EpidemiologicaI-Environmental Study of Diesel Bus Garage Workers: Acute Effects of NO2 and Respirable Particulate on the Respiratory System

ABSTRACT

Personal samples of nitrogen dioxide (NO2) and respirable particulate (RP) were collected over the shift on 232 workers in four diesel bus garages. Response was assessed by an acute respiratory questionnaire and before and after shift spirometry. Measures of exposure to NO2 and RP were associated with work-related symptoms of cough; itching, burning, or watering eyes; difficult or labored breathing; chest tightness; and wheeze. The prevalence of burning eyes, headaches, difficult or labored breathing, nausea, and wheeze experienced at work were higher in the diesel bus garage workers than in a comparison population of battery workers, while the prevalence of headaches was reduced. Mean reductions in forced vital capacity (FVC), forced expiratory volume in 1 sec (FEVt), peak flow, and flows at 50 and 75% of FVC were not obviously different from zero. There was no detectable association of exposure to NO2 or respirable particulate and acute reductions in pulmonary function. Workers who often had respiratory work-related symptoms generally had a slightly greater mean acute reduction in FEV1 and FEF50 than did those who did not have these symptoms, but these differences were not statistically significant.

METHODS

Measurements of respirable particulate and NO2 were made on each worker and collected on the same day as the pulmonary function tests and questionnaires. NO2 was measured using a passive sampler and respirable particulate was determined gravimetrically. These time-weighted average exposure measures are used to estimate dose-response relations, and are also used as surrogate measures of the total diesel exhaust exposure.

Area samples for the irritants SO2, formic acid, formaldehyde, sulfate, phenol, and acrolein were well below existing federal standards, even when we use the highest value obtained in the additive formula recommended by ACGIH for irritants as a group. We have not analyzed for dose-response relations for these compounds due to the low levels and unavailability of personal environmental samples for them. Thus, in the context of this paper, references to exposure to diesel exhaust indicate exposure to NO2 and respirable particulate.

Each worker was administered a respiratory questionnaire which included questions on symptoms occurring at work and thought by the worker to be work-related. Associations between diesel exhaust exposure and work-related symptoms were analyzed using a logistic model. Age and exposure were retained as continuous variables; garage and smoking category were the other independent variables. Comparisons with a population of battery workers were made without adjustments for age and smoking. These comparison workers were administered a similar questionnaire and did not show a statistically significant association of acute symptoms with acid exposure.

For each worker a minimum of five forced expirations were recorded on magnetic tape using an Ohio 800 rolling seal spirometer I before beginning work and then again at the end of the shift (6-7 hr). Maximum forced expiratory volume in 1 sec (FEV1), forced vital capacity (FVC), peak flow rate (PFR), and flows at 50 and 75% of expired FVC (FEF%0, FEF75) were obtained by lining up the after-shift maximum envelope at total lung capacity of the before-shift maximum envelope and measuring flows at before-shift percentages of FVC. Changes in pulmonary function (APFT) were calculated as (after-shift pulmonary function) minus (before-shift pulmonary function). The effects of diesel exhaust on pulmonary function were estimated in two ways. First, it was determined whether mean changes were significantly different from zero, assuming there should be no change in the absence of an environmental exposure. Second, dose-response relationships were analyzed by multiple linear regression techniques.

Characteristics of study population. 

Table 1 summarizes demographic characteristics of the study population of 115 white males, 96 black males, 8 white females, and 8 black females. Age was bimodally distributed, with 59% of the workers between the ages of 18 and 38 and 27% between the ages of 46 and 59.

The distribution by years worked was highly skewed with a mean of 9.5 years and median of 5.2 years. Workers on the afternoon and night shifts were generally younger and had worked about 6 years less than the day shift workers, and measures of respirable particulate (RP) and NO2 on the night shift were about one half those of the day and evening shifts, respectively. Before-shift baseline pulmonary functions were approximately normally distributed with the following mean values (with SD): FVC = 4.64 liters (1.06); FEV a = 3.61 liters (0.95); PFR = 9.02 liters/sec (2.37); FEFs0 = 4.57 liters/sec (1.81); and FEF75 = 1.54 liters/sec (0.81).

RESULTS

Symptoms

The prevalence of acute symptoms ranged from 15 to 54%, but were elevated above expected only in the high-exposure group and in garages A and B. The strongest associations of exposure and prevalences were with the symptoms of eye irritation, labored breathing, chest tightness, and wheeze. Age was commonly associated with a symptom but, except for wheeze, the younger the age, the higher the prevalence of the symptom. Smoking was associated only with cough and wheeze, and was elevated above expected among only the current smokers (Table 2).
The prevalence of acute symptoms in the study population was also compared with the prevalence among workers in a lead acid battery plant. The two populations were similar in age (38.7 with an SD of 12.9 compared with 36.1 with an SD of 12.6) and in smoking composition (7% more smokers, 1% more ex-smokers, and 8% fewer nonsmokers in the battery population).

The majority of workers in the garage and battery populations thought their symptoms were work-related. Except for chest tightness and eye irritation, the proportion often reporting the symptoms was similar in the two populations. The prevalences of eye symptoms, headaches, difficult or labored breathing, nausea, and wheeze were higher among the garage workers, while the prevalence of sneeze was higher among the battery workers (Table 3).

The acute symptom questions were asked such that past exposures or jobs could affect the response; acute symptoms and exposure on the day of the study could be unrelated, or only indirectly related to exposure levels that day. In garage D, workers were also asked at the end of the shift whether they experienced certain symptoms that day. The agreement between symptoms that day, and past work-related symptoms occurring on that job, was as follows: Cough = 73% agreement; eyes itch, burn, or water = 62%; headache = 71%; wheeze = 97%; and chest tightness = 91%.

Pulmonary Function

Changes in pulmonary function over the shift were normally distributed around zero, and no pulmonary function parameter had a mean value that was significantly different from zero (Table 4).
Even the most complex multiple regression model explained less than 10% of the variation in PFT. The only consistently significant explanatory variable was work shift. The least-square means for PFT during the day shift were consistently positive, whereas the averages of the afternoon and night shifts were negative.


Symptoms and Pulmonary Function

The mean decrement in PFT for all workers with acute symptoms was usually greater than the decrement in those without symptoms. However, the variability was quite large and the differences were generally small (Table 5).


