The Promise (the Tyranny?): Some Observations on the Evolution of Evidence-Based Medicine

By Elizabeth Bogdan-Lovis

“She don’t lie, she don’t lie,
she don’t lie–COCHRANE!”
(Refrain from the song “Cochrane” performed by Argentinean physician and Cochrane reviewer, Agustín Ciapponi, at the 2003 Cochrane Colloquium end-of-conference social event.)

A little more than ten years ago a working group at Canada’s McMaster University suggested evidence-based medicine (EBM) as a new approach to teaching and practicing medicine (Evidence-Based Medicine Working Group 1992). EBM is defined as “the conscientious, explicit, and judicious use of the best evidence in making decisions about the care of individual patients” (Sackett 1996: 312). That foundational quote illustrates why EBM carries legitimized political and professional capital among all health care stakeholders, including medical educators, clinicians, patients, governments, and third party payers. EBM is anchored in the grand narrative of modernist scientific medicine. Under the best circumstances, EBM employs science’s empirical testing and measuring tools to free humankind from old fogyism, all the while remaining firm in the belief that this numbered knowledge of the world produces tangible benefits. Lured by this siren’s song of modernity, who can resist the seduction of applying the best science to effectively and efficiently address health needs?

EBM is indeed a new medical paradigm. Although there remain outliers still dubious of its utility, my reliable crystal ball tells me that there is a very promising future for the EBM endeavor, especially for high quality EBM work such as that produced by the Cochrane Collaboration, the ACP Journal Club, the Journal of Evidence-Based Health Care, and the like. That being said, it’s not surprising that relatively rapid, large-scale implementation of EBM has exposed contextual application issues requiring attention.

These issues include the ongoing problem of phronesis, i.e., how to best wed patient-oriented evidence to clinical expertise and judgment; how to effectively extrapolate information from population-based randomized clinical trials and apply it to the individual patient; how to steer the international biomedical research agenda and associated EBM to effectively address prioritized international health needs; and how to incorporate qualitative data into the EBM equation (a task that is being undertaken by the nascent the Campbell Collaboration). In an effort to continually improve EBM, those physicians, biomedical researchers, biomedical statisticians, medical librarians, and policymakers who enthusiastically promote diffusion of this innovation have addressed these and related dilemmas as they emerge; as such it is not my intent to speak to those issues here.

Instead, I will consider the evolution of EBM from the dual perspectives of anthropology and my past life of birth activism. The first (anthropological) perspective allows me to critically examine influential social dynamics within the health care delivery system, paying close attention to the distribution of power in social relations, in clinical authority, and in knowledge production. The complexities of social location and social interactions shape our perspectives and perceptions of the world around us, influencing what we see as risks, as problems, and as solutions. Here, I examine the influence of social context on the evolving practice of EBM.

My second reflexive perspective draws on my long history of birth activism–subtitled “Everything I know today I learned in childbirth preparation classes.” In the 1980s I taught a renegade style of childbirth preparation class. In place of teaching the customary (for the day) Lamaze-style breathing gymnastics, which taught women to be obedient patients, I instead instructed pregnant clients on rebellious strategies to minimize their chances of having an unnecessarily medicalized birth experience. I should add here that I retrospectively view that goal as simplistic and naïve. My clientele included a demographically-skewed, disproportionately large number of physicians, medical residents, medical students, and nurses. As a consequence of their professional orientation, this group was notably conversant with what might be considered rather esoteric medical knowledge. In short, while their social location did not quite equalize the usual provider/patient power differential, it certainly influenced their relationships with their chosen physicians. A curious dynamic emerged.

Not uncommonly, my medical-professional clientele desired a manner of demedicalized birth management that differed from standard institutional hospital fare–such as a desire to avoid an episiotomy, or to birth in something other than the “flat on your back” lithotomy position. Given their relatively privileged insights into the medical world, and knowing that they faced likely resistance from their provider, these pregnant professional women recognized the need to support their unconventional protocol requests in a conventional fashion. They thus armed themselves with supporting medical research, practicing a nascent sort of evidence-based patient choice (Edwards & Elwyn 2001).

