a choice as to how they wish to be regulated—through partnerships with OSHA or through traditional enforcement (OSHA, 1997c). This policy expands on one actually begun in the 1980s, the Voluntary Protection Program, an incentive program that recognized and promoted employers who established successful safety and health programs in partnership with OSHA. Employers who created effective programs became exempt from routine inspections, as long as they met OSHA criteria for quality. OSHA has built on this concept by now offering regulatory relief and penalty reductions to firms that join this program. The additional benefits, according to OSHA, are enhanced worker productivity and motivation, reduced workers' compensation costs, and community-wide recognition.
In addition, OSHA has begun to nationalize a program inaugurated with the State of Maine in 1993. This program—entitled the Maine 200 program, for the 200 firms that first participated—focuses on firms with the highest numbers of injuries. It gives them the choice of forming a partnership with OSHA to forge an effective health and safety program or of encountering enhanced enforcement. An evaluation of the Maine 200 program found that injury or illness rates of participating companies declined from 1991 to 1996 by 30 percent (OSHA, 1997b).
For injury surveillance, OSHA has, until recently, relied on another agency of the Department of Labor, the Bureau of Labor Statistics (BLS). BLS collects and analyzes annual data on the number of workplace fatalities, injuries, and illnesses through two surveys: the Census of Fatal Occupational Injuries and the Survey of Occupational Injuries and Illnesses (see Chapter 3). The BLS census is considered the most reliable because it uses diverse sources to identify and verify fatalities (Leigh et al., 1997); the survey compiles questionnaire data from approximately 250,000 private firms. These firms are a sample of more than 5 million establishments, which are required under the Occupational Safety and Health Act to maintain records of injuries and illnesses. In 1995, for example, employers reported 500,000 injuries and illnesses that resulted in 21 or more days away from work (BLS, 1997). Sprains, cuts, and fractures account for most of the injuries reported by employers. However, the BLS survey is widely considered to underreport injuries and illnesses by as much as 70 percent because of economic incentives on the part of employers (NRC, 1987; Leigh et al., 1997). Furthermore, the survey does not cover the self-employed, government workers, and farms with fewer than 11 employees. Since 1992, the survey also has collected demographic and detailed case characteristics on a sample of injuries and illnesses that resulted in days away from work (BLS, 1997).
OSHA has recently acquired surveillance capacity of its own with the publication of a final rule that requires employers to notify OSHA of their reportable