In 1991, Congress enacted the Intermodal Surface Transportation Efficiency Act (ISTEA), which required each state to use a minimum quantity of "crumb-rubber modified" (CRM) hot-mix asphalt (HMA) paving material. Because of industry and labor concerns over the lack of available information on the environmental and human health effects resulting from the use of CRM-HMA, along with the higher initial cost of using this paving material, a temporary legislative moratorium was passed and the U.S. Environmental Protection Agency (EPA) and the U.S. Department of Transportation, Federal Highway Administration (FHWA) were directed by Congress to evaluate the potential environmental and human health effects associated with the use of CRM asphalt. The National Highway System Designation Act of 1995 eliminated the mandate requiring the use of CRM asphalt but continued to require research concerning CRM asphalt paving. In June 1994, the National Institute for Occupational Safety and Health (NIOSH) entered into an Interagency Agreement with the FHWA to evaluate occupational exposures among asphalt road workers. A study protocol developed by NIOSH included the following objectives: Develop and field test new methods to assess asphalt fume exposures; Characterize and compare occupational exposures to CRM asphalt and conventional (CONV) asphalt; Evaluate potential health effects associated with CRM asphalt and CONV asphalt. The protocol called for up to eight individual site evaluations in different regions of the country. The intent was to allow NIOSH investigators to observe different asphalt pavement formulations, climatic conditions, and paving techniques. Seven site evaluations were completed between 1994 and 1997. The environmental and medical results discussed collectively in this report have been individually published in the following seven NIOSH Health Hazard Evaluation (HHE) final reports: HETA 94-0365-2563, Spartan Paving Company, Lansing, Michigan (March 1996); HETA 94-0408-2564, Granite Construction Company, Sacramento, California (March 1996); HETA 95-0118-2565, Martin Paving Company, Yeehaw Junction, Florida (March 1996); HETA 95-0307-2602, Koester Equipment Company, Evansville, Indiana (December 1996); HETA 96-0072-2603, Staker Paving Company, Casa Grande, Arizona (December 1996); HETA 96-0130-2619, Sim J. Harris Company, San Diego, California (December 1996); HETA 97-0232-2674, Barton-Trimount, Stoughton, Massachusetts (February 1998). A new NIOSH method which simultaneously sampled for total particulate (TP) and benzene soluble particulate (BSP) was developed. Polycyclic aromatic compounds (PACs) were sampled using a new analytical method that included a PAC370 group (2-3 ring compounds, many of which have irritative effects) and a PAC400 group (4- and more ring compounds, some of which are carcinogenic). In addition to PACs, organic sulfur-containing compounds (OSCs, present in crude petroleum or from the addition of rubber) and benzothiazole (a sulfur-containing compound present in rubber tires), were also sampled using a newly developed sampling and analytical method. These compounds were of interest for their potential for respiratory irritation (OSCs) or for their use as an indicator of other chemicals present in CRM asphalt fume (OSCs and benzothiazole). Samples were collected for volatile organic compounds (VOCs, including toluene, xylene, benzene, and methyl isobutyl ketone [MIBK], and total VOCs [TVOCs, quantified as Stoddard solvent]. Both elemental carbon (EC) and organic carbon (OC) were measured and the ratio to total carbon (TC) was compared to ascertain if diesel exhaust was a likely contributor to the air contaminants measured at each site. Area air samples were collected to determine the respirable particulate concentrations. Direct-reading instruments were used to measure carbon monoxide (CO), hydrogen sulfide (H2S), sulfur dioxide (SO2), and ozone. Finally, high volume air samples of both CRM and CONV asphalt fume were collected from the emissions of asphalt cement storage tanks located at the hot-mix plants and analyzed to determine their mutagenic potential. Area air sample results revealed that concentrations of TP, respirable particulate, BSP, PACs, OSCs, and benzothiazole varied between sampling locations and survey days but were generally higher during the CRM asphalt paving than during CONV asphalt paving. In all but two samples, the PAC370 concentrations were greater than the PAC400 concentrations. All of the air samples collected for EC above the screed auger on the paver vehicle had concentrations above the background levels. Since diesel exhaust has been reported to contain EC levels between 60 to 80% of the TC, the relatively low EC:TC ratios measured at all but one of the sites imply that diesel exhaust was not substantially contributing to the air sampling results. None of the asphalt fume samples were found to be mutagenic using a spiral Salmonella mutagenicity assay. Over 50 VOCs were detected in the asphalt emissions, but only the highest peaks were analyzed quantitatively. Although higher concentrations of toluene, xylene, and MIBK were measured during CRM asphalt paving, all concentrations were generally less than 1 part per million (ppm). Concentrations of TVOCs (as Stoddard solvent) ranged up to 224 milligrams per cubic meter (mg/m3). The NIOSH Recommended Exposure Limit (REL) for Stoddard solvent is 350 mg/m3 for up to a 10-hour time-weighted average (TWA). Benzene was detected near the screed auger in area