At my post "BusinessWeek on Health IT: The Dubious Promise of Digital Medicine" I began to discuss the responses made by health IT vendors and organizations regarding HIT problems and defects, raised in a very serious April 23, 2009 BusinessWeek article on these issues.
In this post I address the following claims regarding UMPC and a controversial Dec. 2005 article by a UPMC physician in the journal Pediatrics ("Unexpected Increased Mortality After Implementation of a Commercially Sold Computerized Physician Order Entry System", Yong Y. Han et al. -- my comments on this paper at that time were at this link).
... [HIT vendor Cerner] faced more questions over its technology at the University of Pittsburgh Medical Center (UPMC). In 2005 researchers there [i.e., Han et al. - ed.] found that at the university's Children's Hospital [note that this is a Children's Hospital - ed.], patient deaths more than doubled, to 6.6% of intensive-care admissions, in the five months following the installation of a computerized order-entry system. The research on child patient deaths at the University of Pittsburgh found a "direct association between [computerized records] and increased mortality," according to an article published in December 2005 in the medical journal Pediatrics. Digital technology slowed treatment in several ways, the researchers concluded. One example: Doctors and nurses in the intensive-care unit were accustomed to ordering medications and tests while a sick child was en route to the hospital.
The Cerner system required that orders be submitted only when the patient arrived, costing crucial time. The authors of the Pediatrics article acknowledged that their work clashed with other studies showing that digitization decreases errors and shortens hospital stays [but also is consistent with another body of literature (examples here) that Health IT vendors and others with conflicts of interest simply ignore, as per my numerous HC Renewal posts on the issue. Pharma should only have it as good - ed.]
G. Daniel Martich, chief medical information officer at UPMC, says the Pediatrics study was flawed. Factors other than the installation of computers, such as the centralization of pharmacy services, also disrupted care, he emphasizes. The problems identified in the 2005 paper have all been resolved, Martich adds. "There were workflow issues," he says. "We learned the hard way because we were pioneers." Over the long run, he says, technology has helped decrease mortality rates and cut medication errors in half at Children's Hospital since 2003 .
Again, I emphasize we are considering the healthcare of children in this matter.
Here are my observations and questions:
- Why is the 2005 paper simply being dismissed? This is not a scientific approach to the peer reviewed literature, especially where the deaths of children are concerned. I believed in 2005 and believe now that the red flags raised by this paper may have been "brushed under the carpet" too rapidly. See my 2005 post on the matter.
- The implication that the Cerner CPOE was uninvolved in the increased mortality would itself require a very careful observational study to ferret out the exact contributions of its deficits and that of the "centralization" of pharmacy services and other factors. Has such a study been performed and peer reviewed, and has a rebuttal been posted by the original study's authors? Can such a prospective study even be performed after the fact?
- While CMIO G. Daniel Martich claims "the problems in the 2005 paper have all been resolved" [wait - I thought the paper was flawed, yet it illustrated problems now resolved? - ed.], how can patients be assured of this? Where are the studies that show the issues have all been resolved?
- Does the damning commentary by the UK's former NHS HIT Director Richard Granger, stories such as patients going untreated for months at a time related to Cerner Millennium Care Records Service ("If NPfiT were a jumbo jet heaven help 747 flyers"), an equally damning report in Jan. 2009 on Cerner's EHR products (i.e., "Millenium") as reported by the UK House of Commons Public Accounts Committee, and the impact on the UK's £12.7 billion national Health IT project suggest a systemic problem needing investigation?
Pioneers in what, exactly?
As a former NIH informatics fellow and then former CMIO myself more than a decade ago, it seems more likely UPHS (as well as the children and their parents, and the pediatricians who held the ultimate responsibility per Koppel and Kreda's Hold Harmless paper in JAMA) learned the "hard way" because the organization ignored the work of the pioneers.
There is literature going back fifty years on the challenges of health IT as mentioned in my many posts at HC Renewal, such as here.
