No more dithering on e-health: let’s keep patients safe instead

McGrail et al [1] overlooked several facts when urging that Canada ‘catch up’ with other OECD countries in the use of e- health technologies.

a) While Canada does indeed lag behind most countries reported by the 2009 Commonwealth Fund survey in terms of the use and functionality of EMRs, it is misleading, at best, to link suboptimal performance of primary healthcare in Canada to low penetration of EMRs and EHRs without recognizing other contributing system shortfalls.

b) There continues to be a lack of robust evidence for the effectiveness or cost-effectiveness of e-health applications on improving patient outcomes, including medication safety outcomes.[2-3] Many of the applications that are promoted, such as electronic health records (EHRs) and electronic medical records (EMRs), are particularly speculative in their potential for improving care.[3-4] Other more focused technologies such as disease registries, e-prescribing applications, computerized decision support systems, have some evidence of benefit but the evidence tends to be of low quality, inconsistent and focused on processes rather than clinical outcomes of care. These are more recent findings of inefficacy.

c) Undue haste to implement early stage technologies into complex, busy and constantly innovating clinical practice settings can harm patients by creating sets of new errors, diverting scarce professional resources and limited healthcare funds.

d) Lack of patient involvement and engagement in the roll out of ehealth technologies leaves out an important stakeholder and a potential partner in realizing gains of these new technologies.

Thus far, many of the limited successes reported in ehealth are based in academic centres with decades invested in developing and modifying their home-grown information systems to enhance patient care. These successes are often contextual in nature and have been difficult to replicate in other healthcare organizations.

We agree with the overall recommendation of the authors that physician and other health professional organizations take more of a leadership role, as the e-health agenda will succeed only when clinical needs and patient outcomes drive technical development, not the other way around. We would make 4 recommendations:

1. Focus on rigor in research and evaluation of e-health technologies. Canada is a world leader in health technology assessment and, if there ever was technology in need of rigorous assessment, this is it. Progress is slow, partly because assessments to date have not identified why a system failed or succeeded, whether a system could be implemented in our own settings or whether it is sustainable. In other words, we still don’t understand what does and doesn’t work and why. [6]

2. Focus on end-user benefit. Incentives for technology adaption for clinicians and patients are important to offset the high costs, but adequate support to implement technology correctly has been lacking. Furthermore, if EMRs or ePrescribing, for example, are to drive improved quality of care, they will require improvements in level of sophistication of knowledge base and decision support. For example, there is no evidence that drug interaction checking modules improve the quality of prescribing or patient outcomes.[7] This is because the drug interaction knowledge bases tend to be heavily flawed by inadequate consideration of clinical importance of the potential interaction as well as quality of evidence supporting the interaction.

3. Focus on interoperability. Hospitals, medical offices, pharmacies, laboratories and other clinical points of care, need to be able to easily and securely exchange information regarding patients. Our current CIHR-funded projects will evaluate the interoperability challenges of monitoring and improving drug safety and effectiveness in clinical practice and of delivering electronic health records to Canadians.

4. Focus on health informatics capacity. Just as it is not possible to provide good healthcare without trained doctors, nurses, pharmacists and other health professionals, it is not possible to successfully implement and maintain ehealth technologies without people who are trained in the multiple domains of health, management and information sciences –health informatics.

Sincerely,

Karim Keshavjee,MD,MBA, CEO InfoClin Inc.;

Anne Holbrook, MD,PharmD,MSc, Director, Division of Clinical Pharmacology & Therapeutics, McMaster University;

Mitchell Levine, MD, MSc, Director, Centre for Evaluation of Medicines, McMaster University;

Janusz Kaczorowski, MA, PhD, Professor, Department of Family Practice, UBC;

Malcolm Maclure, ScD, BC Academic Chair in Patient Safety, UBC.;

David Chan,MD,MSc, Associate Professor, Department of Family Medicine, McMaster University;

Robert Bernstein, Professor, Department of Family and Community Medicine, University of Toronto;

Lisa Dolovich PharmD,MSc, Research Director, Dept of Family Medicine, McMaster University;

Mike Green MD, MPH, Associate Director of Research, Department of Family Medicine, Queen’s University;

Francis Lau, PhD, Professor, School of Health Information Science, University of Victoria;

Michael Shepherd, PhD, Dean, Faculty of Computer Science, Dalhousie University.

[1] McGrail K, Law M, Hébert PC. No more dithering on e- health: let’s keep patients safe instead. CMAJ 2010;182 535.

