Is the science of medicine trumping the art of medicine?

In health care today there is a major emphasis on the consistent application of evidence-based elements of care. This shows up in pay-for-performance initiatives, the Joint Commission on Accreditation of Healthcare Organization's accreditation process and the Institute for Healthcare Improvement's "100,000 Lives Campaign."

A recent essay regarding the need to drive system changes that will increase safety appeared in the Annals of Internal Medicine (1) and high on the list of barriers to achieving a safer patient care environment were problems with "worker discretion" and "autonomy." (1)

Practicing evidence-based medicine (EBM) is essential for demonstrating to society that health care professionals can predictably apply the science of medicine. Moreover, it is necessary if medicine is to have any hopes of maintaining its identity as a profession.

Yet across the United States, the physician community has significantly resisted these initiatives. Preservation of individual physician autonomy remains its transcendent value. The emphasis on maintaining individual physician autonomy denies the growing complexity of health care and resists acknowledging the professional interdependencies that are essential to maximizing patient safety and quality of patient care.

The metaphor that best describes the physician's view of teamwork is golf, a game where team success is the additive sum of individual success. What is necessary is a volleyball metaphor, a sport where role clarity coexists with role interdependency. In golf the whole equals the sum of its parts. In volleyball the whole can exceed the sum of its parts.

Appropriate application

The value of the appropriate application of evidence-based medicine, paired as it often is with a long overdue focus on patient safety, cannot be overemphasized. And there is no question that self-interest and worker discretion must bend to higher values. (2)

Proponents of evidence-based clinical guidelines are increasingly outspoken about the importance of minimizing "process variation,' exhorting caregivers to set their sights on 6 Sigma goals--1 part per million variation. (1)

This evolution toward guidelines that are less like suggestions and more like commandments has its roots in two very different phenomena.

1. On the one hand, the data are increasingly convincing that rigid compliance with guidelines works well to minimize morbidity in a number of highly lethal conditions--DVT, postoperative infections, etc.

2. On the other hand, stubborn resistance from those advocating autonomy for its own sake elicits an increasingly strident reaction from what we might call "the EBM community."

We are concerned, however, that the payer and regulator communities, as well as our colleagues who are leading the long-awaited charge for patient safety, may see compliance with these initiatives as a problem to be resolved rather than a polarity to be managed.

It is as though, in direct contradistinction to stubborn autonomy, compliance with evidence-based medicine is being advanced as an absolute value in American medicine.

Without arguing "against" evidence-based medicine, we would like to suggest four boundaries that define its current limits. This exercise should position us to embrace EBM in a reasoned and modulated fashion, assuring that we guide our colleagues and students toward using and advancing the scientific basis of our field in a way that also honors other critical dimensions of medical practice.

[ILLUSTRATION OMITTED]

Inherently conflicting goals

Pain management provides a useful illustration. Currently there is a national emphasis on "eliminating" pain. Our patients are asked to rate their pain using smiley faces and the intent is to avoid any frown.

The communicated expectation is that the patient should remain pain free, and that pain is more easily controlled if anticipated rather than "chased." One frequent consequence is an increased use of opiate antagonists to resuscitate over-sedated patients.

Contrast that to an approach that uses a sedation assessment as the metric for evaluation. In this scenario one can manage patient expectations to accept some pain as a natural consequence and educate the patient to the downside consequences of excessive opiate use.

The tradeoff is virtual elimination of a need for opiate antagonists with acceptance of a necessary measure of discomfort. The point is that the metric used influences the outcome. Might it be appropriate to collaboratively titrate both pain and opiate use in the setting of the doctor-patient relationship, with an emphasis on managing patient expectations rather than rigidly committing to "nothing but smiley faces?"

The pain management situation is not unique. Other examples of clinical situations containing inherently conflicting goals include:

* End-of-life care, where survival, quality of life and quality of death are all valid concerns

* Decisions regarding whether to declare a patient "Do Not Resuscitate"

* How aggressively to treat advanced cancer in the face of patient ambivalence

While EBM offers enormous power when goals are undeniable and technical perfection theoretically possible (preventing DVT), situations attended by ambiguity and conflicting choices require a different set of provider skills.

