Psychometric properties of presenteeism scales for musculoskeletal disorders: a systematic review
Jean-Sébastien Roy1, François Desmeules, Joy C MacDermid
1Department of Rehabilitation, Faculty of Medicine, Laval University, Quebec City, Quebec, Canada. jean-sebastien.roy@rea.ulaval.ca
Objective:
To conduct a systematic review of the psychometric evidence relating to presenteeism scales in workers with musculoskeletal disorders.
Methods:
A structured search was conducted in 3 databases (MEDLINE, CINAHL, Embase) for articles published between 1966 and 2010. Sixteen articles met eligibility criteria. Pairs of raters used structured tools to analyse these articles through critical appraisal and data extraction. Descriptive synthesis of the psychometric evidence was then performed.
Results:
Methodological quality ratings of 56% of the studies reviewed reached a level of 75% or higher. Seven presenteeism scales were evaluated. Overall, presenteeism scales demonstrated acceptable validity content, were moderately to highly correlated (r > 0.50) to each other and to work- and disease-oriented constructs, and were able to differentiate between different populations and disability levels (p < 0.05). Limited evidence exists on the reliability and responsiveness of presenteeism scales, as reliability had only been evaluated for two scales and responsiveness in two studies.
Conclusion:
None of the identified scales demonstrated satisfactory results for all evaluated psychometric properties. For most scales, data regarding properties such as reliability and responsiveness were insufficient. Therefore, there is no substantial evidence to recommend one questionnaire over the others based solely on psychometric properties.
