GAO discussed the Occupational Safety and Health Administration's (OSHA) oversight of federal executive agencies' comprehensive safety and health programs. GAO noted that: (1) OSHA did not annually conduct the required comprehensive evaluations of programs at the agencies identified as having the most hazardous conditions and rarely evaluated the effectiveness of programs at other agencies; (2) of 15 agencies targeted for annual evaluations, OSHA only evaluated 13 of them between fiscal year (FY) 1982 and FY 1990 and only one of those agencies more than twice; (3) agencies' annual reports to OSHA differed greatly in the extent to which they allowed assessment of the effectiveness of their programs and OSHA did not use compliance data to assess agency program effectiveness; (4) the OSHA approach for federal agencies was similar to its private-sector approach, in that it emphasized compliance inspections at specific work sites thought likely to be hazardous; and (5) in FY 1990, OSHA inspected more than half of the federal work sites identified as high-hazard for either safety or health problems, but only 10 percent of the private-sector work sites for safety inspections and only 3 percent for health inspections. GAO also noted that OSHA federal agency and private-sector inspection procedures differed in that federal inspections: (1) were targeted using individual work-site data about injuries and illnesses; (2) did not allow civil penalties for violations; and (3) had no independent agency for appeal if an agency disagreed with the inspection findings.