MATERIALS AND METHODS

All 289 municipalities in Sweden were contacted and asked to supply home addresses of sewage workers and controls. Controls were recruited among other municipal workers not exposed to sewage, such as workers in drinking water plants and gardeners. A total of 257 of the municipalities delivered addresses of sewage workers (n = 1953) and 211 delivered addresses of controls (n = 1425). All subjects received a mailed questionnaire. All questionnaires were checked at arrival, and when incomplete answers were given, the respondents were phoned and asked to give complete answers. After two reminders, the response rate was 74% among sewage workers (n = 1453) and 59% among controls (n = 839). When checking the questionnaires, several subjects recruited as controls were found to be partly exposed to sewage. In the analyses, they were classified as partly exposed (n = 332). Among the sewage workers, 59 subjects were not exposed to sewage. They were excluded in the analyses.

Questionnaire

The questionnaire was based on a standard questionnaire developed in a European Union project on investigation of persons handling waste (www.miljomedicin.gu.se). It contained a series of items on the working environment, the length of employment at the present work site, and the type of work carried out, as well as existing diseases. These were followed by a series of questions on different symptoms and diseases present during the most recent 12 months and whether they were work related. Questions were posed on airway and gastrointestinal symptoms, joint pains, and central nervous system symptoms, including headache, unusual tiredness, and concentration difficulties. Special questions related to physician diagnosed allergy, eczema, and asthma. Finally, the questionnaire contained items on smoking habits and alcohol use. Chronic bronchitis was defined as cough with sputum for at least three months a year for a period of at least two years. The question “Have you had episodes of flu like symptoms such as fever, chills, malaise, muscle or joint pain, and perhaps also cough, breathlessness, and weakness, and felt completely well the following day?” was also posed. A positive response was defined as toxic pneumonitis.²¹

Statistical analyses

The differences between sewage workers (exposed), subjects partly exposed (partly exposed), and controls were analysed using Student’s t test and non-parametric tests (χ², Fisher’s exact test).

Logistic regression analyses were performed to compute adjusted odds ratios (OR) with 95% confidence intervals (CI) while controlling for age, gender, and smoking habits (never-smoker/ever-smoker). As several sewage workers also worked in drinking water plants, which also use chemicals, it is theoretically possible that this workplace exposure could influence the results. Separate analyses controlling also for work in drinking water plants were therefore performed. To analyse the prevalence of symptoms at different work stations, a stepwise multivariate logistic regression model was applied, with adjustments for age, gender, smoking habits (never-smoker/ever-smoker), and interactions between different work stations. Another stepwise multivariate logistic regression model, with adjustments for age, gender, and smoking habits (never-smoker/ever-smoker), was applied to adjust for interactions between different symptoms.

RESULTS

Table 1 shows the characteristics of the study population. There was a larger proportion of males in the exposed groups compared with controls. There were no differences in smoking habits, alcohol use, full time employment, and physician diagnosed asthma or allergy between the exposed and controls. Among the exposed, a larger proportion worked shift/indoors, had a shower and changed clothes after work shift, had cleaning duties, used protective equipment, and reported chemical use and dust exposure at work compared with controls.

Analyses of working characteristics in relation to years of employment as a sewage worker showed that the proportion of workers who reported eating, drinking during work shift, and having a shower after work was higher with an increasing number of years of employment in a time–response fashion (data not shown).

Table 2 shows the proportions and adjusted OR with 95% CI for self reported airway symptoms among the exposed, partly exposed, and controls. Among the exposed, an increased risk for upper and lower airway effects such as nose irritation, congested nose, cough, breathlessness in different situations, wheezing, chest tightness during work shift, chronic bronchitis, and toxic pneumonitis was found compared with controls.

Among the partly exposed, an increased risk for airway effects such as breathlessness when exposed to gases etc, chest tightness during work shift, chronic bronchitis, and toxic pneumonitis was found compared with controls.

Table 3 shows the proportions and adjusted OR with 95% CI for self reported gastrointestinal symptoms and joint pains among the exposed, partly exposed, and controls. Among the exposed, an increased risk for soft stools/diarrhoea related to work, nausea, and stomach pain was found. There was a decreased risk that other family members suffered from diarrhoea at the same time. An increased risk for joint pains was found among the exposed compared with controls, and the joint pains were related to pains in more than four joints but not with loading. Among the partly exposed, an increased risk for soft stools related to work and stomach pain was found compared with controls.

