摘要：Objectives . This study assessed the effectiveness of an intervention to reduce wood dust, a carcinogen, by approximately 26% in small woodworking businesses. Methods . We randomized 48 businesses to an intervention (written recommendations, technical assistance, and worker training) or comparison (written recommendations alone) condition. Changes from baseline in dust concentration, dust control methods, and worker behavior were compared between the groups 1 year later. Results . At follow-up, workers in intervention relative to comparison businesses reported greater awareness, increases in stage of readiness, and behavioral changes consistent with dust control. The median dust concentration change in the intervention group from baseline to follow-up was 10.4% (95% confidence interval = –28.8%, 12.7%) lower than the change in comparison businesses. Conclusions . We attribute the smaller-than-expected reduction in wood dust to the challenge of conducting rigorous intervention effectiveness research in occupational settings. Exposure to wood dust has been associated with several cancers, including those of the nasal cavity, lung, and gastrointestinal tract and Hodgkin disease. 1, 2 Of these, the epidemiological evidence is strongest for cancers of the nasal cavity, 3, 4 and in 1994, the International Agency for Research on Cancer declared wood dust a known carcinogen. 2 Other adverse health effects of wood dust include allergic respiratory diseases, bronchitis, and dermatitis. 5 According to the US Census Bureau, approximately 1.75 million people are employed in wood products, paper, furniture, and related product manufacturing. 6 The state of intervention effectiveness research for occupational health and safety has been detailed in several reviews of studies conducted primarily from the 1980s through the mid-1990s. 7– 10 In general, these reviews concluded that occupational health and safety studies were more likely to focus on worker knowledge and behavior than on engineering or administrative improvements in the workplace, were usually restricted to a limited number of workers within a single worksite, were quasi-experimental or nonexperimental without a comparison group, lacked a theoretical framework, and typically based outcomes on worker self-report rather than on more objective outcomes such as a reduction in injuries or hazardous exposures. In addition, most health and safety research has targeted large industries with many employees, even though small businesses (i.e., those with fewer than 100 employees) constitute 98% of all businesses and employ approximately 57% of the US workforce. 11 The Minnesota Wood Dust Study was designed to address both an important health hazard—wood dust—and many of the limitations of previous research conducted in the occupational setting. To assess the effectiveness of an intervention aimed at reducing wood dust by approximately 26%, a reduction deemed feasible on the basis of published literature and our pilot study results, 12, 13 we randomly assigned 48 woodworking businesses, each of which employed 5 to 25 woodworkers, to intervention or comparison conditions. Drawing from the health promotion field, in which the use of interventions tailored to meet an individual’s specific barriers to change has been effective in changing health-related behaviors such as smoking, 14 diet, 15 and breast cancer screening, 16– 19 the intervention consisted of general written recommendations, technical assistance to enhance engineering, administrative methods to control wood dust, and worker training to modify work practices associated with high dust production, based on an evaluation of the specific needs of each business. Businesses in the comparison condition received written recommendations alone. Changes from baseline to 1-year follow-up in dust concentration and worker behavior were compared between woodworking businesses assigned to the intervention and those in the comparison condition.