[2] Mollon B, Chong JR, Holbrook A, Sung M, Thabane L, Foster G. Features Predicting the Success of Computerized Decision Support for Prescribing: A Systematic Review of Randomized Controlled Trials. BMC Medical Informatics and Decision Making 2009;9(11):1-28.

[3] Garg AX , Adhikari NK, McDonald H, Rosas-Arellano MP, Devereaux PJ, Beyene J, Sam J, Haynes RB. Effects of computerized clinical decision support systems on practitioner performance and patient outcomes: a systematic review. JAMA. 2005 Mar 9;293(10):1223-38.

[4] Poon EG, Wright A, Simon SR, Jenter CA, Kaushal R, Volk LA, Cleary PD, Singer JA, Tumolo AZ, Bates DW. Relationship between use of electronic health record features and health care quality: results of a statewide survey. Med Care 2010;48(3):203-9.

[5] Zhou L, Soran CS, Jenter CA, Volk LA, Orav EJ, Bates DW, Simon SR. The relationship between electronic health record use and quality of care over time. J Am Med Inform Assoc. 2009;16(4):457-64.

[6] Kaveh G. Shojania, Alison Jennings, Alain Mayhew, Craig Ramsay, Martin Eccles, and Jeremy Grimshaw. Effect of point-of-care computer reminders on physician behaviour: a systematic review. Can. Med. Assoc. J. 2010;182:E216-E225; doi:10.1503/cmaj.090578.

[7] Wong K, Yu S, Holbrook A. A Systematic Review of Medication Safety Outcomes Related to Drug Interaction Software. Can J Clin Pharmacol. (in press)

Conflict of Interest:

Dr. Keshavjee consults to physician organizations on how to use EMRs to improve patient care. He also consults to EMR vendors on improving their technologies. He does not own shares in any EMR companies. The remaining authors have no conflicts.

It is my understanding that electronic medical records have been widely and successfully implemented in Newfoundland and Labrador. This was achieved by bringing hospitals and laboratories on board first, creating a stable structure, before engaging care providers. In my mind's eye, I envision a strong stable infrastructure, like the trunk of a tree.

In Ontario the implementation of EMR has relied on the goodwill and financial resources of individual practitioners, especially "early adopters" who make a personal decision to adopt EMR without a solid infrastructure in place. Funding is sporadic and inadequate. Not surprisingly, uptake is low. Without a strong central structure, the branches cannot support the weight of their fruit.

As a family physician practising in Ontario for many years, I am reluctant to dangle from the branches. I am even more reluctant to commit my staff and my patients to a process which historically creates more problems than it solves. In Ontario, EMR is "not ready for prime time".

Conflict of Interest:

None declared

Dr. Majumdar argues that the jury is still out on the safety and effectiveness of electronic health records. Based on what evidence?

On safety, he is concerned with implementation problems and patient harm. For implementation, our successful peer nations can now provide options for phasing in universal electronic systems smoothly and effectively. Regarding patient safety, his only evidence is a methodologically weak, pre-post study covering 5 months of experience in 1 U.S. hospital over 7 years ago. If this is the best “evidence” of harm resulting from over a decade of research in a multitude of settings, it certainly does not justify foregoing the potential benefits.

And the evidence suggests substantial potential benefits. Dr. Majumdar cites a meta-analysis that concludes—in 4 different organizations with strong leadership—electronic systems increased guideline adherence, enhanced public health surveillance, and decreased medication errors (1). Further, he has himself argued that electronic systems reduce medical errors in countries where “physicians are required and trained to use the national systems.” (2) Surely this could be achieved in our universal provincial health systems.

There are tens of thousands of preventable adverse events every year in Canada; an integrated and universal data infrastructure could help reduce them and facilitate coordinated care for the many Canadians with complex chronic conditions. We carry no illusions that electronic records will fix all the problems in health care. However, we’re already spending billions, the potential benefits are large, and the experience of other nations suggests that EHRs are a necessary step toward improved quality and safety. We’ve been conservative for long enough.

Respectfully,

Kimberlyn McGrail, PhD

Michael R. Law, PhD

-----------

1. Chaudry B, Wang J, Wu S et al. Systematic review: impact of health information technology on quality, efficiency, and costs of medical care. Ann Intern Med. 2006;144:742-752

2. Soumerai SB, Majumdar SR. Bad Bet on Medical Records. The Washington Post; March 17, 2009. Available at: http://www.washingtonpost.com/wp- dyn/content/article/2009/03/16/AR2009031602618.html [Accessed March 29, 2010].

