How CPOE violated medical ethics

by MedInformaticsMD

As a result of my mother’s health IT-related cerebral hemorrhage and resultant multi-month hospitalization, with transfers to nursing homes then back to acute care due to setbacks, I recently had the chance to observe her care in a small community hospital.

This was a hospital that, in her last several days there before going back to a nursing home for rehab, went live with a major vendor computerized physician order entry (CPOE) system.  The CPOE was brought in from a parent large hospital where the CPOE had been in use several years.

Just by passing the nursing station/doctor’s charting room on my mother’s floor and opening my eyes and ears, I saw doctors and nurses struggling to take care of patients while “getting the bugs out of the system.”

They had had received some classroom “training” in a static environment, but it was clear they were learning about a lot of “gotcha’s” and unanticipated glitches in vivo. The problems were predictable. In fact, I predicted unexpected difficulties to several of my mother’s clinicians before go live.

There was some skepticism (maybe in my nearly being in tears about my mother, I came off as a bit melodramatic). However, several later told me they “now knew what I was talking about” upon my mother’s discharge, just several days into the go live.

One story I overheard during go live especially sticks out in my mind.

A newly admitted patient who needed urgent heparinization did not receive the medication promptly. The patient’s physician could not order it, as nursing had not yet “admitted” the patient and entered data such as weight. Physicians found no way to override, despite calls to the help desk, attempts by on site IT people and users from the parent hospital.

In the end, the pharmacist simply provided the drug using a weight estimate despite no “official” order having been entered into CPOE. I heard that the delay was on the order of “several hours.”

Clearly, both technology and people issues were involved … but I assure the reader, injured or dead patients really don’t care exactly how their injury occurred, after the fact (other than in litigation, which doesn’t fix the damage or remediate the suffering).

This story sticks in my mind as it was due to disappearance of an anti-arrhythmic med on my mother’s ED admission med list at the parent hospital, without any alerts that a medicine she’d been on for years was somehow removed as an “active medication” that led to her needing heparin acutely. The iatrogenic need for heparinization then led to an iatrogenic cerebral hemorrhage and emergency craniotomy, followed by many complications.

Here, then, is my question.

Where does the moral authority come from to subject live, unsuspecting, uninformed patients to the type of risks the patient whose heparin was delayed was subject to?

What right did the hospital have to not inform this patient before admission that a new critical CPOE system was going “live” that day, and that the patient could consider going to another hospital a few miles down the road instead that had no such potential problems?

From the Belmont Report, the six fundamental ethical principles for using any human subjects for research are:

  • Respect for persons: protecting the autonomy of all people and treating them with courtesy and respect and allowing for informed consent
  • Beneficence: maximizing benefits for the research project while minimizing risks to the research subjects
  • Justice: ensuring reasonable, non-exploitative, and well-considered procedures are administered fairly (the fair distribution of costs and benefits to potential research participants
  • Fidelity: fairness and equality
  • Non-maleficence: do no harm
  • Veracity: be truthful, no deception

On the basis of Belmont Report and other medical ethics regulations, where does the moral authority come from for hospitals to put patients through such risks without informing them ahead of time and offering them an opt-out, even if only the continued use of paper in their care?

MedInformaticsMD is a physician and medical informatics professional who blogs at Health Care Renewal.

Submit a guest post and be heard.

Comments are moderated before they are published. Please read the comment policy.

  • Jennifer

    I am so sorry to hear about your mother. What a horrible sequence of events.

    Your posting raising 2 very good points everyone needs to consider

    1) The first is that computerized orders and medical records do not eliminate erors. In fact, new ones can pop up. While everyone can read the writing and see the meds someone is taking, if the wrong med or the wrong dose ordered it may not be caught. In one study, COE reduce medical errors from 12% to 6% with discharge meds. Not eliminated, just reduced. So everyone still needs to double and triple check they have ordered the right meds. There is no room for complacency.

    2) The second is training and roll out. You can’t give a hospital a few hours of class and then say, here you go. At our instituition when we went to fully computerized records we had extensive class tarining and then out patient schedules were reduced by 50% for several weeks and gradually ramped up to 100%. There were helpers available everywhere to smooth the implementation. That also happened in the OR where surgeons reduced surgeries by 50% to allow for time to get used to the system. There were “super users” everywhere on the inpatient side who could help us regular folk and make sure patient care wasn’t affected. I was very surprised at the transition. It was a highly coordinated roll out. If I had only received the training that you described I have no idea how I would have managed. In the haste to adopt computerized medical records we must not forget that they can become a hazzard if not implemented correctly. Thanks for your posting.