They guessed correctly. Their providers did indeed often respond defensively, manifesting a “biased assimilation effect”–described by psychologists as one’s tendency to look harder for flaws in research with which one disagrees. Seeking weaknesses in the women’s proffered research articles, their providers would cite medical research to support their own routinized medical protocols. A patterned thrust and parry of data vs. data ensued. The metaphorical duel was usually called to a draw when the provider successfully assumed authoritative power with statements such as “if you’d seen what I’ve seen” illustrated with a sufficiently convincing worst-case scenario, or more simply “this is the way I do it.”

I offer this dated description of such fractious confrontations to illustrate the dynamic, complex, and relational nature of knowledge production, the subjectivity of scientific medical research in the clinical context, and the socially-located trump card of professional authority. In this 1980s climate, the parallel knowledge base, summarized here as “birth works” was considered by providers to be woefully ignorant of real medical matters. Social theorist Pierre Bourdieu (1980) would use the term habitus to describe the function of power relationships within this state of affairs. According to Bourdieu, habitus is the social location, the social lens, and the accompanying social structure, that enables an individual to perceive activities in a particular order and in a particular way, and to evaluate them accordingly. The more one is enmeshed within one’s habitus, the more they then conceptualize, interpret, and make sense of surrounding structured activities through that lens. Habitus subtly shapes a tacit tolerance for the status quo and is in turn structured by it (Callinicos 1999: 292-3).

In truth, at the time, neither the birth works nor the standard medical model had been sufficiently evaluated by a systematic review of medical research to adequately establish where the weight of the evidence lay. Fast forward to the end of that decade and enter the EBM research data organization and application tool.

EBM disciples will be familiar with the following piece of the EBM origin myth. Epidemiologist and visionary founding father of EBM, Archie Cochrane, suggested that modern obstetrics in particular deserved the “Wooden Spoon Award” for including a multitude of routine clinical interventions lacking a sufficient, scientifically tested, evidentiary base to justify their practice, and whose hazards were unknown (Cochrane 1979). As a response to “Archie’s” clarion call, a collaborative multinational working group set about the task of systematically identifying and evaluating research on childbirth management. The results of their work, published in the two-volume Effective Care in Pregnancy and Childbirth (Chalmers, Enkin, & Kierse 1989), demonstrated that critically appraised research supported a social model of childbirth management (birth works) over the standard medical model. Yet, as we know, standard medical management continues to define and shape hospital birth.

Equally relevant to this discussion is the observation that most pregnant women actively seek medically managed hospital birth. Advocates of patient-oriented evidence-based decision-making suggest that patients can and should be effectively engaged in certain aspects of clinical decision-making (Edwards & Elwyn 2001). Available access to EBM databases might potentially lead to increased democratization of what was formerly professionally-owned esoteric medical knowledge. But habitus also influences socially embedded patient consideration of what constitutes best care.

So what are we to make of such disconnects? Perhaps modern society's relationship to health care services is similar to its relationship with shopping. In his book I Want That! How We All Became Shoppers (2002), cultural historian Thomas Hine observes that shopping practices reflect our sense of entitlement for increasingly sophisticated commodities. Hine tells us "Shopping is a responsibility, an exercise of power…the choices that we make about these things play a big role in determining who we are. We don't want to surrender such power to a computer program . . ." (p. 208). By comparison, perhaps media-steered health care consumers, and industry-influenced health care providers alike, perceive the most recent medical technologies as delivering optimal care. Encouraged by popular representations which equate "newer, bigger and more" with better, perhaps we selectively view ourselves as entitled to pick and choose among various health services - health care resource thresholds not withstanding. Quite possibly, we will be unwilling to surrender such consumer power to the vagaries of EBM number crunchers. Patient-oriented EBM be damned?

Alternatively, perhaps the aforementioned blindness is caused by "cognitive bias" which gives an illusion of validity - discounting information that goes against one's preferred conclusion. One can hypothesize that EBM might offer a valuable corrective here - it has the potential to objectively assess the cumulative merit of the research, and with such an objective assessment, produce robust knowledge to counter cognitive bias. And in certain cases, EBM has done just that. My colleague Howard Brody reminds me that located within EBM's lore there is the pivotal Cardiac Arrhythmia Suppression Trial (CAST). Here, despite plausible pathophysiological reasoning on the soundness of administering antiarrhythmia drugs encainide and flecainide for survivors of myocardial infarction, definitive outcome evaluation demonstrated otherwise. The CAST study found that even though such therapy successfully suppressed asymptomatic or mildly symptomatic ventricular arrhythmias, it ultimately accounted for a higher total mortality (CAST 1989). In this case, outcome data clearly impacted treatment; it triumphed over cognitive bias and successfully influenced subsequent clinical management.

