This study took place at Cotonou, main economic city of Benin. We have included all the shops that were located on the portion of road in Figure 1 from point A to point C via point B. This portion of the road is 4.5 Km long and the traffic is very high throughout the day because it serves the largest market in the city and the port of Cotonou. This road included three roundabouts represented by points A, B and C. For example, the main roundabout B is represented on Figure 2. One person per shop was included in the current study. When there

was more than one person in the office, we interviewed the oldest one. It allowed us to associate at each respondent a unique GPS coordinates of his workplace. We counted a total of 203 shops in the study site. In 194 of these shops, people accepted to participate in the study so we interviewed 194 individuals.

Data collection

A validated questionnaire https://www.ncbi.nlm.nih.gov/pubmed/30065166 on respiratory symptoms (cough, expectoration, wheezing, breathlessness, itchy nose) was administered to the participants of the study. The questionnaire also collected information on the ventilation of buildings, the duration of occupation, GPS coordinates and some socio demographic characteristics of the respondents. For data collection we developed an electronic case report form that integrate taking GPS coordinates. The data were collected in November 2018.

In the first part, we presented the socio demographic characteristics of the respondents and the frequency of respiratory symptoms. The test of chi-square or t-student was used to compare respectively proportion and mean in the two groups that are roundabout and outside the roundabout. The second part was devoted to the cartography of the different respiratory symptoms.

A logistic regression model was performed to assess the association between the occurrence of respiratory symptoms and the geographical location of participant's workplace (roundabout vs outside of roundabout). For the first model, the dependent variable was cough (presence or absence of symptom). For the second model, the dependent variable was itchy nose (presence or absence of symptom). The both models were adjusted on age, level of instruction, seniority on the site, and aeration of the occupied room. Analysis was performed using Stata software and p-value was considered significant when p-value < 0.05.

The present study is part of the project ChairePOL that received the final approval ethic on the number N˚032/MS/DC/SGM/DFR/CNERS/SA du 15 octobre 2018. Based on this, the participants were interviewed after giving their informed consent.

The characteristics of respondents are presented in Table 1. The sex ratio (M/F) was 1.8. It was 1.6 at roundabout against 1.8 outside the roundabout. The average age of respondents was 36.26 ± 11.65 years. There was no statistically significant difference between the two zones (p = 0.38). The study population was predominantly educated (86.6%). This situation was very different between the two groups. At roundabout, 36.54% of respondents had not attended school compared to only 4.93% among respondents outside the roundabout. About 20% and 60% of participants have spent respectively at least ten years and at least four years in their job. The distribution of seniority at the place was significantly different between the two zones (p < 0.001). At roundabout nearly half of the respondents had already more than 10 years of seniority at the place compared to only about 10% of those interviewed along the road. 69.07% of participants worked in the building that was not well ventilated.

The proportion of each type of respiratory symptom is presented in Table 2. 72.68% of respondents reported having at least one of the five respiratory symptoms. This proportion was significantly different (p < 0.001) between the subpopulation at roundabout (96.15%) and out the roundabout (64.08%). The most reported symptoms by the respondents were respectively cough (47.94%),

itchy nose (42.27%) and breathlessness (28.35%). For each of the respiratory symptoms, the proportion was significantly higher at the roundabout than outside the roundabouts.

Figure 3 shows the spatial distribution of symptom of cough.

The multivariate logistic regression to estimate the risk of having a respiratory symptom at roundabout compared to outside the roundabout is presented in Table 3.

The results of logistic regression model showed that the presence of cough and itchy nose were high significantly associated with the place. The risk of cough symptoms was 5.15 higher in a person exercising at the roundabout compared to a person exercising along the road (OR = 5.15; 95% CI = [2.21; 12.04]; p < 0.001). The risk of itchy nose symptoms was 3.44 higher in a person exercising at the roundabout compared to a person exercising along the road (OR = 3.44; 95% CI = [1.55; 7.63]; p = 0.002).

