Association between exposure to a stenching agent in a herbicide packing plant and the occurrence of headache and lethargy.

Association between exposure to a stenching agent in a herbicide packing plant and the occurrence of headache and lethargy. Scand J This study quantifies the relationship between subjective complaints of symptoms and certain work activities and exposures in a plant where herbicide is packed. Data relating to symptoms were collected from 27 subjects (20 men and 7 women) on a daily basis with the use of a questionnaire. In addition, data were collected relating to work activity and exposure to the stenehing agent added to the herbicide, atmos pheric levels of which were measured with personal monitoring, on a daily basis. Although associations were found between the occurrence of lethargy and certain factors of job history, there was no significant association with exposure to the stenching agent as measured by personal monitoring during the study.

The study was undertaken in respon se to complaints of symptoms including headache and lethargy from personnel working in a plant where a herbi cide is formulated and packed . A preliminar y investigation into the repor ted symptoms revealed that the re had been specific cases of the occurr ence of the symptoms of headache, lethargy, and nausea associated with spillages of the fully formulated product. A review of the toxicology, kno wn properties, and likely exposure of all chemicals in use in the plant identified the stenching agent as the onl y cred ible chemical cause of the symptoms reported.
The stenching agent by definit ion and purpose has an unplea sant smell, and the employees who had suffered from headach e, nau sea, and lethargy had often associated these symptoms with the smell. The stenching agent is a mixture of alkyl pyridines, primarily dimethylpyridine with some ethylpyridine and some trimethylpyridine. There are insufficient data to calculate a median lethal concentration (LC so ) for the inhalation of any of the main components. The LC so (for rat) for 2,4-dimethylp yridine lies between 650 and 4300 ppm (2.0 h). The lowest concentration of any of the other main components found to cause 100 % mortality was inhalation of 2500 ppm (2.0 h). Acutely high concentrations of pyridine vapor cause irritation to the eyes and mucus membranes in man and can result in headache, giddiness, and nausea, possibly leading to uncon sciousness (1

440
The main objective of the study was to estimate the risk of symptoms occurring in associat ion with exposure to the stenching agent and to investigate any other possible causes of work -related symptom s.

Materials, subjects and methods
Subjective symptoms of a general nature such as headache and fatigue have been shown to be common among work ing groups and may be related to man y factor s, such as age, gender , physical act ivity, stress, bor edom, type of job, and number of days since last day off (2)(3)(4). In view of this fact, we felt it appropriate to study the occurrence of these symptoms in relation to quantitative measures of exposure to the stenching agent.
The investigation of this type of symptom in this work for ce did not permit the use of a suitable reference group. As an alternative, analysis of the doseresponse relationships was used to identify any possible bias. The study involved 27 subjects (20 men and 7 women) who comprised the total work force of the plant and thu s included the workers who had originally complained of headache or letharg y. Fourteen study days were identified durin g a three-week period. The total number of person-days included in the study was 294, of which 72 were not workd ays.
In addition to personal monitoring for the stenching agent, each subject was also monitored for noise and temperature when practicall y possible. At the end of each stud y day each individual at work had a work activity questionnaire completed for him by his line supervisor. The work activity questionnaire included a record of the job station for that day, the grade of produ ct being handled, and any spillages or unu sual incidents which may have given rise to unusual ex- Table 1. Results of the personal atmospheric mon ito ring for the sten ch ing agent.

Discussion
Th e study was carri ed out as a result of employees' complaints of symptoms related to the smell of a stenchin g agent. Unfortunately it was not possible to conduct the study on a different work group which was exposed to the same substances and which worked under the same conditions. However, the lack of a relationship between the symptoms and exposure to the stench ing agent is a valid and significant finding, as any bias would be expected to operate in a positive direction . As the workers had a preconception about the relationship between the symptoms and the stenching agent , one might expect mor e symptoms to be report ed when th e stenching agent was noti ceable in the plant. Th is was not true for headache, and in fact the reverse was found in the case of lethargy.
It is diffi cult to account for the actu al cau se of the increased estimated risk of letharg y found to be associated with work . Without a comparison gro up one cannot conclude that this increase was cau sed by any facto r specific to the job. It is possible that lethargy could be caused simply by the work itself.
An investigation of other possible causes of wor krelated symptoms unco vered only one significa nt association (work in the bottle filling shed). Th is finding was thought to be arti factual, and indeed more associations might have been expected due to the nature of the possible subjective bias.
In conclusion, there was no evidence that symptoms occur in relation to exposure to the stenching agent in the range of exposures encountered durin g the study period. From past experience it is accepted that overexposure to the stenching agent, as a result of spillages, may give rise to symptoms, but these should not occur if exposure is con trolled with in the range found in this study . posure. Result s from personal stench monitors were entered into the work activity questionnai re. The timeweighted avera ges for the shifts are summarized for the entire study period in table I. App roximately 80 070 of the results were measured at or below 0.1 ppm and over 98 % were at or below I ppm.
On the morning following the study day, the subjects attend ed the medical center and completed a questionn aire regarding the occur rence of sympto ms during the previous 24 h. In each case the questionnaire was adm inistered by the same person (the nur sing officer).
Th e relationship between the repor ted symptoms of letha rgy and headache and exposure to the stenching agent was assessed with logistic regression methods. Associations between symptoms and variables measuring work activity were assessed with a subjectadjusted Pearson chi-square test, the Cochran-Mantel-Haen szel sta tistic. Seven features of the job histories were examined for an association with symptoms of lethargy or headache. These comprised noise, temperature, exposure to the stenching agent , working with two grades of product (6917C and 7452), and working in two areas (bottle filling shed and the old drum shed). All statistical anal yses were performed with the SAS (statistical analysis system) package (5).

Results
The logistic regression anal yses of the relationships between the symptoms and exposur e to the stenching agent showed no significant associat ions. Indeed the relationship between lethargy and exposure almost achieved statistical significance in a negative direction. Th is result was confirmed when the exposure levels were split into detectable and less than detectable levels and analyzed with the Cochran-Mantel-Haenszel statistic. Personal monitoring devices could det ect stenching agent down to 0.1 ppm.
Nevertheless there were significant differen ces in the symptom of lethargy between day s spent at work and off work . (Workers were nine times mor e likely to report symptoms on workdays.) There was a suggestion of a difference in the symptom of headache between days spent at work and off wor k, but the assoc iatio n was not significant (P = 0.07).
For the female workers none of the seven factors of job histor y showed a significant associa tion with the symptom of lethargy. In the male gro up there was a significant association between the symptom of lethargy and work in the bottle filling shed (P < 0.05).
There was a positive association between lethargy and work with one parti cular product grade, but it did not reach statistical significance.
The re were no significant associat ions between any o f the work activities and the head ache symptoms .