  • The Happy Hospitalist

    I think the same thing should apply to every new nurse, respiratory therapist, doctor, ward clerk and supply closet manager that starts working at the hospital.

    There’s a learning curve to every new job. Shouldn’t I have the right of informed consent to be transferred to another hospital if I find out the new night stock guy didn’t put the central line kits in the closet and I had to wait 45 minutes for my central line, almost dying in the process?

    • gzuckier

      Exactly. Witness the advice to stay out of the hospital in the months following the graduation of a new class. Or to avoid operations on Fridays. etc. etc. etc.

      Heck, compare this to the tolerance of doctors who everybody but the patient knows should be restricted to doing routine autopsies rather than operating on people, but are still allowed to practice for whatever reasons. Just another piece of evidence that the practice of medicine isn’t really THAT different from any other field of human endeavor, we just really really really would like to think it was.

  • MedInformaticsMD

    Dear Happy Hospitalist,

    Health IT is not a “job” requiring a learning curve. It is, as admitted by those in the know such as HIMSS Board chair Barry Chaiken, an experimental device.


    … We’re still learning, in healthcare, about that user interface. We’re still learning about how to put the applications together in a clinical workflow that’s going to be valuable to the patients and to the people who are providing care. Let’s be patient. Let’s give them a chance to figure out the right way to do this. Let’s give the application providers an opportunity to make this better.

    By definition, if “we’re” still learning how to do health IT to be valuable to patients, it’s an experimental, not a proven device.

    See the above internationally recognized rules on medical experimentation.

    As I noted here:

    * If ‘we’re’ still learning (and I don’t include people with genuine clinical computing expertise in that subgroup, but it does include the plethora of amateurs in the commercial health IT industry), then the technology is experimental.
    * Worse, it’s unregulated – a major special accommodation [to the health IT industry] in and of itself.
    * These sentiments about “being patient” would be appropriate – if the subjects of this experimental technology that vendors need to be “given a chance” to make better were experimental lab rats.

    • The Happy Hospitalist

      Then health IT falls right in line with all of medicine in that everything we do to patients is an experiment. From medication management to surgical interventions, everything we do to patients is an experiment. We have no idea what the outcome will be until after the fact.

      And every day I take care of patients, I’m learning something new.

      • jp

        As a GP (outpatient only), I can see both sides of the debate, but tend more to agree w/the author of the article.

        Comparing what Docs do everyday to transitioning to CPOE and by extension, EMR, across a huge system requiring coordination across multiple levels, fields and specialties, is comparing apples to oranges. Potential for error seems exponentially greater.

        Perhaps the real issue is that the rollout was poorly executed. Other posts have noted more methodical transitions which included reduced OR schedules, etc. Then again, there are other issues involving Docs that pop up in hospitals which compromise patient care, but this issue (order entry) just seems to be so critical to good outcomes that it should be taken more seriously than it was.

    • gzuckier

      To rework an old joke; “I gave the patient a beta-blocker, and now he doesn’t show up on the computer system”

  • MedInformaticsMD
  • MedInformaticsMD

    Re: Jennifer

    “I am so sorry to hear about your mother. What a horrible sequence of events.”

    Thank you.

    More training would have helped at the secondary hospital, but the error that nearly killed her occurred in the parent hospital where the system in question had been in use for > 2 years.

    I think better fundamental system design has to preempt the expectation that training is the solution to preventing HIT errors.


  • Dr Geek

    Coming from a similar background, I agree with about 75% of your argument. There is something to be said about implementing systems that clarify rather than obfuscate and no one argues that medical errors, especially avoidable ones, shouldbe ignored, but (having an IT background) you know that the systems most hospitals are implementing an that a lot of vendors are selling don’t address things outside the hospital. I don’t know which system this hospital used, or if they had integration with community physician records, but most don’t. Therein lies the rub. I, as a community doc, may have a perfect list of all the medicines you take, but if there isn’t effective communication between my system and the hospital (whether electronic or good ole fashioned paper) the point becomes moot, regardless of new, old or antiquated system age.
    The real problem is the ‘copy-of-a-copy’ problem. We’re replicating information hundreds of times, and just as humans are wont to do, we make errors which then propagate (ala Mutations in DNA), leaving us with bad inputs and, thus, bad outputs. GIGO

  • MedInformaticsMD

    Dr Geek, I agrre with all you wrote But I think we are talking about different issues.

    My concern is this post is about the ethical issues of implementing experimental devices that mediate or regulate care, becoming a cybernetic intermediary between clinician and patient, without patient informed consent.

