Should O&P Develop Its Own Approach to Evidence-Based Practice?
July 2006 Issue
Judging from trends and predictions, a "perfect storm" would appear to be gathering strength in O&P. Healthcare cost containment will bring major changes none of us is looking forward to, cut-throat competitive practices are a growing concern, and there is the added fear that turf is being lost to other allied healthcare professions or even to sales representatives. Evidence-based clinical practice can be seen emerging through the rain and mist. Presently the image is murky and dim, but as it grows stronger, some will perceive it as a rescue ship coming to help the profession, while others will perceive it as a giant wave having the potential to sink O&P. To truly meet the needs of O&P, the concept of evidence-based practice may need to develop a unique approach of its own within the industry.
The Challenge of a 'Gold Standard'
There are troubling aspects of evidence-based practice that lend credence to the perception that it might be a giant wave. The requirements of "gold standard" randomized controlled clinical trails are virtually impossible to attain in O&P. The need for large sample sizes, the need to blind both practitioner and patient, and the necessity of having only a minimal number of confounding factors which can be controlled for though experimental design, are barriers which can be surmounted for only a limited number of O&P clinical research problems. Funding for large clinical trials and studies is nonexistent anyway.
Also troubling are the studies conducted to date such as on alternative energy-storing prosthetic feet, for example, which often find no significant differences with respect to gait outcomes (1).
Contrasting O&P with other medical disciplines, it may be argued that it is one thing to determine whether one drug therapy or surgical technique reduces a symptom more quickly or to a greater extent than another drug therapy or surgical technique, but it is quite another matter to establish whether one foot performs better than another foot, or one knee performs better than another. Outcomes are contingent upon variables that are unique to each patient.
Patient response to prosthetic feet involves learned motor-control behavior which attempts to compensate for a profound neuromuscular asymmetry, and the asymmetry itself can vary widely from individual to individual. Many subjective and volitional factors, as well the quality of rehabilitation therapy, exert causal influences on outcomes-outcomes involving factors unrelated to the purely physiologic mechanisms often associated with drugs, or the healing processes associated with surgery.
An additional concern is that insurance providers, who are knowledgeable about evidence-based methods in medicine, criticize the O&P profession because it is far behind other medical professions in developing "gold standard" data. Will this lack of high-quality data cause insurers to deny payment for expensive state-of-the-art interventions if gold standard studies cannot be conducted or if less robust studies result in findings of no significant differences? And what if findings are statistically significant but not clinically important? Will the practitioner be bound by study results?
How We Got Here
The origins of evidence-based medicine date back to the 1960s, and rose from a concern about the well-being of patients. There was a growing recognition by physicians that some diagnostic tests and treatments were causing more harm than good. The concept evolved as a reactionary response to a teaching paradigm employed in medical residencies at that time in which eminent practitioners (mentors) dictated diagnostic tests and therapeutic treatments with no need to justify them on the basis of evidence. The authority of the practitioner was cited as reason enough, and it was assumed implicitly that good medical training together with common sense enabled the neophyte physician to be appropriately critical of the literature (2).
What Should EBM Foster in O&P?
What an evidence-based approach should foster in O&P is a questioning attitude with respect to:
What caused the downfall of this paradigm was the recognition that harm was being caused to patients because residents were not being taught to question the effectiveness of the diagnostic tests and therapeutic treatments and to ask for evidence to support the claims their mentors were making. Originally conceptualized as "clinical epidemiology," the evidence-based medicine approach emphasized the need to evaluate the quality of the available data when making decisions in the clinic.
One need is to develop good methodologies for measuring cost-effectiveness that utilize frameworks insurers can understand and will accept, and to explicitly recognize in these frameworks any associated uncertainties. Because most prosthetic and orthotic interventions all work to some degree and inherently produce benefits, harm may not be an issue in O&P to the same extent that it is in many medical specialties. Only a limited set of problems might merit the cost and effort required for controlled clinical trials. Instead, the issue in O&P clinical decision-making, particularly with respect to the selection of a particular prosthetic component from among a family of similar components, often concerns the relative cost-effectiveness of the alternatives.
Thus, the role of the evidence-based approach in mainline prosthetics intervention need not be driven by the need to improve how life-and-death decisions are made or to evaluate the potential for causing serious harm to a patient. Whereas large, randomized, controlled clinical trials are appropriate for determining what works and doesn't work-whether a diagnostic test or treatment does more harm than good on the average-they may not be the best means for evaluating prosthetic components. Cost-effectiveness studies featuring smaller sample populations, and emphasizing a range of performance variables or featuring scoring systems which weight and combine variables, might be more appropriate.
There may, however, be a larger set of problems in orthotics that merit the clinical trials approach, because with orthotic devices, the forces and moments normally applied to living tissues are intentionally being manipulated (increased or decreased) to correct for a neuromuscular imbalance, and this carries with it the potential for causing harm. Orthotic interventions span a wide range of pathologies and design concepts, and at one end of the spectrum shade into the ergonomic devices worn by individuals to protect vulnerable neuromuscular systems from harm when performing work tasks. Clinical trials may be the appropriate means for evaluating them.