I began my informatics postdoc in 1992-4 working on CPOE implementation at Yale-New Haven Hospital. I'd learned lessons from the writings of others who'd experienced CPOE difficulties years before that, such as (emphases mine):
Massaro TA. Introducing physician order entry at a major academic medical center. I: Impact on organizational culture and behavior. Acad Med 1993;68:20–5. [PubMed]
Summary: In 1988 the University of Virginia Medical Center began implementation of a medical information system based on mandatory physician order entry. The implementation process was much more difficult than expected. The program experienced considerable delays, and cost much more than was originally estimated. Although there were some legitimate questions concerning the user-friendliness of the new technology, these were less significant than the cultural and behavioral problems encountered. The new system challenged basic institutional assumptions; it disturbed traditional patterns of conduct and forced people to modify established practice routines ... The author describes the problems that occurred and the organizational behaviors on which they were based, analyzes the lessons learned, documents the progress that has been achieved, and outlines the challenges that remain. The center's experience provides insight into the issue of technology-driven organizational transformation in academic medical centers. Recommendations for successful introduction of similar agents of institutional change are presented.
Massaro TA. Introducing physician order entry at a major academic medical center: II. Impact on medical education. Acad Med 1993;68:25–30. [PubMed]:
Summary: The introduction [in 1988] of an information technology (IT) system that mandates order entry by physicians had significant and often unexpected effects on medical education at the University of Virginia Medical Center. The system was deactivated briefly after the introduction of laboratory ordering, and frustration with the pharmacy ordering pathways provoked a major confrontation between the residents and medical center management. Changes in responsibilities, patterns, and priorities of work introduced by the system also contributed significantly to the general dissatisfaction. These issues had not been thoroughly considered in the planning stage, but it was only after accommodation was made to these changes that integration of the technology into routine practice could proceed. The author emphasizes the importance of extensive involvement and leadership of attending physicians in the planning and implementation of such a system. He presents a set of recommendations to those considering similar IT initiatives and wishing to reduce the disruptions that may accompany their introduction. With time and experience, however, the housestaff have adjusted to the system and developed facility in using it. Much of the dissatisfaction was derived from the perception that "doctors spend too much time on the computer." In fact, less than 10% of the physicians spent more than an hour each day. However, a small group of residents on call for the busier services were sometimes at the computer for more than four hours each day.
The entire thirty year old field of Social Informatics spells out the dangers of unintended consequences, and books such as Lorenzi and Riley's "Managing Technological Change: Organizational Aspects of Health Informatics", which contains enough information to have allowed avoidance of the issues in the 2005 paper, was first published in 1994.
My website on HIT difficulties would have helped avoid these issues as well, and it has been online and nearly unique (and the top hit on queries on 'healthcare IT failure' - click here - and similar) for a decade. Did anyone in this CPOE project do such a search, read it, and perhaps query its author about the material [of course not - ed.] before spending tens of millions of dollars and putting patients at risk?
In other words, was true due diligence performed? Considering this was a Children's Hospital, one would have expected extraordinary levels of due diligence.
Pioneers learning the hard way? No. It seems this medical center was a pioneer in ignoring the lessons of the past. I believe a simpler explanation of the difficulties in this CPOE project might be that, like many HIT projects, the project was mismanaged.
Perhaps Dr. Martich meant UPMC were pioneers in this particular new CPOE system. As in this article in the Pittsburgh Tribune:
The University of Pittsburgh Medical Center is taking another step in a quest to commercialize new medical technology."Proving ground?"
UPMC on Monday signed a three-year deal with health care information technology provider Cerner Corp. to develop and market medicine-related technological advances. Both parties will contribute $10 million in cash, services and intellectual property to the effort.
The deal is a smaller version of an April 2005 deal between UPMC and information technology behemoth IBM.
As is the case in the IBM deal, UPMC will serve as a built-in proving ground for jointly developed technologies and products, with Cerner marketing the products and UPMC awarded a share of profits.
