The goal of the study was to assess the test-retest reliability of the Work Ability Index (WAI) questionnaire. Reliability was tested using a test-retest design with a 4 week interval between measurements. Valid data were collected among 97 elderly construction workers aged 40 years and older. We analysed the test-retest reliability of the WAI score itself (range 7-49 points) and classification in one of the four WAI categories based on this score: poor (7-27 points); moderate (28-36 points); good (37-43 points); and excellent work ability (44-49 points). Exactly the same WAI score on both measurements was reported by 25% of the subjects and 95% of the individual differences between measurements were found to be <6.86 points (two times standard deviation). Despite the individual changes between measurements, no significant difference was reported in the mean WAI score at group level between test and retest measurements (40.4 versus 39.9). The percentage of observed agreement for the classification of subjects in one of the four WAI categories on both measurements equalled 66%. The results of this study provided evidence of an acceptable test-retest reliability of the classification of subject's work ability by means of the WAI questionnaire. At group level, the mean WAI score and classification into WAI categories were found to be a stable measure over a 4 week interval. These results give additional support for the applicability of the questionnaire in occupational health research and the daily practice of occupational health care.