So just why is it that doctors and patients sometimes capriciously discard a working hypothesis because of new data, while at other times they cling to a hypothesis despite evidence to the contrary? Nineteenth-century philosopher William James posited that one set of looks can indeed overcome another way of looking. But he also noted the countervailing tendency. "Most of us grow more and more enslaved to the stock conceptions with which we have once become familiar, and less and less capable of assimilating impressions in any but the old ways. Old fogyism, in short is the inevitable terminus to which life sweeps us on. Objects which violate our established habits of 'apperception' are simply not taken account of at all; or, if on some occasion we are forced by dint of argument to admit their existence, twenty-four hours later the admission is as if it were not, and every trace of the unassimilable truth has vanished from our thought" (James 1950, c 1918).

In the case of childbirth management, EBM-supported reasoning is sacrificed to pragmatic reasoning that sees truth in its consequences. Returning to my birth works example, it is casually observed that medicalized childbirth results in healthy mothers and healthy babies, so it is concluded that medicalizing birth works. The prevailing medical gaze interprets childbirth as risky; the resultant cultural prescription then views birth as in need of precautionary "just in case, just to be safe…" medicalization. According to this popular cultural metanarrative birth does not work; it is instead an untidy experience that predictably flirts with disaster, and only when aided by bigger and presumably better medical technologies and more intervention does it then emerge triumphant.

Problematically, reliance on the body of systematic reviews of relevant research argues more convincingly for how to proceed with medical interventions than whether to proceed with them in the first place. This curious paradox is linked in part to human nature, and human perception. "If people want to believe there is an effect, it can be very hard to persuade them that any effect is too small to be important" (Alderson & Groves 2004: 473).

Yet another confounding obstacle to implementing effective and efficient EBM policy is the powerful influence of stories - especially those stories with "tragic outcomes…[that] seem to offer a solution" (Newman 2003: 1426). In his article "The Power of Stories Over Statistics" epidemiologist and biostatistician Thomas Newman describes the compelling power of stories, noting how an especially poignant story might make statistical probabilities excessively vivid, and in so doing, inadvertently distort the appropriate application of the statistical data (Newman 2003: 1426). This dynamic might partially explain why the American College of Obstetricians and Gynecologists has recently been willing to entertain the notion of patient choice for elective primary cesarean section (Bump 2002: 823).

The 2003 Institute of Medicine report Health Professions Education: A Bridge to Quality suggests that "it is critical for interdisciplinary health teams and each of the disciplines to be able to tap this evidence base effectively at the point of patient care, determining whether an intervention, such as a preventive service, diagnostic test, or therapy, can be expected to produce better outcomes than alternatives - including the alternative of doing nothing (emphasis mine)" (Institute of Medicine 2003: 56). Ironically, EBM's pattern of reliance on the results of randomized controlled trials focuses the discourse more on doing something than on doing nothing. Thus, it is the case that the medical management model as well as societal opinion are equally biased towards action over inaction.

Returning to my example, childbirth management screening, testing, and measuring are socially and medically reified as the best way to know birth, and within this empirical model a parallel birth works model lacks equivalent authority; it is viewed as backward, less desirable, and unsafe. As two general practitioners cogently point out "perhaps it is societal opinion (for which one ear of the medical profession is always pricked) that sins of omission are more reprehensible than errors of commission that is at fault. Is missing a rare diagnosis so much worse than harm from over-testing?" (Doust & Del Mar 2004: 474). The social response appears to be a deafening "yes."

While early EBM argued for a social model of childbirth management, a review of current childbirth management practices demonstrates that EBM failed to effectively challenge societal opinion. Paradoxically, expanding reliance on randomized controlled trials to empirically know birth, such as is required by EBM, now subtly but surely steers birth towards an increasingly medicalized model - the tyranny?

Perhaps my concerns about the evolution of EBM are accurate only when considering unnecessarily medicalized aspects of life and quite possibly EBM is the very best corrective that we have to offer in that epistemological regard; but in today's increasingly medicalized world I simply don't trust that we can accurately distinguish the difference. Medical facts and social values are intimately intertwined. Medical science is dialectical, structuring societal opinion while in turn being structured by it.