samples collected during CRM asphalt paving in concentrations up to 0.77 ppm. Lower, but still detectable, concentrations of benzene were measured during CONV paving. NIOSH classifies benzene as an occupational carcinogen with a REL of 0.1 ppm, 8-hour TWA, noting that exposures should be controlled to the lowest feasible level (LFL). All personal breathing-zone (PBZ) TP exposures were below 1.4 mg/m3, expressed as TWAs for the workday. These PBZ results cannot be compared to the NIOSH REL for asphalt fume of 5 mg/m3 for a 15-minute exposure since the samples in this study were collected over the full work-shift. For six of the eight job categories studied, the geometric mean (GM) PBZ exposures to TP during CRM asphalt paving (range 0.17 to 0.48 mg/m3) was higher than during CONV asphalt paving (range 0.06 to 0.81 mg/m3). However, only the screed operators and roller operators were exposed to significantly more TP during CRM asphalt paving than during CONV asphalt paving (p 0.01). The GM BSP concentrations were higher for four of the six jobs evaluated during CRM asphalt paving (range 0.02 to 0.25 mg/m3) compared to CONV asphalt application (range 0.02 to 0.44 mg/m3). The average concentration by job was below the American Conference of Governmental Industrial Hygienists (ACGIH) Threshold Limit Value (TLV) of 0.5 mg/m3 , 8-hour TWA for asphalt fume. Some paver operators and truck dumpers, however, had individual exposures above the TLV. The BSP concentration differences by asphalt type were not significantly different for any job category. The GM PBZ concentrations for PAC370, PAC400, OSCs, and benzothiazole were higher during CRM asphalt paving than CONV paving. As was observed in the area air samples, PAC370 concentrations exceeded PAC400 concentrations. Of the jobs evaluated, only the screed and roller operators had significantly higher PAC exposures during CRM asphalt paving when compared to CONV paving (p 0.01). The paver, screed, and roller operators were all exposed to significantly more OSC during CRM paving than CONV paving (p 0.01). All paving jobs were exposed to significantly more benzothiazole during CRM paving than during CONV paving (p 0.01). With the exception of the first survey site in California, benzothiazole was only detected during CRM asphalt paving. There are currently no occupational exposure limits for PAC370, PAC400, OSCs, or benzothiazole, and little occupational exposure data exists for comparison beyond what is reported in this study. TWA CO area concentrations ranged up to 24 ppm during paving, with peak values as high as 910 ppm. The NIOSH REL is 35 ppm for up to an 8-hour TWA, with a ceiling limit of 200 ppm. Concentrations of H2S, SO2, and ozone were well below their respective occupational exposure limits. Prior studies of the acute toxic effects of asphalt fume exposures in workers have repeatedly reported eye and upper respiratory tract (nasal) irritation. These health effects have been well documented in asphalt road pavers and typically appear to be mild and transient. In this study, both CRM and CONV asphalt workers experienced symptoms (eye, nose, and throat irritation being most commonly reported), but the occurrence rates of these symptoms were higher among the CRM asphalt-exposed group. With one exception during CRM asphalt paving, specific contaminant concentrations did not differ between days when eye, nose, or throat irritation or cough were present and days when these symptoms were absent. Concentrations of TP were nearly twice as high on days when eye, nose, or throat irritation was present compared to those days when eye, nose, or throat irritation was not reported (p 0.01). During CONV asphalt paving, concentrations of TP, BSP, and PACs were significantly higher on days when eye, nose, or throat irritation was present (p 0.05). Concentrations of BSP, OSC, and benzothiazole were significantly higher on days when cough was present (p 0.05). Although an exposure-response relationship has not been established in this study, the identification of health effects related to personal exposures to contaminants (measured as TP and BSP) during both CRM and CONV asphalt paving indicates that such a relationship may exist. Exposures to TP, BSP, PACs, and OSCs were generally higher during CRM asphalt paving compared to CONV asphalt paving. The highest PBZ exposures were measured on paver and screed operators, and truck dumpers. Screed and roller operators were exposed to significantly more TP during CRM paving than during CONV paving, while no significant difference existed between CRM and CONV paving for BSP exposures. Workers on all paving jobs had significantly higher exposures to benzothiazole during CRM paving. Symptom occurrence rates were higher during CRM asphalt paving, and eye, nose, and throat irritation were the three most frequently reported symptoms among all asphalt fume exposed workers. At some paving sites, worker exposures to CO exceeded the NIOSH ceiling limit of 200 ppm. Area air samples measured generally low concentrations of selected VOCs at the paving sites, with the exception of benzene. Recommendations are included for lowering the asphalt application temperature; prohibiting eating, drinking, and smoking near asphalt fume emissions; providing washing and clothes-changing facilities at the work site; reducing the use of diesel fuel for tool cleaning; and wearing protective clothing or appropriate sun screen to protect exposed skin from harmful effects of sun exposure when working outdoors.