    Either HIT systems are experimental, or they are not; either they are medical devices that mediate care, or they are not, and either informed consent is needed for their use in the care of patients, or it is not.

    Those are the issues I see as potentially debatable related to my posting, and I hope I’ve made my position on these points clear.

  • MedInformaitcsMD

    Happy Hospitalist writes

    Author: The Happy Hospitalist
    Then health IT falls right in line with all of medicine in that everything we do to patients is an experiment. From medication management to surgical interventions, everything we do to patients is an experiment. We have no idea what the outcome will be until after the fact.”

    This is not a serious argument but seems more a “moving of the goalposts” in unfettered and uncritical support of the status quo with regard to clinical IT.

    I woud hope, for example, that you practice sound medicine based on scientific research, clinical trials-proven safety and efficacy, and informed consent. And that you do not give someone in diabetic ketoacidosis, say, therapeutic touch or Eye of Newt as the sole therapy as an experiment.

    I also assume you do not ignore the medical literature out of convenience or bias.

    With regard to clinical IT, here’s just a small sampling of research and journalism casting doubt on health IT’s safety and even efficacy:

    1. Health IT Project Success and Failure: Recommendations from Literature and an AMIA Workshop by Bonnie Kaplan and Kimberly D. Harris-Salamone. From the May/June 2009 issue of JAMIA.

    2. “E-Health Hazards: Provider Liability and Electronic Health Record Systems.” Hoffman and Podgurski’s followup paper on EHR medical and legal risks

    3. Role of Computerized Physician Order Entry Systems in Facilitating Medication Errors. Ross Koppel, PhD, et al, Journal of the American Medical Association, 2005;293:1197-1203

    4. Electronic Health Record Use and the Quality of Ambulatory Care in the United States. Arch Intern Med. 2007;167:1400-1405. The authors examined electronic health records (EHR) use throughout the U.S. and the association of EHR use with 17 basic quality indicators. They concluded that “as implemented, EHRs were not associated with better quality ambulatory care.”

    5. Pessimism, Computer Failure, and Information Systems Development in the Public Sector. (Public Administration Review 67;5:917-929, Sept/Oct. 2007, Shaun Goldfinch, University of Otago, New Zealand)

    6. Bad Health Informatics Can Kill. his site contains summaries of a number of reported incidents in healthcare where IT was the cause or a significant factor. It comes from the Working Group for Assessment of Health Information Systems of the European Federation for Medical Informatics (EFMI).

    7. The U.S. National Research Council’s “Current Approaches to U.S. Health Care Information Technology are Insufficient.”

    8. The UK Public Accounts Committee report on disastrous problems in their £12.7 billion national EMR program.

    9. Gateway reviews of the UK National Programme for IT from the Office of Government Commerce (OGC) (released under the UK’s Freedom of Information Act).

    10. A report on the serious problems with the Department of Defense’s AHLTA system, Electronic Records System Unreliable, Difficult to Use, Service Officials Tell Congress. (This system, as I wrote here, is slated for abandonment. I cannot imagine it was greatly improving outcomes).

    11. A New York Times report “Little Benefit Seen, So Far, in Electronic Patient Records” on Jha’s research at the Harvard School of Public Health, that compared 3,000 hospitals at various stages in the adoption of computerized health records and found little difference in the cost and quality of care.

    12. An American Journal of Medicine paper “Hospital Computing and the Costs and Quality of Care: A National Study” by Himmelstein and Woolhandler at Harvard Medical School, that also concluded “as currently implemented, hospital computing might [very] modestly improve process measures of quality but not administrative or overall costs.”

    13. A Milbank Quarterly article “Tensions and Paradoxes in Electronic Patient Record Research: A Systematic Literature Review Using the Meta-narrative Method” by Greenhalgh, Potts, Wong, Bark and Swinglehurst at University College London.

    14. Health Affairs, 29, no. 4 (2010): 639-646 Electronic Health Records’ Limited Successes Suggest More Targeted Uses, Catherine M. DesRoches et al.

    15. NORCAL Mutual Insurance Company: “Electronic Health Records: Recognizing and Managing the Risks”

    16. FDA. Health IT risks, Feb 23, 2010 internal memo not intended for public use (via Huffington Post),

    These just skim the surface. There are numerous others. (Links to many of these can be found at my health IT difficulties site, , in the “2009″ section and the ‘other resources’ section.)

    Now, either we are scientists and address these issues, or we are voodoo practitioners who ignore science and do what our own preconceived notions, biases, or whims tell us to do.

