Volume 47 Number 5, 2010
Pages vii — viii
Probably no fashion has gripped first medicine; then the rest of health care; and finally social services, vocational rehabilitation, education, and other professional fields as quickly and deeply as evidence-based practice (EBP). There have been claims that there is nothing new here, that professionals have always been required to base treatment of their patients/clients on evidence as to what are effective, efficient, and reliable methods of diagnosing/assessing, treating, or offering prognosis, and have done so. However, the core message of EBP has been a worthwhile one whatever one thinks of that claim: not any odd piece of evidence will do. To provide patients/clients with optimal service, one needs to base it on a systematic study of the most recent and most appropriate evidence, carefully evaluated for potential biases and errors. Probably as a result of the prominence of the originators of EBP in medicine, as well as the appealing and easy quantitative and mathematical approaches to evidence evaluation, EBP is now a bandwagon few people dare to disdain publicly. In fact, adding the label "evidence-based" to about anything and everything done professionally is seen as a way of giving it a modern, up-to-date, scientific cachet that will appeal to other professionals, if not patents/clients. EBP has become a holy cow to whom tribute is due.
A case-in-point is a recent paper by Ottomanelli et al. in the Journal of Rehabilitation Research and Development . The authors present the "Methods of a multisite randomized clinical trial of supported employment among veterans with spinal cord injury" and some preliminary results of the study, and in doing so use the term "supported employment" or its abbreviation SE 37 times. In 21 of those instances, the term is preceded by the expression "evidence based," including in the abstract (3 times), in the introduction (4 times), in the methods (11 times, including in the main hypothesis), in the discussion and conclusions (2 times), and even in the acknowledgments (1 time). It makes one wonder what "evidence-based supported employment" (EBSE) is, and how it differs from plain vanilla "supported employment." The authors' report is not helpful here; they use EBSE and SE apparently interchangeably, and when they use EBSE they only seem to take the opportunity to emphasize that SE is an evidence-supported practice.
And here exactly is the rub: either SE is a practice that is supported by evidence or it is not. In the former case, there is no need to do research on whether SE works-there is evidence, and it would be a waste of time and resources to further investigate it, at least in the broad application implied by the authors' hypothesis: "Evidence-based SE will improve competitive employment outcomes and general rehabilitation outcomes significantly more than conventional vocational rehabilitation (i.e., standard care) among veterans with SCI." Indeed, in a recent individual-case meta-analytic study, Campbell et al. conclude, "Rather than additional [randomized controlled trials] comparing [Individual Placement and Support (IPS)] to other vocational models, future research should emphasize enhancing the IPS approach for clients who do not benefit from a trial of supported employment" [2, p. 7]. (IPS is the term Campbell et al. give to SE as Ottomanelli et al. use it.) On the other hand, if there is no evidence for the effectiveness of SE, research is needed. However, then there is no basis for claiming that there is an evidence base, and the term EBSE should be avoided until such time as there is evidence, and not just any evidence, but "proof" at least of a strength and diversity acceptable by minimal EBP standards.
Quite likely Ottomanelli et al. will argue that there is evidence from the severe mental illness field that SE is superior to sheltered workshops, extensive skills training prior to placement, and other traditional vocational rehabilitation approaches (and that therefore the term "evidence-supported" is appropriate), but that there is no evidence that such an approach is effective, let alone superior, for individuals with physical disabilities, specifically spinal cord injury (SCI). They could quote, for instance, Corbi??re et al.: "Given the extensive literature supporting their effectiveness, SE programs are now considered `evidence-based practices' for helping people with severe mental illness obtain competitive jobs . . ." [3, p. 45]. However, making that claim is admitting there is no evidence for SE in SCI, and consequently using the expression EBSE for SCI is inappropriate. One cannot have the cake and eat it too.
The problem, of course, originates in the authors' and many other researchers' and clinicians' desire to see their work aligned with the EBP movement. However, we should be careful not to turn EBP into a holy cow. Not everything that claims the label EBP satisfies the minimal requirements. We would do better to acknowledge that we have no or very weak evidence rather than slapping the label EBP on everything. That will just weaken our practices, rather than strengthen them as the founders of EBP intended.
Dr. Dijkers makes a valid point about the importance of terminology. He is correct that the term evidence-based supported employment (EBSE) is justified for persons with serious mental illnesses. Abundant evidence for the use of supported employment in persons with serious mental illness was produced as a result of a number of randomized controlled trials (for a recent review see Bond et al. ). The generic term "Supported Employment" refers to a defined model within the Rehabilitation Services Administration (RSA) and Department of Education (DOE) for persons with disabilities. It was developed and implemented in the 1980s through today. The RSA/DOE model is not an evidence-based model, but it is widely used in community rehabilitation.
In describing the methods of our study, we sought to communicate that we used an intervention that adheres to the evidence-based supported employment principles for serious mental illness among a population of persons with spinal cord injury (SCI). We were not suggesting that at this point there is sufficient evidence to use the term EBSE for populations other then severe mental illness, including persons with SCI. We hope that the outcome data from our study will provide useful information on whether this model is applicable to a new population of persons with disabilities. We appreciate the opportunity to respond and clarify our use of these terms.
Go to TOP
Last Reviewed or Updated Thursday, July 29, 2010 8:37 AM