RESEARCH REPORT POSTER DISPLAY

PSYCHOMETRIC EVALUATION OF WORK DISABILITY OUTCOME MEASURES FOR INDIVIDUALS WITH MUSCULOSKELETAL INJURIES: A SYSTEMATIC REVIEW. Williams R¹, Allwood S, Sanchez M, Shea R, Wang J, Wark G; ¹School of Rehabilitation Science, McMaster University, Hamilton, Ontario. Canada

PURPOSE: This systematic review evaluated the psychometric properties of outcome measures used to assess work disability in individuals with musculoskeletal injuries. RELEVANCE: Although there are several measures of work disability, they have not been adequately evaluated. PARTICIPANTS: Three pairs of reviewers evaluated the articles. METHODS: MEDLINE, CINAHL (Cumulative Index to Nursing and Allied Health Literature), EMBASE, AMED (Allied and Complementary Medicine), and PsycINFO between 1980 to March 2006 were searched for peer-reviewed studies that evaluated self-administered instruments that assessed work disability for individuals with musculoskeletal injuries. Examples of key words were “work disability”, “worker role”, “return to work”, “instrument validation”, “validity”, and “reliability”. The eligibility criteria were articles that: (1) were peer-reviewed, (2) were published in English, (3) involved an evaluative component of a work disability outcome measure, and (4) were conducted with individuals who had sustained musculoskeletal injuries. References cited in the retrieved articles also were screened for possible inclusion. ANALYSIS: The reviewers independently evaluated the outcome measures for the following psychometric properties: (1) content validity, (2) construct validity, (3) internal consistency, (4) test-retest reliability, (5) responsiveness, (6) time to administer, and (7) ease of scoring. A modified rating form was used to assess the instruments. A scoring method was developed whereby a value was assigned for each psychometric property evaluated (e.g., “1” was given for a “+” (adequate methods/results); “0.5” for a “±” (doubtful methods/results); and “0” for either a “−” (inadequate methods/results) or “?” (no data available)). The total possible score was 7. When discrepancies occurred between reviewers in the rating of the article, the reasons for the discrepancies were identified and consensus between reviewers was reached. If consensus could not be obtained, a third reviewer independently evaluated the article until agreement was determined. RESULTS: Of the 767 articles that were identified, 12 measures met the inclusion criteria. Most studies were found for the Work Limitations Questionnaire (WLQ). None of the instruments demonstrated satisfactory results for all properties. Five instruments had data on content validity with the methodology being questionable for two of the Instruments. Eleven studies reported on construct validity, however, the methodology was doubtful for three of the measures. Internal consistency was addressed for eight instruments. Test-retest reliability was reported for five instruments, with only one instrument achieving an intra-class correlation coefficient greater than 0.7. Responsiveness was addressed for only two questionnaires. Instruments with an administration time of less than 10 minutes included the Functional Abilities Confidence Scale (FACS) and the Stanford Presenteeism Scale (SPS). Scoring was considered easy for the WLQ and the SPS. Overall, the highest ratings (total scores of greater than 4 out of 7) were given to the FACS, the WLQ-25, and the Occupational Role Questionnaire. CONCLUSIONS: Further studies should be conducted that evaluate the psychometric properties and clinical practicality of outcome measures for assessing work disability. IMPLICATIONS: There is a need to continue to validate work disability outcome measures and to make them more meaningful for physical therapists. KEYWORDS: work disability, outcome measures, systematic review. FUNDING ACKNOWLEDGEMENTS: unfunded. CONTACT: rwilliam@mcmaster.ca