Introduction It was postulated that workers, at the sub-acute stage after injury, respond differently to clinical and occupational interventions offered in a workers’ compensation environment. Individual worker risk of disability, it was further believed, would influence the effectiveness of early intervention. The objective of the current pilot study was to evaluate return to work (RTW) outcomes following proactive, combined clinical, occupational and case management-based interdisciplinary early intervention, provided in a workers’ compensation environment 4–10 weeks of onset of back pain, to workers with medium and high risk for disability. Methods The project was a controlled study comparing conventional workers’ compensation case management with integrated, interdisciplinary and multimodal early intervention (hereinafter referred to as “EI”). At baseline, risk status was determined by a validated Risk for Disability Questionnaire by Carragee et al. (Spine 5(1):24–35, 2005). Seventeen workers at high risk of protracted disability and 20 workers at moderate risk of disability received conventional case management, and 17 workers assessed at high risk of protracted disability and 18 workers at moderate risk of disability received the Early Intervention. Results At 3 months post back pain onset, no statistically significant differences were identified in RTW outcomes between conventional case management and the Early Intervention. However, by 6 months post back pain onset, workers at high risk of work disability who received the Early Intervention were significantly more likely to RTW than high risk workers who received conventional case management. In contrast, moderate risk workers continued to exhibit no statistically significant differences in RTW outcomes. Conclusion Multimodal Early Intervention in the workers’ compensation case management context is likely effective for workers with sub-acute back pain who are at high risk of occupational disability. The comprehensive Early Intervention is, however, likely redundant for workers who are not at high risk for disability and should not be applied indiscriminately. Further studies are required to determine longer-term Early Intervention outcomes, and to replicate the findings using a randomized control design. Also, with a larger sample size, it will be possible to determine predictors of occupational outcomes.