December 2013

Using OSHA inspection data to analyze respirator protection program compliance


We also carried out multivariate analyses of the factors that are associated with a greater likelihood of RP violations a) being cited at an establishment and b) being cited in a subsequent inspection at an establishment that had an RP violation. We used negative binomial regressions because a goodness of fit test rejected the assumptions required for Poisson regression.


The OSHA IMIS reports more than 30,000 violations of the respiratory protection program standard and associated overexposures in manufacturing between 1999 (which was shortly after the standard became effective) and the end of 2006. Table 1 shows the number of establishments in each industry where OSHA inspections cited violations of the RP standard.11 The first column provides the 1997 estimate of the number of establishments in each industry, based on County Business Patterns. (We exclude the tobacco and leather industries because they have so few establishments and, as explained, we exclude establishments and inspections in California, Oregon, and Washington.)

The second and third columns show the number of establishments inspected during this period and the number of them that had at least one health inspection. The fourth column presents the number of establishments where OSHA cited a firm for overexposure to a toxic substance or to the RP standard.12 Column 5 presents the number of establishments with one or more RP violations where the establishment was cited for those violations in a health inspection.

Columns 7 through 9 in table 1 provide a basis for estimates of the prevalence of violations of the RP standard using the three measures described in the methods section. For example, the 3.9 percent for the food industry in column 7 would be a valid estimate assuming that the establishments actually cited for noncompliance were the only ones that were noncompliant. The 10-percent figure in column 8 (measure 2) would be a valid measure if inspected and uninspected manufacturing establishments in each industry were equally likely to be cited for RP violations. The 18-percent figure in column 9 (measure 3) is, as we indicated, the one that we think comes closest to measuring the use (and misuse) of respirators in places where they are required. That 18-percent projection would indicate, for example, that about 3,300 food industry establishments are estimated to have RP violations.13

Finally, column 10 shows the RUPS estimate of the percentage of establishments in each industry where respirator use was required during the last 12 months. (The exact period would depend upon when in 1991 the survey was given to the establishment.)14

The disparities between column 9 and column 10 appear to provide insights into the extent to which respiratory programs in different industries are well implemented. Most strikingly, RUPS indicates that 48 percent of chemical industry establishments require respirator use but only 27 percent of health inspections in the industry cited RP violations. Another industry where respirator violations appear to be less frequent than respirator use might suggest is transportation equipment (SIC 37), where RUPS shows required respirator use in 41 percent of establishments but inspectors cited violations in only 29 percent of health inspections.


11 We linked all inspections that occurred at the same establishment during this period in order to provide measures of establishment coverage. The 1997 County Business Patterns was the last to use SIC codes (not NAICS codes) and thus facilitates analysis for OSHA, which continued to use SIC codes through this period.

12 The general standard for overexposures (1910.1000) includes section E, which requires employers to first attempt to use engineering controls or administrative controls before resorting to the use of respirators to reduce exposures.

13 Here and elsewhere in the report, the figures for the percentage of establishments with health inspections that had violations refers only to violations cited in the health inspections.

14 T figure is drawn from table 6 in the RUPS report.

prev page4next page

View full article
About the Author

John Mendeloff

John Mendeloff is a professor at the University of Pittsburgh and the director of the RAND Corporation Center for Health and Safety in the Workplace.

Maryann D’Alessandro

Maryann D'Alessandro is the director of the National Personal Protective Technology Laboratory, Centers for Disease Control and Prevention.

Hangsheng Liu

Hangsheng Liu is a policy researcher at the RAND Corporation.

Elizabeth Steiner

Elizabeth Steiner is a research assistant at the RAND Corporation.

Jessica Kopsic

Jessica Kopsic is a technical research analyst at MDRC.

Rachel Burns

Rachel M. Burns is a project associate at the RAND Corporation.