Summary Problem: Generalized anxiety disorder (GAD) is more prevalent in people with low back pain than in the general population. When GAD is present, intolerance of uncertainty is also higher and anxiety is maintained through ineffective strategies. In the long term, reduced self-efficacy in problem solving and the onset of depressive symptoms may hinder the return to work. The general aim of this exploratory study was to enhance understanding of the nature of the anxiety symptoms found in workers with a persistent musculoskeletal disorder (MSD). The primary objective was therefore to document the presence, intensity, and temporal changes in GAD and its maintenance factors, using the model developed by Dugas et al. (1998). These factors include intolerance of uncertainty, worries, negative problem orientation, beliefs about the usefulness of worrying, cognitive avoidance, and depression. The secondary objectives were as follows: (1) to document, on an exploratory basis, the relationships between the component factors of the GAD model developed by Dugas et al. (1998) and the biopsychosocial factors already recognized in the MSD field, and (2) to document the relationship between all these factors and the return to work. Method: A prospective, repeated-measures observational design was employed and a convenience sample of 39 workers was recruited. The inclusion criteria were (1) having an MSD that was accepted and compensated by Québec’s worker compensation board, the CSST (Commission de la santé et de la sécurité du travail) and that had caused a work absence of more than three months; (2) being between 18 and 64 years of age; (3) speaking French, and (4) having a work exposure component in their treatment plan. The exclusion criteria were (1) having an MSD related to a specific pathology and (2) the presence of a severe mental disorder identified in the medical file. The participating workers were evaluated (1) at the start of their rehabilitation program (2) during the initial hours of their return to work, (3) when they had resumed 50% of their full work hours, and (4) at the end of their rehabilitation program, by means of validated self-report questionnaires measuring the factors in the GAD model, biopsychosocial factors recognized in the MSD field, and medico-administrative factors. Results: A total of 50% of the participants presented with symptoms of GAD. However, with respect to the intensity of the symptoms pertaining to GAD development and maintenance factors, the profile obtained was not typical of GAD. Intolerance of uncertainty, worries, negative problem orientation, beliefs about the usefulness of worrying, cognitive avoidance, and depression were significantly reduced during the rehabilitation program. At the end of rehabilitation, only 21% of the participants still met GAD diagnostic criteria. The biopsychosocial factors already recognized in the MSD field did not appear to correlate significantly with the component factors of the GAD model. However, the perception of benefitting from a safe work environment correlated with a lower risk of presenting GAD symptoms. Regarding the return to work, the related factors were beliefs about the usefulness of worrying, kinesiophobia, pain catastrophizing, and the perception of benefitting from ergonomics and disability management. Conclusion and contributions of the study: The comparison of our results with empirical or normative data also allowed for the interpretation and a better understanding of the magnitude of the difficulties experienced by workers with a work disability.