Objective: To evaluate the predictive value of the Work Ability Index (WAI) for different indicators of the need for rehabilitation at 1-year follow-up. Design: Cohort study. Methods: Data were obtained from the Second German Sociomedical Panel of Employees, a large-scale cohort study with postal surveys in 2009 and 2010. Results: A total of 457 women and 579 men were included. Confirmatory factor analysis confirmed the one-dimensionality of the WAI. Regression analyses showed that poor and moderate baseline WAI scores were associated with lower health-related quality of life and more frequent use of primary healthcare 1 year later. Subjects with poor baseline work ability had 4.6 times higher odds of unemployment and 12.2 times higher odds of prolonged sick leave than the reference group with good or excellent baseline work ability. Moreover, the odds of subjectively perceived need for rehabilitation, intention to request rehabilitation and actual use of rehabilitation services were 9.7, 5.7 and 3 times higher in the poor baseline WAI group and 5.5, 4 and 1.8 times higher in the moderate baseline WAI group, respectively. A WAI score ≥37 was identified as the optimal cut-off to predict the need for rehabilitation. Conclusion: The WAI is a valid screening tool for identifying the need for rehabilitation.