Featured Post

Musing on the Interaxon Muse Meditation Headband

"For this calibration, find a comfortable position and take a deep breath". The computer brain interface world is getting int...

Thursday, February 7, 2013

Research Ethics Review for EMR Systems?

This article by Wes Fisher found on the KevinMD.com news captured my interest because what I think the author is saying is that we should have Research Ethics Boards (REB) or Institutional Review Boards (IRB in the United States) to review implementations of Electronic Medical Record systems. It is interesting because I am a research ethics professional and I have a graduate degree in eHealth. He compares developing and use of EMRs to human experimentation, and as we know today, no human research experimentation can be done without clearance by an REB or IRB.  He uses quotes from the Belmont Report, which is one of cornerstone research ethics policies, to make comparisons for the need for human participant protection from possible harms from EMRs.  In the United States there is a technology push for EMRs and a lot of money mandated for development of EMRs. This could and probably does lead to the development of systems that are not going to be successful, and possible pose harm for patients. The ethics discussed in the article is also more about economic ethics, or the ethics of pushing a technology in advance of knowing net benefits, user acceptability, user satisfaction.

I have recently been thinking that EMR systems should not only have Privacy by Design principles built into their conception, development, and implementation, but also Evaluation By Design. We know that IT systems are notorious for high failure rates - up to 70% of them. It is no different in Health IT. In Canada, Health Infoway does have a certification system for EMRs, which costs about $100,000 and several companies have obtained that certification. This is not same as having an REB review the system.  As well, Health Canada can define some EMRs as class II Medical Devices, which require a certification. Class I medical devices only display or transport data.  Class II devices manipulate the data for decision support purposes, and thus, act more like humans - prone to making error. This Health IT failure website has great case studies.

From what I have studied, a lot of private companies are developing innovative healthcare applications to help people with chronic conditions. Often these applications are tested with patients without having gone through an REB review, unless the companies are working directly with university or hospital researchers who have requirements for REB review. At minimum, the clinical efficacy of these systems needs to be proven in such things as controlled clinical trials, before governments and businesses decide to spend vast sums for patient healthcare. As Dr. Norm Archer in the McMaster School of Business says, the return on investment (ROI) in Healthcare has to be improvement in quality of life.

I have been doing some research lately on information system success models, such as the DeLone and McLean model. This model for IS success has been well tested and measure in information systems studies, but I don't know how it has been applied to healthcare. I have seen the Technology Acceptance Model (TAM) used, but I am told this model is overused and many journals don't accept articles anymore using it. I am going to look into how the Delone and McLean model is used in Healthcare, because it could be one of the cornerstones for an Evaluation by Design model for EMR systems.  I also want to think more along the lines of this article by Dr. Fisher, and think about how the Canadian policy for research ethics, the Tri-Council Policy Statement, could apply to the review of EMRs.

The ethics of EMR: How unproven technology affects patients

The ethics of EMR: How unproven technology affects patients
The implementation of the electronic medical record (EMR) in American medicine gained a powerful foothold in medical care with the passage of the American Reinvestment and Recovery Act (ARRA) in 2009. With the passage of this act came the promise of improved efficiencies, safety and ultimately reduced cost delivery for health care. Also, some $18 billion dollars in financial incentives were offered to physicians to offset costs to deploy these systems nationwide. To assure adoption, if the systems were not implemented by 2015, doctors and care providers will suffer payment penalties from the government. For physicians who care for Medicare patients, there was no alternative than to deploy these systems.
In 2010 alone, the EMR market was pegged at $15.7 billion dollars, a cost that is ultimately passed to all Americans. In addition, despite all of the changes that health care reform has brought to date, people in some states continue to see their insurance premiums mushroom over 20% in 2013 from the preceding year. Simply put, patients are finding health care anything but “affordable.”
We should acknowledge that there might be cause, ethically, to deploy a technology that truly benefits patients at some cost. After all, you have to break a few eggs to make a good omelet. If interoperability of EMR systems between facilities were commonplace and clinical data were shared with ease while patient privacy was vigorously upheld flawlessly, the cost of these systems might be ethically justified.
But the promise of improved efficiencies to our health care system, improved patient safety and (especially) reduced cost for our health care system remain elusive. More importantly these goals remain unproven. In fact, examples that the opposite is occurring abounds as doctors struggle to enter ever-increasing amounts of information of no relevance to the patient’s presenting problem just to prove they’re using the EMR in a “meaningful” way, health data security breeches continue, data-mining of patient information is occurring not just for patient care, but for marketing purposes, and the direct costs of health care for patients continues to rise, not fall. Proponents of these systems will argue these issues are nothing more than “growing pains” of these novel systems.
So should we step back for a moment and ask ourselves if we are being ethical to patients with the deployment of this technology? Does the ends of presumed cost savings to our national health care system justify the deployment of poorly integrated, difficult-to-use systems? Are patients being subjected to new risks heretofore never considered with the adoption of this technology? Could a tiny programming error occur that negatively impacts not just one patient, but millions? If so, what are the safeguards in place to prevent catastrophic error? Who will be responsible? Who is the oversight body that assures the guiding principles of the Belmont Report (respect for persons, beneficence and justice) with respect to EMR deployment are followed? The Secretary of the Department of Health and Human Services or a more nebulous body like Congress?
If we accept that the benefits of the EMR are at least uncertain to patients in terms of risk and cost, we should demand they be studied before deploying them. The guiding medical ethics tenets would demand nothing less. So, would not such study qualify as human research? After all, we should remember that the United States and other countries have a precedent of human research programs performed by government agencies that were usually highly secretive, and in many cases information about them was not released until many years after the studies had been performed.
From a sentinel paper in 1966 by Henry J. Beecher, MD on Ethics in Research:
I should like to affirm that American medicine is sound, and most progress in it soundly attained. There is, however, a reason for concern in certain areas, and I believe the type of activities to be mentioned will do great harm to medicine unless soon corrected. It will certainly be charged that a mention of these matters does a disservice to medicine, but not one so great, I believe, as a continuation of the practices cited.
Experimentation in man takes place is several areas: in self-experimentation; in patient volunteers and normal subjects; in therapy; and in the different areas of experimentation on a patient not for his benefit but for that, at least in theory, of patients in general.
While Beecher’s paper was addressing ethical research errors in general, his words are oddly prescient for EMR development. Ethical errors, as he pointed out, “are increasing not only in numbers but in variety.” He points to one of the biggest drivers of ethical conflict: money.
Of transcendent importance is the enormous and continuing increasing in available dollars for research, as shown below:
Money Available for Research
YearMassachusetts General HospitalNational Institutes of Health