The seniority of the participants on their work is very important. Approximately 20% of them have already spent at least ten years and approximately 60% have already spent at least four years. The duration of exposure to air pollution, especially road traffic, is an important risk factor in the occurrence of respiratory diseases. Indeed, several epidemiological studies have demonstrated that long-term exposure to road traffic as indicated by residential proximity to major roadways or residential traffic intensity is associated with adverse cardiovascular outcomes [14] and cardiopulmonary mortality [15]. Hoek et al. have conducted a pilot cohort study in the Netherlands and found that there was evidence of increasing that long-term exposure to ambient air pollution is associated with deaths from cardiopulmory diseases [16]. In the same vein, Stevanović et al. found that a long term exposure to air pollutants unfavorably influences the health of population and is a risk factor for development and aggravation of certains symptoms and diseases [17]. The analysis of the results of this study also revealed that the majority of the participants worked in non-ventilated rooms. This lack of ventilation leads to a lack of air exchange in the rooms. The compilation of pollutants in the rooms has an accumulating effect, resulting in a condensation of polluted air in them. Thus, in a study conducted in Helsinki and Amsterdam [1], it was estimated that 70% to 84% of some pollutants, measured indoors, came from outside (the internal source linked to smoking being absent in this study). The poor ventilation of work rooms close to road traffic is an important risk factor in the occurrence of respiratory diseases.

Our results show that the majority of participants (72.68%) reported the presence of at least one respiratory symptom. The cough, the itchy nose, the breathlessness and the spits from the chest were the main respiratory symptoms reported by participants. These results are consistent with the results of Garshick et al. who have found that exposure to vehicular emissions by living near busy roadways might contributed to symptoms of chronic respiratory disease in adults [18]. The similar result was found by Bayer-Oglesby et al. who concluded that living close to main streets or in a dense street network increases the risks for certain respiratory symptoms in adults [19]. In Beijing, long-term exposure to traffic-related air pollution in people who live near major roads was associated with a higher prevalence of chronic cough [20]. And finally, Hegseth et al found statistically significant associations between self-reported TRAP (traffic-related air pollution) exposure and respiratory symptoms in a large, random sample from a population living in an area with modest TRAP levels [21].

For the multivariate logistic regression analysis, age, education level, seniority at the place, gender and room ventilation were used as factors to fit the model. The main result highlighted in this work is the difference in risk of respiratory symptoms between roundabouts and outside roundabouts. The risk of cough and itchy nose symptoms were higher in individuals working at the roundabouts compared to those working part from the roundabouts. Zmirou et al., in 2004, showed through their study that the risk of developing asthma and respiratory symptoms such as exacerbation of asthma and infections is higher, depending on the location and the distance occupied from each other to road traffic [22] [23].

As we have shown in previous work [13], concentration levels of pollutants, particularly PM2.5, were higher at roundabouts than outside roundabouts.

During periods of low traffic, pollution levels are equivalent to the general atmosphere of the neighborhood where the axis is located. On the other hand, when the road traffic is important, to this general atmosphere is added the pollution directly induced by the traffic [24]. Along the road traffic, it is observed a dispersion of pollutants which reduces the risk incurred by workers near road traffic. The three roundabouts of our study, were located downtown, most of the companies are in town, there is a strong traffic often observed in this area especially to the hours of return of the service. This result is consistent with the results found by Wu et al in 2005 in Southern of California [25]. The presence of traffic lights at roundabouts is an important factor in the health risk associated with air pollution. In peak areas, pollutants released into the air by motorcycles and vehicles create a fog effect in this area. As for stationed workers, their sales room serves as a receptacle for these pollutants in suspension. Hence a transfer of pollution from the outside to the inside of the premises [26]. These indicate that premises near traffic are likely to be more polluted than those located in an environment less directly influenced by pollution sources, although indoor pollution levels are also conditioned by ventilation and the indoor environment [12].