    Technoligical determinism (loosely, “computers added to any domain automatically improve the domain”) and irrational exuberance about an experimental technology have no place in medicine, I believe.

    Perhaps I’m wrong and we should just practice medicine as the IT marketers, TV ads, and pundits on late night television tell us to, but I’ll stick to the scientific method, critical inquiry and accepted ethical principles for human subjects research for now.

  • MedInformaticsMD

    Addendum – on “moving the goalposts” in debate:

    “Moving the goalposts, also known as raising the bar, is an informal logically fallacious argument in which evidence presented in response to a specific claim is dismissed and some other (often greater) evidence is demanded. In other words, after an attempt has been made to score a goal, the goalposts are moved to exclude the attempt. This attempts to leave the impression that an argument had a fair hearing while actually reaching a preordained conclusion.”

  • DrA

    As someone involved with CPOE implementation at a small hospital, I agree with the post. I and the staff and HIT are doing our best to make this a safe, effective rollout. However, any mistakes caused by cpoe errors are iatrogenic, they are not random unpredictable adverse side effects that Happy Hospitalist implies.

    I belive patients should be Informed. I also believe hospitals are woefully understaffed – they need nurses, not aides and computers. I also belive that all patients need to double check every medicine administered to their bodies.

    I also believe in unicorms, fairys, and that lawyers will support tort reform.

  • DrA

    I practical wa to resolve this is to have CPOE deployment also approved by a hydrid ethics/IRB board/patient safety committee.

  • Marc Gorayeb, MD

    The proposition that hospitals or offices are experimenting with CPOE or EHR is a bit off the mark. No one is trying to determine whether the technology is safe and effective, so there is no experimenting going on. Therefore, no investigational informed consent is required.
    The technocrats have baldly declared that no further investigation or clinical trials on this technology are needed, that these systems should be implemented forthwith. Your issue is really with the government.
    A more fundamental question is whether these systems would survive in a free market (which we don’t currently have in health care). Under a free market scenario, new technology would not be rolled out widely until it has proven itself either in better outcomes or in cost savings.

  • MedInformaticsMD

    Marc Gorayeb, MD writes:

    The technocrats have baldly declared that no further investigation or clinical trials on this technology are needed, that these systems should be implemented forthwith. Your issue is really with the government.

    Agreed, this is a major point of my writings the past few years. Let me add that it’s not just my issue; it’s medicine’s issue as a profession, and society’s issue regarding healthcare.

    No one is trying to determine whether the technology is safe and effective, so there is no experimenting going on. Therefore, no investigational informed consent is required.

    Not to be darkly humorous, but “we were just following orders” is not a desirable excuse for going along with the mandates without medical pushback and critical thinking! Implementers who’ve taken the Oath should be evaluating just those issues.

    As a matter of fact, in NIH grant proposal reviews in my study section activities, consideration of just those issues is given regarding proposals involving some “new or innovative” use of health IT. If those issues are not addressed, for instance if the proposal does not satisfactorily safeguard patients and clinicians, then the proposal will get a poor score.

    Yet hospitals and practitioners seems to be under no regulations whatsoever.

    This is an inconsistency.


  • MedInformaticsMD

    Dr A writes:

    I practical wa to resolve this is to have CPOE deployment also approved by a hydrid ethics/IRB board/patient safety committee.

    While I prescribe no one solution, this is one possible “front end” method to improving HIT quality (a necessary precondition to improving healthcare itself).

    Another is the back end method used in other areas of medicine: regulation.

  • Happy Hospitalist

    I guess I haven’t had my question answered. Should every patient in the hospital be notified when there is a new nurse, doctor, floor tech or supply closet stocker hired and be given the opportunity of informed consent to transfer to another hospital. I mean, if were talking about giving patients informed consent about initiation of technology, should we not be given the same courtesy when the new supply clerk is hired and responsible for placing the central lines in the closet?

  • ninguem

    Why does everyone go gaga over the computer in medicine?

    The patient needs the human services provided by the physicians, nurses, staff. Everything else is support.

    The information system is of a kind with the power supply, the plumbing, and the cafeteria for that matter.

    There are a whole bunch of hospital-specific things we do because patients got hurt from error, and improperly designed systems that did not take into account the work flow and the possibility (certainty?) of human error, fatigue, and the confusion of emergency/crisis situations that hospitals see every day.

    Piped gases, pin-index systems, electrical grounding, special outlets in operating rooms, were done because of fires, nitrous oxide in the oxygen lines, explosions, accidental power disconnects, etc.