Conflict of Interest:

None declared

It was with some level of bemusement that I read your editorial on e- health, especially when putting into context a prior editorial on the pursuit of excellence. In a chapter in Alice in Wonderland, Alice finds herself running with a rabbit as fast as she can, while surprisingly staying in place; the rabbit explains that everything else is moving, and that they had to run just as fast simply to keep up.

Unfortunately, medicine in Canada does not work this way: when the regulators, legislators, and medical insurers all belong to same bureaucratic provincial institution, medical practice becomes an extension of this bureaucratic practice. As a result, the need to “run” is non- existent in certain fields as competition becomes supplementary as opposed to compulsory. To believe that physicians, if given the proper carrot, will adopt e-health, fails to recognize the many facts on the grounds:

1. Incentives have already been in place, with little visible change.

2. Hospitals still have a certain philosophy of ownership of information with archaic notions of document security.

3. The questionable (even suspicious) implementation of poor quality EHR in many institutions.

4. The hostility many physicians have towards change, and technology in general. (see Liette Lapointe and Suzanne Rivard: Getting physicians to accept new information technology: insights from case studies, CMAJ 2006 174: 1573-1578.)

Bureaucratic organizations require bureaucratic methods: doctor fees will only be paid if they can prove, electronically, that their work had been done. While some may view this as brash, this is but the first baby- step for medicine as it's about the enter the information age. That is, unless we really are seeking to have "third-world" medical practices.

Conflict of Interest:

None declared

One can but agree with message of the editorial by McGrail, Law and Hebert (CMAJ, 6 April 2010) that “It’s time to change our approach.” However, I disagree with the changes needed and take particular exception to two statements: "Unfortunately, mandating the use of electronic health records may be the only way to avoid long delays.", and "Our laggard position is certainly not a question of software design or lack of access to user-friendly technologies."

Regarding my first exception, why is it that countries like the Netherlands (curiously not mentioned in the editorial since for the last two years it received by far the highest score in the EuroHealth Consumer Index [www.healthpowerhouse.com]), Denmark and Portugal (the country with the most advanced e-Health score) did not have to resort to mandating the use of EMRs? The answer lies in part in having provided remuneration for the use of EMRs, not simply subsidizing their purchase, leaving physicians in Canada paying out thousands of dollars for a technology that benefits the payer (read provincial governments) and the patient, but not physicians.

The other part of the answer lies in my second exception: the software design and user-friendliness of the EMRs approved for subsidization (in British Columbia) are severely lacking in functionality. As a specialist in endocrinology and metabolism I have had to reduce the number of patients I see in a given period of time since adopting my current EMR a year ago. Yet, I have used EMRs in Europe that are far superior to what is available to us. The countries mentioned do not exclude any vendors unlike Canadian provinces but, rather, insist on some basic functionality and inter- operability. The net result is intense competition and better products. As an example of problems with my approved EMR: there is no automatic backup feature during a computer crash or when one inadvertently accesses another function: the whole record is lost (up to 45 minutes of work for a consultant). It takes 12 steps, and 80 seconds, to get from a patient’s open ‘chart’ to write a prescription and back to the open chart (Question: why must the chart be closed before undertaking another task for the same patient?). At that point I must sign the prescription and hand over a piece of paper 6 times larger than my written ones (which takes me 20 seconds to complete legibly). Whatever happened to the paperless record? Similarly, it also takes 12 steps and 90 seconds to prepare a laboratory requisition before I can sign it and hand it to the patient. A third huge time-waster is to continue to receive laboratory data on patients that I no longer follow. The problem there would appear to be that once my name has been added to a requisition filled out by another physician I now continue to get copies ad infinitum. The same happens, of course, with paper reports but it takes 2 to 3 seconds to dismiss them to the shredder while accessing, viewing and processing the electronic results takes 7 steps and 35 seconds. That is, if the results are bundled together: it is not unusual to receive 10 separate results as separate reports, each requiring the 7 steps and 35 seconds (and that is not even mentioning the results that arrive with the notation: “Patient Unmatched”, requiring several steps to reassign each result to the correct patient and which then have to be dealt with individually).