These data, rough as they are, illustrate vast opportunities and concomitantly expanded responsibilities.
Taking into account the sound and increasing emphasis of recent years that experimentation in man must precede general application of new procedures in therapy, plus the great sums of money available, there is reason to fear that these requirements and resources may be greater than the supply of responsible investigators.
The need for “responsible investigators” remains significant; funding for all of the National institute of Health in 2011 was $142.5 billion dollars. Annually, EMR companies have received the equivalent of 11% of the entire NIH annual research budget from US citizens without having to prove their safety or value to patients.
Again, from Beecher’s paper:
The ethical approach to experimentation in man has several components; two are more important than others, the first being informed consent. The difficulty of obtaining this is discussed in detail. But it is absolutely essential to strive for it for moral, sociologic, and legal reasons. The statement that consent has been obtained has little meaning unless the subject or his guardian is capable of understanding what is to be undertaken and unless all hazards are clear. If these are not known this, too, shall be stated. In such a situation the subject at least knows that he is to be a participant in an experiment. Secondly, there is the more reliable safeguard provided by the presence of an intelligent, informed, conscientious, compassionate, responsible investigator.
Because EMR deployments are cloaked in intellectual property, non-disclosure and restrictive hospital employment agreements, doctors are often prohibited from voicing specific concerns about an EMR system publicly. In addition, by adopting EMR systems as cornerstones of the American health care system, Congress, the President and the ARRA side-stepped patients’ informed consent regarding the short-comings of these systems, advertising only their desired benefits instead. Furthermore, rather than Congress turning to “conscientious, compassionate, responsible investigators,” they turned to lobbyists when deciding to fund the deployment of unproven EMR systems. As a result, doctors were relegated to becoming nothing more than stewards of data entry subject to new, ever-evolving documentation requirements as these systems evolve for cost-saving benefits and care “efficiencies.”
Patients and doctors alike understand the need for improved efficiencies and value in our era of exploding health care costs. We must strive to find a solution to our health care cost crisis that is transparent, cost-effective and ethical. Without such an effort, our health care system will collapse. Only recently has the Office of the National Coordinator of Health Information Technology recognized the problem and opened their Health IT Patient Safety Action and Surveillance Plan for public comment. This plan asks the EMR companies and interested stakeholders to develop their own methods to assure patient safety and reporting systems – a move that approaches the same ethical standards as equivalent of asking the foxes to watch the henhouse. Nonetheless, we should acknowledge their efforts.
But we should be cautious of EMR systems as we move forward. After all, these clinical systems have not been subjected to the same cost-benefit and ethical scrutiny as other clinical tools we use in health care. The scrutiny of EMRs should be no different than that found with pharmaceutical or medical device research where Institutional Research Board approval and proof of no conflict of interest is demanded. Why should clinical EMR systems be any different?
Given the profit motives and market consolidation occurring amongst the purveyors of these EMR systems and the potential for lethal EMR errors both from software and human interface issues, doctors and patients must especially question the ethics of the movement to deploy untested, novel technology on our patient population under restrictive covenants. As part of informed consent, patients should have full understanding of how and where their clinical data are used, including when it will be used for direct-marketing campaigns, prioritizing care delivery, or for research. Patients should be able to opt out of the use of their clinical data for these or any other purpose if desired, without restricting payment for care. Finally, physician and patient concerns about EMR systems should be allowed to be vetted publicly and without threat of professional or personal reprisal or the withholding of payments for care rendered, especially and particularly if these disclosures are performed in the best interest of patient care.
To do otherwise is unethical for our patients and the public at large.
Wes Fisher is a cardiologist who blogs at Dr. Wes.
Image credit: Shutterstock.com