe Bugajski, “The Data Model That Nearly Killed Me,” Syleum.com 3/17/09).

So why did Congress authorize $20 billion for HIT in the stimulus package? Proponents relied on a 2005 RAND estimate of $77 billion in savings—based on the assumption of an error-free system that would be rapidly implemented by 90% of all facilities. Even if achieved, $77 billion would be only 4.5% of total costs, placed at $1.7 trillion by RAND, writes Greg Scandlen (Heartland Institute 2/20/09).

Most likely, “every penny of the $20 billion will be wasted on systems that don’t work and can never be implemented. That was the outcome of federal attempts to upgrade technology at the IRS, the FBI, and the air traffic control system.”

Additional information:

• “Two New Challenges a Healthcare Cybernetic Utopia: Yet More Hurdles Exposed,” Health Care Renewal 1/7/10.
• “Health IT Promises and Threats” by Scot Silverstein, M.D., PowerPoint presentation from 2009 AAPS annual meeting.
• “The Syndrome of Inappropriate Overconfidence in Computing: An Invasion of Medicine by the Information Technology Industry?” by Scot Silverstein, M.D., J Am Phys Surg, summer 2009.
• “EMR—a Non-consented Experiment,” AAPS News, July 2008.
• “Electronic Panacea,” AAPS News, April 2008.