• Maria B Ospina, Liz Dennett, Arianna Waye, Phillip Jacobs, and Angus H Thompson.
• Institute of Health Economics, Ste 1200, 10405 Jasper Ave NW, Edmonton, AB, Canada T5J 3N4. E-mail: gthompson@ihe.ca.
• Am J Manag Care. 2015 Feb 1; 21 (2): e171-85.

BackgroundPresenteeism (decreased productivity while at work) is reported to be a major occupational problem in many countries. Challenges exist for identifying the optimal approach to measure presenteeism. Evidence of the relative value of presenteeism instruments to support their use in primary studies is needed.ObjectivesTo assess and compare the measurement properties (ie, validity, reliability, responsiveness) and the quality of the evidence of presenteeism instruments.Study DesignSystematic review.MethodsComprehensive searches of electronic databases were conducted up to October 2012. Twenty-three presenteeism instruments were examined. Methodological quality was appraised with the COSMIN (COnsensus-based Standards for the selection of health status Measurement INstruments) checklist. A best-evidence synthesis approach was used in the analysis.ResultsThe titles and abstracts of 1767 articles were screened, with 289 full-text articles reviewed for eligibility. Of these, 40 studies assessing the measurement properties of presenteeism instruments were identified. The 3 presenteeism instruments with the strongest level of evidence on more than 1 measurement property were the Stanford Presenteeism Scale, 6-item version (content validity, internal consistency, construct validity, convergent validity, and responsiveness); the Endicott Work Productivity Scale (internal consistency, convergent validity, and responsiveness); and the Health and Work Questionnaire (HWQ; internal consistency and structural validity). Only the HWQ was assessed for criterion validity, with unknown quality of the evidence.ConclusionsMost presenteeism instruments have been examined for some form of validity; evidence for criterion validity is virtually absent. The selection of instruments for use in primary studies depends on weak forms of validity. Further research should focus on the goal of a comprehensive evaluation of the psychometric properties of existing tests of presenteeism, with emphasis on criterion validity.

18 keywords...

hide…