 

·Like photography, EBM is "a way of depersonalizing our relation to the" patient (p.167).
·Like photography, "it is a way of appropriating the objective world" (p.122).
·Like photography, EBM "can be seen as a faithful recording of what is evident" (p.118).
·Like photography, "it is a way of finding a place in the world … to relate to it with detachment" (p.167).
·Just as photography turns living beings into things, so, too, EBM turns embodied processes of the doctor-patient interaction into commodified things, things to be measured and recorded (p.98).
·EBM "creates another habit of seeing - both intense and cool, solicitous and detached, charmed by the insignificant details, addicted to incongruity/congruity" (p.99).
·In photography it is only by looking at reality in the form of an object does it become real. So too, EBM requires that aspects of a healing encounter be subjected to testing, measuring, and within such a robust numbered equation, the option of "doing nothing" pales by comparison (p.168).
·Like photography, EBM "cannot create a moral position, but [it] can reinforce one and can help build a nascent one" (p.17).
·Like photography, EBM "may be more memorable than moving images, real people, because [it is] a neat slice of time, not a flow" (p.17).
·EBM is in "a chronic voyeuristic relation to the world which levels the meaning of all events" (p.11).
·To reduce an aspect of the healing encounter to the level of a systematic review is to "appropriate" [it]…It means putting [it] into a certain relation to the world that feels like knowledge-and, therefore, like power" (p.4).
·EBM "alters and enlarges our notions of what is worth looking at and what we have a right to observe. [It is] a grammar and, even more importantly, an ethics of seeing" (p.3).
· EBM "can convey some kind of stable meaning, [it] can reveal truth" (p.106).

Elizabeth Bogdan-Lovis, MA
Assistant Director, the Center for Ethics and
Humanities in the Life Sciences

Citations

Alderson, P. and T. Groves. (2004) What doesn't work and how to show it: Ineffectiveness is hard to prove and accept. BMJ 328: 473.
Bourdieu, P. (1990) The Logic of Practice (Translated by Richard Nice). Cambridge: Polity Press.
Bump, R. (2002) Advising prospective mothers about the maternal morbidity of vaginal childbirth. Am J of Obstet and Gynecol 187: 823.
Callinicos, A. (1999) Social Theory: A Historical Introduction. New York: New York University Press.
Campbell Collaboration http://www.campbellcollaboration.org/index.html
CAST (1989) Preliminary report: effect of encainide and flecainide on mortality in a
randomized trial of arrhythmia suppression after myocardial infarction. The Cardiac Arrhythmia Suppression Trial (CAST) Investigators. N Engl J Med 321: 406-412.
Chalmers, I., M. Enkin, M.J.N.C. Keirse. (1989) eds. Effective Care in Pregnancy and
Childbirth. Oxford: Oxford University Press.
Cochrane, A. L. (1979) 1931-1971: A critical review, with particular reference to the
medical profession. In Medicines for the Year 2000, pp.1-11. Office of Health Economics London.
Cochrane Collaboration http://www.cochrane.org/index1.htm
Doust, J. and C. Del Mar. (2004) Why do doctors use treatments that do not work? (Editorial) BMJ 328: 474.
Edwards, A. and G. Elwyn. (2001) Evidence-based Patient Choice: Inevitable Impossible? New York: Oxford University Press.
Evidence-Based Medicine Working Group. (1992) Evidence-based medicine. A new
approach to teaching the practice of medicine. JAMA 268: 2420-5.
Hine, T. (2002) I Want That! How We All Became Shoppers. New York: Harper Collins.
Institute of Medicine. (2003) Health Professions Education: A Bridge to Quality.
Washington DC: National Academy Press.
James, W. (1950, c 1918) The Principles of Psychology. New York: Dover Publications.
Newman, T. (2003) The power of stories over statistics. BMJ 327: 1424-27.
Sackett, D.L., W. Rosenberg, M. Gray, B. Haynes, & S. Richardson. (1996).
Evidence-based medicine: What it is and what it isn't. BMJ 312: 71-72.
Sontag, S. (1973) On Photography. New York: Farrar, Straus & Giroux.
Stapleton H, Kirkham M, Thomas G. (2002) Qualitative study of evidence based leaflets
in maternity care. BMJ 324: 639.