The impact of OSHA inspections on manufacturing injuries.
NIOSH 2003 Jan; :1--98
This study examines the injury prevention impact of OSHA inspections at manufacturing plants from 1992 to 1998. The data come from matching establishments in the OSHA IMIS and the BLS Survey of Occupational Injuries and Illnesses. We matched 16,036 establishments and obtained a sample of 50,076 observations of annual changes in injuries. The study carries out 3 main sets of analyses. The first compares outcomes in 1992-98 with those from very similar studies that examined 2 earlier periods: 1979-85 and 1987-91. The second examines how the characteristics of workplaces and inspections affected the impact of inspections in the 1992-98 period. The third disaggregates both interventions and outcomes to examine how both inspections and the citation of particular violations affected particular injury event types. The comparison across periods found that the sizable effects of inspections found in 1979-85 had declined steadily over time. Depending on what model was used, the reduction in lost workday (L WD) injuries from an inspection with a penalty was either 15% in the first period, 8% in the second period, and 1 % in the most recent period; or 22% in the first period, 15% in the second period, and 8% in the most recent period. These reductions were seen across all size classes of establishments and across all inspection types. We also looked separately at the two components of the L WD rate: injuries with days away from work (DA W) and injuries with restricted work activity only (RW A). In all 3 periods we found that inspections had had no impact on RW A injuries and thus that all of the decline in effectiveness had occurred among DA W injuries. We examined several candidates for explaining the declining impact, but were not able to arrive at a satisfactory explanation. Our study of the 1992-98 period showed that the only clear evidence of effectiveness occurred among establishments with fewer than 100 workers. Inspections at unionized workplaces were found to lead to increases in the numbers of RWA injuries, probably because of changes in management practices that the inspections induce. Surprisingly, health inspections led to sizable reductions in injuries while safety inspections did not. Complaint inspections showed more evidence of impact than programmed inspections, but we found no robust evidence that inspections in particular industries were particularly effective. The third set first categorized DAW injury events into those that were more or less preventable through enforcement of OSHA standards. Contrary to expectations, we found that injury types not addressed by standards were reduced by inspections more than those that were; for example, injuries due to "bodily reaction and exertion" were affected by inspections, but injuries due to getting caught in machinery were not. We also looked at the effects of citing 5 frequently violated standards. Among them, only citations of the general personal protective equipment standard led regularly to injury reductions. We concluded that OSHA inspections have effects not only through a detection/correction mechanism, but also through inducing employers to pay more attention to safety problems as a whole.
Workplace-studies; Injuries; Health-protection; Health-hazards; Construction-industry; Construction-workers; Injury-prevention; Accident-prevention; Statistical-analysis; Behavior-patterns
John Mendeloff, Graduate School of Public Health and Graduate School of Public and International Affairs, University of Pittsburgh, 3E34 Posvar Hall, Pittsburgh, PA 15260
Final Grant Report
NTIS Accession No.
Research Tools and Approaches; Intervention Effectiveness Research
National Institute for Occupational Safety and Health
University of Pittsburgh, Pittsburgh, Pennsylvania