A systematic review was undertaken using multiple databases (MEDLINE/PubMed, CINAHL®, EMBASE, PsycINFO) to identify and synthesize all relevant literature published from 1980-2005 An initial broad search was performed with five types of SCI therapies searched: drug therapy, radiotherapy, diet therapy, rehabilitation therapy and therapy. To further refine the search, the search was limited to human subjects and articles published in English.
Based on the above search criteria, the total number of references from all databases was 8007. Two investigators reviewed both the title of the citation and the abstract (of all 8007 references) to determine its suitability for inclusion. Articles’ suitability was based on the above inclusion criteria as well as the following exclusion criteria: less than half the reported population had a spinal cord injury; no measurable outcome associated with treatment; or animal studies. Unless there were no other supporting literature, studies with less than 3 subjects were excluded.
Meta-analyses, systematic reviews and review articles were identified at this point and studies cited with these works that were not identified in the original literature search, were also sought, through hand searching. The review was restricted to published works.
MeSH headings were used with the keywords. Key words were paired with spinal cord injury, tetraplegia, quadriplegia or paraplegia
Specific SCI rehabilitation topics (e.g., pressure ulcers) were identified by a multi-disciplinary team of expert scientists, clinicians, consumers with SCI and policy-makers. These specific topics were searched with additional keywords generated from expert scientists and clinicians in SCI rehabilitation familiar with the topic and more titles and abstracts were reviewed. The reference lists of previous review articles, key articles, systematic reviews and clinical practice guidelines were hand searched. It is known that hand searching may provide higher rates of return than electronic searching within a particular subject area (Hopewell et al. 2002). The number of titles and abstracts reviewed is approximately 8400. Additional keywords used for each specific topic are outlined in Appendix 1.