The independent effects of physical and psychosocial job factors on occupational LBP and disability are still under debate, and their impact during different phases of the disabling and recovery process are essentially unknown. The goal of this project was to prospectively examine the role of physical and psychosocial job factors in the etiology and prognosis of occupational low back pain (LBP) during five successive phases in the course of occupational LBP: (1) the pre-disability symptom phase, (2) the pre-disability formal injury report phase, and (3) three disability phases - acute, subacute, and chronic - defined by increasing durations of lost work time. It was hypothesized that both physical and psychosocial job factors are independent predictors of LBP at all five phases and that their relative effect sizes change across phases. The main objective was to determine phase-specific risk factor profiles with particular focus on the relative impact of psychosocial and biomechanical risk factors. The long-term goal is to yield useful information for the design of workplace interventions which combine organizational and ergonomic job redesign to prevent low back injuries and associated work disability. This project combines the resources of two existing data sets - two prospective cohorts of San Francisco urban transit operators (n = 1,449, n = 1,640). These two longitudinal studies provide workers' compensation data with up to 7.5 years of follow up, allowing for the study of each phase of the disability process, including the chronic disability phase (>90 days off work), which accounts for 80% of the costs associated with work-related low back injuries. Both studies provide comparable information on job related, sociodemographic, injury, medico-legal, and economic factors. Primary analyses examined the independent effects of physical workload (measured by duration of professional driving, vehicle type, and ergonomic problems) and psychosocial job factors (including self-reported psychological and physical job demands, job control, job strain and social support at work). In addition, unique data on job stressors, assessed by observers with an innovative job analysis instrument independent of worker appraisal, were used as an alternative objective measure of job stress in a sub-study of one-third of transit lines. Secondary analyses aimed to determine the predictive validity of these observational measures for LBP, neck pain, and musculoskeletal disorders of the extremities. Separate analyses were conducted for less and more severe injuries based on medical criteria. Key findings of this project can be summarized as follows: Both physical and psychosocial job factors were independently associated with all outcomes under study. Representative for physical job factors, the amount of professional driving in hours per week showed a strong and statistically significant dose-response relationship with 12-month prevalence of LBP and with incidence of a first compensated work -related low back injury during 7.5 years of follow-up, even after controlling for psychosocial job factors and individual demographic and anthropometric variables. Similarly, cable car operators performing heavy physical labor consistently had a higher risk for LBP and work-related injury compared to diesel bus drivers. Self-reported ergonomic problems were also associated with LBP prevalence and low back injury incidence. Furthermore, long weekly driving hours constituted a major barrier to return-to-work after a disabling low back injury during all disability phases. It is noteworthy that the effects of physical job factors were strongest for severe low back disorders confirmed by medical diagnosis such as postlaminectomy syndrome, spinal stenosis, or herniated lumbar disc with myelopathy. The diagnosis of such disorders is based on objective signs of structural and functional damage to the lower spine and, in contrast to less severe low back injuries, is therefore virtually independent of subjective worker or physician appraisal. These findings in particular represent new strong evidence for a causal role of physical job factors in the development of occupational low back injuries. Among psychosocial job factors, self-reported high psychological job demands and the combination of high psychological demands and low decision latitude at work emerged as risk factors for both low back pain prevalence and injury incidence, independent of physical job demands or individual worker characteristics. Observer-based job stressors showed strong and statistically significant associations with LBP, with odds ratios ranging from 2.76 for vehicle moving barriers to 8.96 for barriers for locomotion and equipment handling. For each increment of 10 minutes of extra work due to all six barrier categories combined, the prevalence of LBP increased by 46%, even after adjustment for age, gender, and physical workload. Other musculoskeletal disorders (with the notable exception of neck pain) were also strongly associated with this observational measure of job stress. The results demonstrate a high predictive validity of this new observer-based measure of job stress and support a causal role of job stress in the development of low back pain and other musculoskeletal disorders independent from the appraisal of the worker and independent from physical workload and demographic factors.
Department of Occupational and Environmental Medicine, San Francisco General Hospital, University of California - San Francisco, San Francisco, California