Two studies are reported that tested the fear-avoidance (FA) model using path analytic techniques. In study 1, 429 employees with back pain at baseline and back pain at 18 months follow-up completed questionnaires assessing sociodemographic information, pain severity, negative affect, pain-related fear, and disability. Results indicated that pain severity at baseline predicted pain-related fear and disability at follow-up, and that pain-related fear is rather a consequence than an antecedent of pain severity. Results further revealed that the disposition to experience negative affect has a low impact upon pain severity and disability, and is best viewed as a precursor of pain-related fear. Study 2 included 238 employees without back pain at baseline, but who developed back pain at 1 year follow-up. A similar model as in study 1 was tested. Overall, results are in line with those of study 1. Results are discussed in terms of theoretical relevance and clinical implications.