ObjectivesThe aim of the present study was to examine the relationship between knee-straining work and radiological findings of knee osteoarthritis and to study the relation between radiological findings and self-reported knee complaints and clinical signs of knee osteoiuthitis. Methods The material consisted of 133 floorlayers, 506 carpenters, and 327 compositors, 26-72 years of age, who had completed a questionnaire and reported no previous knee trauma. A stratified sample of these questionnaire respondents, 50 floorlayers, 51 carpenters, and 49 compositors, were radiologically examined for knee osteoarthritis by 2 radiologists. The X-ray films were independently assessed by 2 radiologists and blinded with respect to knee complaints, trade, and age. R~s u~~s The radiological investigation showed estimated prevalences of knee-osteoarthritis (grades 2 4 ) for 14% of the floorlayers, 8% of the carpenters, and 6% of the compositors (not significantly different). For the subjects 250 years of age the estimated prevalences of the combination of radiological grades 2 4 for knee osteoarthritis and knee complaints during the last 12 months were 29% [95% confidence interval (95% CI) 17-44%], 9% (95% CI 1-26%), and 1% (95% CI 1-10%) for the floorlayers, carpenters, and compositors, respectively. Radiological grades 2 4 were mainly found for subjects above the age of 50 years, subjects with knee complaints, and floorlayers. Radiological knee osteoiuthritis was positively associated with self-reported knee complaints and with clinical signs of intraarticular and retropatellar crepitation. C O~C~U S~O~S The data suggest that work in which a considerable amount of time is spent in knee-straining positions may be a risk factor for the development of knee osteoarthritis above the age of 50 years.