Table 4 shows the proportions and adjusted OR with 95% CI for self reported central nervous system symptoms among the exposed, partly exposed, and controls. Among the exposed, an increased risk for headache, concentration difficulties, unusual tiredness, and heaviness in the head was found compared with controls. Among the partly exposed, an increased risk for unusual tiredness was found.

Table 5 shows a multivariate logistic regression analysis, controlling for age, gender, and smoking habits, for interactions between different symptoms among the exposed in relation to controls. In this model, an increased risk for toxic pneumonitis, unusual tiredness, congested nose, and soft stools related to work was found among the exposed.

Table 6 shows a multivariate logistic regression analysis of reported symptoms in relation to work at different work stations. Adjusted OR with 95% CI are presented, controlling for age, gender, smoking habits, and interactions between work at different work stations.

Some of the above reported symptoms were increased in relation to work at pump stations or filter units or sludge handling. Work at covered basins was not associated with an increased risk for any symptoms.

DISCUSSION

The controls were recruited among other municipal personnel not exposed to sewage and with similar socioeconomic status as the sewage workers. The response rate was lower among controls compared with sewage workers. This is often a practical reality, as the working environment of the controls was not investigated. However, the prevalence of different symptoms among the controls was similar compared with other similar control groups in previous studies.^{7–9,13,14,22} The results were based on self reporting of certain symptoms and diseases, working environments, and exposures by a mailed questionnaire. The sewage workers may be more prone to report their symptoms, especially if the symptoms have been discussed in association with working with sewage. Thus, there is a possibility that over reporting to some extent can be responsible for the differences found between the exposed and controls.

A significantly increased risk for several airway symptoms, chronic bronchitis, and toxic pneumonitis was found among the sewage workers compared with controls. Several previous studies have shown that airway symptoms are more common among sewage workers than in controls.^3–10 An increase in airway symptoms has also been reported in earlier investigations on workers in organic dust environments in general.²¹ An increased prevalence of toxic pneumonitis is a common finding in investigations of organic dust environments and has been related to endotoxin exposure.²¹ In sewage workers handling sludge, similar reactions with flu like symptoms have previously been described.^4,6 In such workers, flu like symptoms have also been related to Legionella pneumophila infection.¹¹ Furthermore, an increased risk/prevalence of asthma and chronic bronchitis has been reported among sewage workers,^7,8 as well as decreased lung function values.^6,7,12

An increased risk for non-specific gastrointestinal symptoms was found among the sewage workers. Work related lower gastrointestinal symptoms were the most common. There was a decreased risk for other family members to suffer from diarrhoea at the same time, indicating that the symptoms were related to the work and not to a family derived infection. Several previous studies have shown that gastrointestinal symptoms are more common among sewage workers than in controls.^3,5,9,13,14 The symptoms reported from the gastrointestinal tract are not diagnostic of specific diseases such as stomach ulcer or ulcerative colitis.

An increased risk for central nervous system symptoms such as headache, heaviness in the head, unusual tiredness, and concentration difficulties was found among the sewage workers. Similar findings have been reported previously.^5,10,23,24 An increased prevalence of unusual tiredness is frequently reported in studies investigating workers/persons in organic dust environments.^21,25 An increased risk for joint pains was found among the sewage workers, and similar results have been reported previously.⁹ In this study, the joint pains were related to pains in more than four joints but not with loading, indicating an inflammatory/systemic response among the sewage workers. However, this must be further investigated in future studies, including clinical markers of inflammation.

Analyses controlling for interactions between different work stations at sewage treatment plants, indicated that work at covered basins is not associated with an increased risk for symptoms. This finding can be the result of a different exposure situation at these work stations, including less aerosolisation of the sewage water.

Regarding possible causative agents for the symptoms observed, viruses such as Norwalk agent, other microorganisms, or endotoxin have been suggested.^3,9,10 Of the different components in microorganisms, bacterial endotoxins are of particular interest. They can be present in sewage treatment plants in amounts that exceed those that give rise to symptoms and illness.^9,21 The acute effects of endotoxins are well documented in several inhalation experiments in man.^26–30 Relations between symptoms such as those found in this study (for example, airway symptoms, toxic pneumonitis, and unusual tiredness) and endotoxin exposure in different environments have previously been reported.²¹

In conclusion, the results of this questionnaire survey show an increased risk for airway, gastrointestinal, and general symptoms such as joint pains and central nervous system symptoms among sewage workers. Clinical investigations are needed to determine the cause of the reported symptoms among sewage workers, and further field studies are required to assess the causal agents.