Background: rehabilitation interventions are usually complex and include different aspects that make difficult to accomplish with classical measures of methodological quality in clinical research. The aim of the present study was to identify, synthesize and categorize the main methodological issues in rehabilitation research to guide the development of methods for reporting and evaluating evidence in the rehabilitation field.
Methods: a scoping review was conducted on the PubMed, Cochrane Library, Embase, Web of Science, Scopus, PEDro and Google Scholar databases up to August 2018. we included methodological studies, special communications and literature reviews addressing any methodological issue in rehabilitation research. For each study, we identified the methodological issues addressed. We conduced quantitative (frequencies of issues addressed in the studies) and qualitative (content analysis of the issues) synthesis.
Results: after removing duplicates, we screened 2879 citations, and 71 studies were finally included. Of these, 69% (n = 49) were narrative reviews, 14% (n = 11) systematic reviews, 7% (n = 5) editorials, 4% (n = 3) meta-epidemiological studies, 3% (n = 2) cross-sectional surveys, 1% (n = 1) mapping reviews, and 1% (n = 1) overviews.
The methodological problems in rehabilitation research included: poor description of data collection and statistical analysis methods (56%); problematic application of randomized controlled trials (38%); description of interventions (35%); the definition of core outcome sets for different clinical problems (31%); lack of blinding of assessors (17%); clinical practice applicability (11%); description of randomization method (10%); description of participants' characteristics and recruitment (8%); methodological and reporting quality (10% vs 8%). Studies also discussed other issues more closely related to the peer review process, such as the methodology training need (7%), low-quality of the peer review process (6%), funding (6%) and ethical statements (3%), lack of protocol registration (3%), and conflict of interest declarations (1%).
Conclusion: this study highlights several methodological and reporting issues in rehabilitation research and in the peer review process. Research looking at methods to improve reporting, as well as to improve the conduct of trials in the rehabilitation field is needed. The first step forwards would be to evaluate the influence of all these issues on the validity of trial results, and the development of a check-list for the evaluation of rehabilitation research specifically.
Patient or healthcare consumer involvement: not applicable