I cannot believe I am reading such a claim. A hospital and patients, as I have written before, are not a learning lab for HIT vendors. The appropriate "proving ground" for new medical technology is the controlled clinical trial where participants (in this case, patients and healthcare professionals alike) have freedom of choice whether or not to participate, and a chance to give (or deny) consent after being fully informed of potential risk.
Were patients or their parents asked to give consent to the use of HIT (such as here)? Were they even aware they were serving as subjects of a "proving ground?" Likewise for clinicians?
It should be kept in mind that in this "proving ground" (i.e., experimental) situation we are dealing with children. Using a hospital as a "proving ground" for unproven HIT on unconsented or coerced subjects (e.g., children, their parents, and the doctors themselves) probably amounts to battery at the very least, is consistent with violations of fiduciary and Joint Commission safety standards, and I am concerned about it representing human rights violations as well.
In view of the recent Koppel and Kreda paper in JAMA on "hold HIT vendors harmless" and "defects gag" clauses, this entire arrangement seems incredibly disturbing at the very least.
I also believe the conflict of interest represented by a healthcare organization partnering with a vendor to "commercialize new medical technology" on unconsented subjects while sharing profits is a horrendous development, for as I stated, the appropriate settings for such practices are properly conducted, impartial clinical trials. The potential for abuses in the current arrangement seems uncomfortably high.
My sentiments seem aligned with IOM's recommendation 4.1 and others on Conflict of Interest as pointed out by fellow HC Renewal blogger Bernard Carroll at this post:
Academic medical centers and other research institutions should establish a policy that individuals generally may not conduct research with human participants if they have a significant financial interest in an existing or potential product or a company that could be affected by the outcome of the research. Exceptions to the policy should be made public and should be permitted only if the conflict of interest committee (a) determines that an individual’s participation is essential for the conduct of the research and (b) establishes an effective mechanism for managing the conflict and protecting the integrity of the research…” (page S-14).
One wonders if the individuals involved in the CPOE project had unreported conflicts of interest with the HIT industry. Perhaps, as in pharma, this needs to be explored.
Han's 2005 article in Pediatrics suggesting children died as a result of the "experiment" should probably be further explored on the basis of the issues I raise above. I also note Han is no longer at UPMC; perhaps his treatment after publication of this article should also be investigated.
In Dec. 2005 on this blog I wrote:
... This study is dealing with children, of course, and is perhaps a flag that much more detailed study of these systems, especially in socially-sensitive environments such as pediatrics, need to be performed.
For if this study's findings are not just due to serendipity and do reflect some underlying causation, the medical, ethical and legal issues could be enormous.
I stand by that assessment.
Finally, putting the camera to my statement that medicine is suffering a cross occupational invasion by the IT industry, in Business Week the picture below appeared of a handsome and proud physician in his office:
That person is not a physician nor does he have any medical experience of which I am aware. He is the CEO of the Cerner HIT company.
This picture is the future face of medicine, and it will not be the brightest of futures if this cross occupational invasion continues unopposed by those of a scientific and biomedical mindset much longer.
I really do believe it is the time for a Congressional investigation of these issues, and strong consideration of FDA regulation of health IT, a precedent set long ago in the pharmaceutical industry.
Addendum: I note in Dr. Martich's bio an impressive list of medical credentials; however I find formal Biomedical Informatics and/or computer science training lacking. This is not to attack the person, but the culture of HIT, as this scenario is common in HIT and represents what I refer to as "amateurs" running health IT.
I use the term "amateur" not derisively, but in the same sense that I am a radio amateur (albeit licensed after examination at the Extra class and in the "old days", when 20 WPM morse code proficiency was also required). I am not a telecommunications professional. While I have excellent skills in telecommunications theory, hardware, antennas, and operations, I would not consider myself even remotely qualified to, say, run a strategic telecommunications project for a large organization.
In medicine, I always believed education was critical. I may have been mistaken.