    I responded to a hospital code and found central supply had the wrong jacks for the piped oxygen. There WILL be problems when rolling out a new system. There’s something wrong if you can’t do a certain service until the computer lets you. For Pity’s sake, fix it.

  • Jo

    The best way to resolve this is for all physicians and hospitals to demand that the EMR companies prove that thier systems work in the real world or they will be fined by the government and open to lawsuits by the users for harm done to their customers (the patients).

    “Meaningful Use” of EMR is only meaningful if the EMR is meaningfully useful. The criteria to certify these systems is that they must prove certain “computer” functions in an “outside of office visit/hospital” setting, not that it is actually useful with patient care. There are no “real world” worse case scenerio tests that these companies have to pass, yet hospitals and physicians have to pay out thousands of dollars for malpractice if that “system” is faulty, does not have fail-safes or just doesn’t work on a given day, and they physicians and hospitals will be fined by the governement in just a few years if they do not make sure the systems work a certain way.

    Physicians/hospitals have to show that they are compliant with the “Meaningful Use” criteria and be compliant with the rules but the EMR systems that the government are certifying to be used are not required to perform these tasks or even show that they have trained the end user properly in order for there not to be any harm.

  • MIMD

    Regarding “I guess I haven’t had my question answered. Should every patient in the hospital be notified when there is a new nurse, doctor, floor tech or supply closet stocker hired and be given the opportunity of informed consent to transfer to another hospital. I mean, if were talking about giving patients informed consent about initiation of technology, should we not be given the same courtesy when the new supply clerk is hired and responsible for placing the central lines in the closet?

    I think it was answered. Appears you are shifting the issue but I’ll try again.

    Doctors and nurses are subject to qualifications and credentialing checks before hire. Supply closet stockers are subject to criminal background and often drug tests. Medical trainees are supposed to be adequately supervised. Patients sign consents to treat on admission or visits that state or imply the hospital or doctors office meets JC and other standards for qualification of practitioners, and standards of care. When a procedure is done that is even a little off the ordinary and could cause harm, e.g., a PIC line, another more detailed informed consent is usually obtained.

    Health IT on the other hand is experimental, uncontrolled, unregulated, and known to be able to cause harm. Yet there is no consent process to its use in mediating care.

    I ask once more as in the main essay’s focus…how is this ethical? What gives us the right?

    • gzuckier

      With respect, no. The difference between being the first patient a new graduate works on vs a patient during a doctor’s “peak” years (vs a patient of a doctor who’s kind of in the retirement phase, while we’re on the subject) is probably as large, and probably much more frequently experienced, than the difference between being the first patient experiencing a new IT system and a patient when that system has become routine, despite the doctors having been properly educated, tested, and credentialed. That just provides a maximum size for how bad the difference could be.

      • MedInformaticsMD

        With respect, no.

        I,m not sure what the “no” refers to.

        However, if I interpret your statement correctly … that we should not try to be ethical and have stringent safety requirements in healthcare because ethics and requirements cannot guarantee perfection … then I have little else to add other than we live in different worlds.

      • MedinformaticsMD

        This is probably a good time to introduce the observations that led me to write the hosted essay.

        Those observations are in my JAMA letter to the editor published July 22, 2009 entitled “Health Care Information Technology, Hospital Responsibilities, and Joint Commission Standards” at

        For those without JAMA access, a fuller and free version is at this link:

        I observed that cavalier, fatalistic or other inappropriate (for medicine) attitudes about health IT have led to hospital executives signing health IT contracts that violate not only Joint Commission safety standards, but also their fiduciary responsibilities to those who work under their roof.

        I believe these clauses:

        * Are responsible for stagnation in health IT industry thinking that requires an unrelenting focus on the needs of clinicians and patients;

        * Cause hospital executives to violate their fiduciary and Joint Commission safety obligations to patients and staff;

        * Perhaps even put clinicians themselves in a conflict of interest with their own professional ethics (that call for widely sharing information about potential risks to patients).

        This is what led me to consider the essay above after my mother’s injury.

        – SS

  • MIMD

    Re: Jo & “Meaningful use”

    I disliked this term and stated so in government testimony during open commenting period.

    Good faith use” [that use will cause benefit] is an accurate term that fits the reality on the ground much better.

    “Meaningful use” is actually Orwellian; it assumes the outcome. It is amazing how propaganda can be diffused with only two words.

  • gzuckier

    It’s not directly related to patient treatment, but I found it somewhat symbolic at the time; when one local hospital first implemented ther PC based system, they required a hardware “key” as well as a password to log in, to preserve patient security. There followed very very quickly the first modification to the system, which bypassed the requirement for that key.

Most Popular