Objective: To clarify examine the epidemiologic evidence linking work related exposure to osteoarthritis (OA) of the hip and knee.
Methods: We investigated MEDLINE and EMBASE 1966-1994 with search terms of osteoarthritis, osteoarthrosis, arthrosis; risk factors, exposure; occupational diseases, agricultural workers' diseases, work. From 123 original studies on risk factors for OA, 17 studies were identified as providing a comparison group and relating the presence or absence of radiologically diagnosed OA to occupational factors. The quality of the methodology of each study was evaluated independently by 4 reviewers using a standardized protocol.
Results: Common methodological problems were encountered in areas such as representatives, recall of exposure, and evaluation of exposure and outcome. These problems often made a cause-effect relationship difficult to interpret. However, 7 of the 17 studies met our criteria for good methodological quality. A critical analysis of their results led to the following conclusions: (1) A consistently positive relationship exists between work involving knee bending and knee OA in men (range of odds ratio: 1.4-6). (2) The evidence suggesting a relationship between knee OA and occupational exposure in women is inconclusive. (3) A consistently positive but weak relationship exists between work related exposure (i.e., farming in particular) and hip OA in men. We felt we could not conclude with confidence that this relationship is strong due to the potential biases that exist in each of these studies. (4) No study attempted to investigate occupational exposure and hip OA in women.
Conclusion: Studies suggest a strong positive relationship between work related knee bending exposure and knee OA. The evidence between work related exposure, farming in particular, and hip OA is consistently positive but weak.