Author Ronald Heifitz' distinction between problems requiring technical solutions and problems requiring adaptive solutions may offer guidance here. (3) To the extent that we can isolate aspects of the clinical encounter that are purely technical, we are on safer ground advocating unyielding compliance with EBM.

But if we are in a realm of less certain and more complex challenges, we need different guidelines--gentle guidelines regarding "best process" rather than rigorous guidelines about best practice.

Advancing clinical medicine

Demanding adherence to prescriptive practices risks compromising creativity and discovery. While acknowledging and supporting the upside of prescriptive patient care processes and the downside of an emphasis solely on a craftsman's approach to medical practice, there is the potential to advance medicine by breaking rules that can constrain medicine through unyielding adherence.

This is a true polarity where the entrenched support of one pole to the exclusion of the other leads to a circumstance where the downside of the advocated pole becomes dominant. (4)

Atul Gawande describes the commitment of Warren Warwick, MD, director of the Cystic Fibrosis Center at Fairview-University Children's Hospital in Minneapolis, to advance the care of patients with cystic fibrosis. (5)

The results achieved by Warwick and his team have been consistently superior to those expected by adherence to the evidence-based guidelines currently accepted in the field. Warwick approaches existing evidence-based guidelines as a baseline, rather than an absolute. He cautions against accepting our notion of "best practice" as anything other than "best practice according to current knowledge," and urges us to push further for more powerful results.

When, as in this work, excellence reflects performance outside two standard deviations of the mean, it is by definition a reflection of deviant behavior. (6) If reimbursement were tied to compliance, rather than to outcomes, the deviant physician would in fact be penalized for doing the potentially better thing, and the lessons of innovation would go unlearned.

What protects us against idiosyncratic practice that increases risk to patients with little promise of gain? Here, we have solid structures in place--the peer review process, the IRB process, and federal guidelines for grant review, all of which can serve to protect us by encouraging innovation based only upon cogent hypotheses and solid safeguards.

Appropriately individualizing care

Another risk of algorithmic medicine is that it may drive toward impersonal care--care in which rules take precedence over individuality. Our public already rails against care that has become so technologically focused as to be virtually "dehumanized." (7)

A groundswell of physician emphasis on lifestyle needs, payment for call, and using paraprofessionals to do pre- and post-operative care, together with a growing reliance on data generated outside of direct patient contact, can render the fundamental, humanizing interaction between doctor and patient virtually incidental to the practice of medicine.

A physician can access the database from home over the Internet, view digitalized images, respond to prompts and edits, and submit electronic orders without ever touching or directly communicating with the patient. E-ICU's are the current cutting edge of this movement toward efficiency.

The science of medicine risks trumping the art of medicine, efficiency risks trumping compassionate communication. The unique clinical needs of the patient, not to mention every patient's right to be regarded by the care-giving team as a unique individual, are all vulnerable.

Our commitment to perfectibility must be matched by an equally tenacious commitment to patients as individuals who suffer, and whose clinical difficulties may indeed be unique variants of more universal conditions. Efficiency is appropriate to technical processes. It should never apply to the doctor/patient relationship.

All experienced clinicians have met patients who "know" something about some hidden and undiagnosed pathophysiology. Our plea is for an EBM humble enough to forestall prescriptive care when wisdom--even intuitive wisdom--suggests that the patient's immediate need is not "process constancy" but to be heard and treated individually.

Again, we see this as a polarity to be managed--between the wisdom of scientifically based consistency on the one hand, and the wisdom of a humility that can cause us to question our own certainty on the other.

Maintaining the doctor's presence

The art of history taking and physical diagnosis are giving way to the primary application of technology.

Abdominal pain equals CT scan, chest pain demands heart catheterization, side effects of medications demand more medications rather than a reassessment of the patient.

With each breakthrough, we are further removed from the private encounter of sufferer and healer. The issue here has nothing to do with a Luddite desire to return to some romanticized past, but a desire instead to rescue the power and meaning of the doctor/patient relationship, and assert that we need not forgo human contact in our pursuit of science or our embrace of technology.

