Determinants of long duration sickness absence after an occupational back injury in the Belgian working population

Aims
Occupational low-back injuries are a major cause not only of lost time but also of job loss and early retirement for health reasons. To promote the employability of an aging workforce, a better understanding of the prognostic factors for a long duration sick leave after such injuries is needed. The Belgian national database of work accidents was used to investigate these factors.

Methods
The data were retrieved from the Belgian Fund for Work Accidents (FAT) database over a three-year period (2001-2003). The population source involved all the workers under a job contract in the private sector recorded during that period as compensated cases for an accident that occurred at the workplace (n=558.276); 6,6% of those cases had the back as “location of injury”. Taking into account the calendar year system for counting time on benefits, the analysis was restricted to the 18936 workers who had reported a back injury during the first 6 months of each year. Eight factors (gender, age, region, professional category, size of enterprise, seniority, sector of activity and accident circumstances) were analyzed in relation with the duration of sick leave as attributed by the insurance company. For each factor, the relative prevalence of cases with more than 92 days of work absence (outcome variable) was compared using the Pearson’s 2 in the bivariate analysis. In the multivariate analysis, a Stepwise Logistic Regression was performed to obtain the final model.

Results
The proportion of back injured workers with more than 92 lost days was 6.8 % (CI 95%: 6.4-7.1). In the bivariate analysis, all the 8 factors under study, gender and seniority excepted, were correlated with sick leave duration. After adjustment for confounding factors, the final model involved 5 factors statistically associated to a long duration sick leave: age (older than 40 years : OR=1.45(1.25-1.68)); geographical location of company (Wallonia: OR=1.70(1.47-1.98); Brussels: OR=1.61(1.29-2.02)); professional category (blue collar: OR=1.71 (1.32-2.20)); circumstance of accident (falls : OR=1.26 (1.03-1.52); overexertion: OR=0.75(0.64-0.89)); and sector of activity (building industry: OR=1.47(1.22-1.78)).

Discussion and conclusions
This study showed that in the Belgian working population, 6.8 % of the compensated back injuries can be considered as chronic back pain cases. This figure is however likely to underestimate the actual incidence of the problem as the FAT statistics showed that 21.7% of the back injury claims were rejected by the insurers. The proportion of back injured workers with a long sick leave (chronic low back pain) is clearly related to the worker age, blue-collar status, accident precipitating cause, and some high-risk activities; these observations are in line with the literature. The observed regional differences in incidence could not be explained by a differential distribution of high risk sectors in the country; they are probably to be ascribed to cultural and social factors but these could not be studied on the basis of the variables recorded in the FAT database. In conclusion, this study shows that considerable prevention efforts have to be focused on some categories of workers with back injury, who are at risk for a long sickness absence leading to chronic low back pain.

References
1. Nielens H., Van Zundert J., Mairiaux P., Gailly J., Van Den Hecke N., Mazina D., Camberlin C., Bartholomeeussen S., De Gauquier K., Paulus D., Ramaekers D. Chronic Low Back Pain. Good Clinical Practice (GCP). Bruxelles: Centre fédéral d'expertise des soins de santé (KCE); 2006. KCE reports 48B (D/2006/10.273/64).
2. Crook J, Milner R, Schultz IZ et al. Determinants of occupational disability following a low back injury: a critical review of the literature. [Review] [68 refs]. Journal of Occupational Rehabilitation.12:277-95, 2002.