Reading this must make you wonder why anyone should want to adopt an ‘approved’ EMR, at least in British Columbia. Most of the problems I have mentioned affect both GPs and specialists. However, the approved EMRs in British Columbia are very GP-centric, adding to the problems for specialists: in spite of promises made I have virtually no templates, essential to follow chronic diseases, nor have been given the tools to create my own. Rather, I continue to use my Excel templates to record the so essential sequential data for patients with chronic diseases. Attaching these and other attachments, takes 9 steps and 45 seconds per attachment. Retrieving each attachment takes 5 steps and 30 seconds. Data entry also continues to be a problem. In spite of what is claimed the reality is that the accuracy of speech recognition in a practical setting is at most 95%. The same is true, at least in my hands, of handwriting recognition. By the way, the only physicians I can communicate with electronically are the ones in my office using the same EMR! It’ll surprise you that I still have faith in the future of EMRs. However, much can and has to change before i can recommend their adoption. Having been on a number of computer evaluation committees I am disappointed by how little heed is paid to the consumer, that is, the physician. I am particularly concerned about the failure to adopt a simple, common, ‘medical problems and treatments profile’ for each patient, one that would make it unnecessary for each physician seeing a patient for the first time to laboriously review and reenter it in the record.

I have four suggestions to improve this dreadful situation:

1. Do not provide subsidies to physicians to acquire an EMR; rather, pay them extra for the patient record they provide with it (I would suggest a 30% premium on top of the usual fees);

2. Open up the market of ‘approved’ EMRs to all vendors, providing the competition so necessary for innovation and efficiency. Physicians will soon sort out which ones are worth the money and which are not. It was this approach used in Denmark and the Netherlands with more than 20 EMR vendors that led to their high penetration, unlike England where the single centralized record is many years behind and has already cost more than C$3000 per citizen;

3. Do away with the paper prescriptions and requisitions, as is the case in Denmark, the Netherlands and many other places including parts of Mexico.

4. Develop a simple one-page Personal Health Record Summary, carried by patients on a flash memory stick and readable by all EMRs, instead of a nation-wide EMR for each patient: patients’ medical records are needed in their local region, seldom across the country. Such summaries would also serve as an emergency back up at times of computer crashes and inaccessible internet connections, events that inevitably occur.

The reason the EMR uptake in Canada is so far behind other countries is because in its present form it isn’t worth the time, effort and cost to physicians. Provide a better product and physicians will flock to it. Why haven't we looked more thoroughly at what is available elsewhere and adopted the best of it? And where are the head-to-head, randomized, controlled comparisons of the superior performance of the EMR versus paper records? Hint: there aren’t any! The industry relies on endorsements, e- evangelists and the proven mantra of successful promotion: repeat the same message repeatedly and people will begin to believe it. In the mean time my advice to physicians considering buying an EMR is simple: insist that the company lets you use one, free of charge, for one month before signing the contract.

Gerald Tevaarwerk, Endocrinologist, Victoria, British Columbia.

Conflict of Interest:

Nil

Editorialists believe Canadians (presumably physicians) are “dithering” - nervously irresolute in doing or acting - on e-health. Unfortunately, wishful thinking about e-health by the editorialists seems to disparage any specialist or primary care physician that is not adopting e-health.

They state e-health (electronic health records with some decision support) will save lives and improve outcomes. Really? E-health has never consistently and demonstrably done either [1,2]. Why should more information, alone, lead to better outcomes? E-health has also never consistently and demonstrably increased clinical productivity or saved money [1,2]. Furthermore, most good evidence comes from four American institutions that used home-grown systems [1]. The “many vendors…with outstanding products and services” are not subjected to the even the minimal standards that we try and apply to other healthcare interventions.

Indeed, there are many examples of proprietary e-health systems being strongly pushed onto providers that led to organized physician revolts (Cedars-Sinai, Los Angeles), abandonment of too-rapidly implemented systems (Kaiser-Hawaii), and maybe increased mortality (Childrens’ Hospital, Pittsburgh).

Were physicians who did not quickly adopt thalidomide, rofecoxib, gastric freezing, or EC-IC bypass because of the lack of evidence for benefit and safety “dithering?” A conservative approach, weighing pros and cons, and watchful waiting for evidence of benefit without harms or unintended consequences does not seem unreasonable and should not be mistaken for dithering.

I agree Canadians deserve nothing less than physicians who use proven effective interventions and technologies to help their patients feel better and live longer. Until e-health has proven itself as safe, effective, and affordable, maybe “festina lente” is rational: make haste, slowly.

Respectfully,

Sumit R. Majumdar, MD MPH FRCPC FACP

Associate Professor, University of Alberta

REFERENCES

1.Chaudry B, Wang J, Wu S et al. Systematic review: impact of health information technology on quality, efficiency, and costs of medical care. Ann Intern Med. 2006;144:742-752

2.Majumdar SR, Soumerai SB. The unhealthy state of health policy research. Health Aff (Millwood). 2009;28:w900-908.

Conflict of Interest:

None declared