Not only is the doctor/patient relationship potentially therapeutic, it is the ultimate source of joy that attends the practice of medicine. Relating to patients in a transactional manner reduces the intangible rewards of the profession.

Technical competency, no matter how expertly applied, in the absence of human context is singularly without joy.

We are at risk of confusing two very separate issues:

* The march of science, which does indeed allow increased precision

* The substitution of technology for caretaking, which threatens the soul of medicine

Increasingly, with hospitalists, intensivists and a generation of young physicians asserting their prerogative to a balanced lifestyle, the idea of "owning" a patient is fading from view.

Ownership, in this context, is all about responsibility, not about paternalism or objectification. (8) Our argument is for the steady advance and application of science (but not its deification) in a context that equally values and respects individualized human needs. It is the distinction between curing and healing.

Should we use evidence-based medicine? Should we address autonomy as a problem when it threatens patient safety? The answer to both questions is an unqualified yes. When to insist on EBM, how to constrain autonomy and how to identify what limits the perfectibility of clinical medicine are more difficult questions to answer.

There may indeed be aspects of medical practice that can be known with near-absolute certainty and practiced with 6-Sigma perfection, but we believe that these are likely to be "slices" of clinical medicine only--bundles of interventions applied in conditions of relatively linear clarity and in the absence of profound system complexity.

This line of reasoning suggests, in some contradistinction to Alemberti's thesis, that progress in the field not only involves "perfecting" our application of abundantly defensible pathways, but developing ever-increasing skill at the appropriate integration of linear processes (such as pathways) in a world that can never be reduced to the purely linear.

Our arguments are meant to bolster what might seem obvious, yet, in polar arguments, the obvious is all too often lost. For us, the embrace of scientific medicine and the preservation of the art of medicine are not an either/or argument, but rather a both/and issue, a polarity to be managed and not a problem to be solved.

Hopefully the push toward an emphasis on evidence-based medicine, public reporting of performance data and pay for performance won't serve to extinguish the art of medicine, devalue transformational doctor/patient relationships or penalize creativity.

Those aspects of the physician's role are intrinsic to the joy and privilege that attend this unique profession, and live, always, in some tension with our obligation to make rigorous use of "best science."

This tension is best honored by the exercise of humility when we find ourselves at the outer frontiers of statistical knowledge and defensible certainty.

Joseph Bujak, MD, FACP, is vice president of medical affairs for Kootenai ter, Coeur D'Alene, Idabo. He can be reached at 208-666-2014 or [e-mail]

[ILLUSTRATION OMITTED]

Eric Lister, MD, is the managing director of Ki Associates, a consulting firm dedicated to working with the boards and executive leadership teams of hospitals and physician practices. He can be reached at 603-433-2305 or [e-mail].

[ILLUSTRATION OMITTED]

References:

1. Amalberti R, Auroy Y, Berwick D, Barach P. "Five system barriers to achieving ultrasafe health care." Ann Intern Med. 2005 May 3; 142(9) May 3, 2005.

2. Lister ED. "Profession versus guild." Health Aff (Millwood). 24(2) Jul-Aug, 2005.

3. Heifetz R. "Leadership without easy answers." Interview by Joe Flower. Healthc Forum J. 38(4) Jul-Aug, 1995.

4. Johnson B. Polarity Management: Identifying and Managing Unsolvable Problems. Amherst, Mass; HRD Press, 1996.

5. Gawande A. "When doctors make mistakes." The New Yorker, February 1, 1999.

6. Quinn RE. Deep Change: Discovering the Leader Within. San Francisco; Jossey-Bass, 1996.

7. Carey B. "In the hospital, a degrading shift from person to patient." New York Times. August 16, 2005

8. Stephanie Starks Hope Foundation. Beyond the Best Interest of a Child National Initiative. http://sshopefoundation.org/sshf_child/

By Joseph S. Bujak, MD, FACP, and Eric Lister, MD