The practice of evidence-based medicine requires healthy skepticism and critical thinking. At the core of evidence-based medicine is a desire to know what works and what doesn't work, and why. Controlled clinical trials feature elegant experimental designs intended to answer basic questions about what works. However, determination of cause and effect-the why-is more complicated, and it can involve many additional steps in research design.
In O&P, better cause-and-effect research needs to be undertaken to find answers to the question of why some interventions work better for some individuals than for others. This may mean placing more emphasis on models which examine how and why individuals respond in their own unique ways, and placing less emphasis on models which examine group means (3). The work being done at Northwestern on foot rollover shape is an example of good cause-and-effect research (4). What would enhance the contribution of many biomechanics studies to evidence-based clinical practice are improved definitions of effectiveness, so that cause-and-effect studies can be designed to better address clinically relevant questions concerning cost-effectiveness.
Cause-and-effect relationships for humans entail interactions among highly complex neuromuscular, physiologic, and psyÂchological systems. Human subjects also are enmeshed in and dependent upon sociologic, environmental, and economic systems which exert causal influences. When measured appropriately, patient preferences offer an efficient means of collapsing many of these additional factors into a format that can be managed within research designs that are modest in scale. Preferences form an important category of data not only because they may reveal additional insight into behavior, but because the collection and use of them in clinical decision-making supports the principle of respect for persons. Preferences, along with other data of a subjective nature, merit inclusion as a form of evidence. Effort should be directed toward the development of "gold standard" methods for measuring preferences, so that consumer values would carry weight in the evidence-based approach.
Clinical medicine has shifted toward a probabilistic paradigm. Contemporary clinical reasoning involves the examination of facts, the development of hypotheses, assumptions about cause and effect, and estimates of uncertainty. Even with good evidence, the need to make educated guesses exists. The evidence-based approach forces explicit recognition of the source and quality of the information underlying clinical decisions. Did the evidence come from experience, a colleague's suggestion, a manufacturer, hearsay, or a refereed journal article? To what extent did the research meet a set of agreed-upon standards for evidence? What is the level of uncertainty?
The epidemiologic approach underlying controlled clinical trials is not inherently clinician-friendly because it operates within a traditional framework of deductive logic. A hypothesis is formed, an experiment is designed specifically to test the hypothesis, and on the basis of the resulting data, the hypothesis is accepted or rejected. In a deductive framework, the general results obtained (i.e. from a controlled clinical trial) then can be applied to individual cases (specific patients). In contrast, clinical practice is inherently inductive in nature. Clinicians develop expertise in the opposite direction. Based on experience with a number of specific patients, clinicians tend to reach general conclusions or confirm privately held hypotheses. These hypotheses or conclusions then become the basis for making future decisions. This approach does not always produce a correct hypothesis or conclusion, however.
Abduction Logic Effective
A third form of logic is abduction. It explicitly recognizes the uncertainty and incompleteness of information that underlies clinical decision making, and shows how the best available information can be linked together with the specific needs of individual patients. It lends itself nicely to evidence-based clinical practice, because it requires the clinician to make a distinction between evidence derived from external sources and evidence based on his or her own personal experiences and hypotheses. It requires the practitioner to assess the completeness of the evidence from both sources for a particular patient (5).
In conclusion, O&P must approach evidence-based clinical practice from its unique position within the allied health disciplines, and develop methodologies that serve it best. The skills required in O&P clinical practice combine elements of engineering, medicine, rehabilitation, psychology, and craftsmanship. Each of these disciplines has its own way of perceiving and structuring problems and solution strategies. The evidence-based approach that needs to be developed must be sufficiently broad to be compatible with the world-views of these distinctly different disciplines as they interact, yet precise enough to benefit both patients and clinicians when applied in the context of day-to-day routine decision-making in the clinic.
Edward S. Neumann, PhD, PE, CP, is professor of civil and environmental engineering at the University of Nevada, Las Vegas; director of the Center for Disability and Applied Biomechanics; and adjunct professor of biomedical engineering as well as kinesiology. He has developed and taught courses in orthopedic biomechanics, human motion analysis, prosthetic systems engineering, assistive technology, and ergonomics.
Hafner BJ. Overview of Outcome Measures for the Assessment of Prosthetic Foot and Ankle Components. Proceedings - Outcome Measures in Lower Limb Prosthetics. American Academy of Orthotists and Prosthetists. No 6, Jan 2006, pp 105-112.
Daly J. Evidence-Based Medicine and the Search for a Science of Clinical Care. University of California Press, 2005.
Bates BT, James R, Dufek JS. Single Subject Analysis. Chapter 1 in Innovative Analyses of Human Movement, (edited by N Stergiou). Human Kinetics, Champaign, 2004.
Hansen AH, Sam M, Childress DS. The Effective Foot Length Ratio: A Potential Tool for Characterization and Evaluation of Prosthetic Feet. Journal of Prosthetics and Orthotics. April 2004:16(2), pp 41-45.
Jenicek M, Hitchcock DL. Evidence-Based Practice - Logic and Critical Thinking in Medicine. American Medical Association, 2005.