OBJECTIVES: Health impairment often leads to work impairment in the form of both absenteeism and presenteeism, which is reduced productivity at work. To date, few instruments are available to measure this impact. The objective of this study was to review and evaluate instruments designed to measure health-related productivity at work. METHODS: A literature search was conducted using ABI Info, EconLit, PsychInfo, MEDLINE, CANCERLIT, AIDSLINE, HealthStar; databases were searched from 1990-2001. Articles identified comprised productivity measured by subjective (self-report questionnaires) and objective (computer-based tracking systems) techniques; subjective instruments included both generic and disease-specific measures. Identified instruments were evaluated on psychometric properties, generalizability to different populations, and administrator/respondent burden. RESULTS: Five generic subjective measures were identified – the Endicott Work Productivity Scale; Health and Labor Questionnaire; Health and Work Questionnaire; Work Limitations Questionnaire; and the Work Productivity and Activity Impairment Questionnaire (WPAI) – that could theoretically be used in any working population. All subjective instruments showed adequate to good reliability. However, these instruments were usually validated against other subjective measures (such as health-related quality of life), as a uniform measure of “productivity” does not exist. The WPAI was the most frequently used instrument and has also been modified to measure productivity reductions associated with specific diseases (e.g., allergic rhinitis, gastro-esophageal reflux disease). Objective assessments have been carried out to measure employee productivity for tasks where individual output could be quantified (e.g., claims processed, telephone inquiries handled). To avoid respondent bias, these studies utilized computer-based tracking systems to determine productivity reduction attributable to illness. CONCLUSIONS: Subjective and objective productivity assessments offer different benefits: while subjective instruments can be used in populations with different illnesses and/or jobs, objective measures minimize respondent bias but are associated with restricted generalizability and are burdensome to carry out in practice.