Abstract
Dear Editor,
Geertzen et al. (2004) raised important issues regarding the design, conduct and reporting of clinical trials in prosthetics. The general issues are of course also relevant to orthotics and it is on this subject I would like to expand the discussion in the journal.
There have been numerous examples of excellent technical orthotic research and innovative solutions to biomechanical problems. There has however been a dearth of systematic and unbiased evaluations of the clinical effectiveness of most orthoses. In an editorial in 1993, Lehneis mused on the state of the art in orthotics (Lehneis 1993). Paraphrasing a colleague he proposed that “the ideal orthosis replaces function, is weightless, invisible and costs nothing”. Simply reversing this statement, the worst orthosis impedes function, is cumbersome, unsightly and expensive. The importance of evaluating the relative benefits from using interventions against any possible harm, including psychosocial effects, has been reinforced with the era of evidence-based health care (Sackett et al. 1996).
An increasing body of orthotic research is already published. Critical appraisal, such as the constructive comments made by Geertzen et al. (2004), and judicious application of this evidence can inform clinical practice and identify topics requiring further research. The report of the consensus conference organized by ISPO on the lower limb orthotic management of children with cerebral palsy concluded “The existing body of literature on the effects of orthotic intervention in cerebral palsy is, for the most part, seriously scientifically and experimentally flawed” (Condie and Meadows 1995 p 5). In 2003, ISPO organized another in its series of consensus conferences, this time on the orthotic management of stroke. This meeting similarly found that the methodologies employed by researchers were largely unable to answer the questions being posed. The effectiveness of orthotic intervention for stroke was based on small numbers of subjects, too short follow up periods and often an inadequate control for comparison (Morris 2004).
An issue of current debate in paediatric orthotics is whether “cranial remoulding orthoses” improve head shape for infants with non-synostotic plagiocephaly better than “repositioning and advice”. Two reviews of the evidence have been published. In the UK the reviewers conclude categorically that treatment with helmets is not effective and the treatment is rarely offered in the UK (Bridges et al. 2002). In North America however, where helmets treatment is more common, the American Academy of Pediatricians concluded that the issue is unclear and would benefit from further research especially to justify the cost of treatment (Persing et al. 2003). This prevailing uncertainty would seem to provide ideal circumstance for a clinical trial; however this would require clinicians with strongly held professional preferences on either side to shift towards the consensual equipoise.
The current challenge is finding out how we can improve clinical research in orthotics. Firstly, we must ask clear and important questions about the effectiveness of orthoses about which clinicians and patients are uncertain. Good clinical questions consist of (i) a group of well defined patients, by diagnosis or clinical problem; (ii) plainly described orthoses, using generic terms and incorporating the biomechanical effect; and (iii) simple measurable outcomes. Secondly, in the absence of compelling evidence to guide practice, we must design experiments using an appropriate research methodology that is capable of providing answers to clinical questions by measuring outcomes at an appropriate time. Rigorous clinical research requires the avoidance of bias at all stages in the selection of patients, choice of an appropriate comparison, how treatment is allocated, who does the assessment and what outcome is measured, and also how the study is reported (The James Lind Library 2004).
Jannick et al.(2004) have argued quite correctly that the outcomes measured in orthotic research should include the wider aspects of the usability of the orthosis, efficiency, satisfaction and context of use, rather than focussing only on effectiveness. Standardized patient reported outcomes are increasingly used as key outcome measures in a variety of clinical research including orthopaedics. Further work is clearly needed to formalize the appropriate steps in designing rigorous orthotic studies.
There are individual and institutional barriers to the evolution of clinical research in orthotics. There is an inherent resistance among some clinicians to accept that research in orthotics is either possible or necessary. There is also lack of research skills among orthotists as training is necessarily stronger in biomechanics and practical skill rather than clinical epidemiology. The incentive of demonstrating the effectiveness of interventions, as opposed to safety, in order to obtain a licence from country-specific regulatory agencies to market new medical devices is absent. The infrastructure required to support multi-site research is also lacking.
Reassuringly there are also many facilitating factors. Purchasers of health care are increasingly unwilling to pay for interventions for which there is no evidence of effectiveness. The orthotic profession is becoming increasingly scientific and has formally accredited academic training in many countries. These courses must include training in searching for and critically appraising the literature as well as teaching clinical as well as technical research skills. The Journal of Prosthetics and Orthotics included a series of educational articles focussed on teaching research skills between 1996–7. An excellent editorial by Childress (2002) acknowledges the increasing scientific basis for orthotics and notes the opportunity afforded to interested clinicians at this time.
It is a time of opportunities. Orthotists and those who prescribe orthoses need to recognize their uncertainty about the effectiveness of specific applications of orthoses, as these are opportunities for research. The specific challenges for clinical research in orthotics should not deter collaboration in large simple studies as part of routine clinical practice. The defining of simply measurable outcomes for orthotic intervention is a priority. The establishment of national and international networks of centres willing to collaborate in clinical research would overcome some of the infrastructure problems and enable the science of orthotics to evolve further and faster in the future.
Yours sincerely