Health impact assessment of auto rickshaw and cab drivers due to exposure to vehicular pollution in Delhi: an integrated approach

Vehicular emission is an important contributor to air pollution in the urban environment and impacts the health of commuters as well as drivers. The in-vehicle concentration of pollutants is known to be higher than the ambient environment and varies with the mode of transport. Thus, this study attempts to assess the health impacts of air pollution exposure on auto rickshaws and cab drivers. The study was conducted in Delhi using a triangular approach involving a health perception survey, lung function test and in-vehicle monitoring of particulate matter (PM1, PM2.5, PM10) concentration to assess the health impacts of air pollution on auto rickshaw and cab drivers. A total of 150 respondents (75 from each occupation) were surveyed, and spirometry was performed for 40 respondents (20 from each occupation). Binary logistic regression showed auto rickshaw drivers were exposed to significantly higher in-vehicle PM concentrations in summers and winters and, thus, had a significantly higher risk of developing respiratory, ophthalmic and dermatological health symptoms (p< 0.05 and relative risk >1). Pulmonary function test showed obstructive lung impairment was reported only among auto rickshaw drivers (6%) and restrictive lung impairment was also more prevalent among auto rickshaw drivers (48%) than cab drivers (33%), suggesting a greater vulnerability of auto rickshaw drivers to respiratory health issues. Lung function impairment was associated with age (p= 0.002). The health and well-being of individuals is a matter of global concern, also highlighted in sustainable development goal no. 3. However, it was observed that neither auto rickshaw drivers nor cab drivers used formal/standard protective measures mainly due to unawareness or unaffordability. The study suggests increasing awareness and formulating guidelines to highlight the use of proper protective measures by these vulnerable groups and specific policy measures to protect outdoor workers like auto rickshaw drivers.


Introduction
Urban air pollution is a serious problem worldwide, which directly and indirectly affects the health of a large portion of the global population, predominantly in developing nations. It is ranked as one of the top ten causes of death by a recent study on the Global Burden of Disease (GBD) conducted by the World Health Organization (WHO), contributing to~4.2 million premature deaths every year (WHO 2018;Landrigan et al. 2017;GBD 2016). It is essential to highlight that increasing concentration of air pollutants like particulate matter (PM), ozone (O 3 ), carbon monoxide (CO), oxides of sulphur (SO x ), oxides of nitrogen (NO x ) and volatile organic compounds (VOCs) etc. has resulted in poor air quality which affects local atmospheric visibility, built infrastructure, human health and global climate . Vehicular emission is an important source of air pollution in the urban environment and is known to impact the health of commuters, drivers and those living on or near major or congested roads (Bowatte et al. 2018;Mindell 2017;Monrad et al. 2017;Jain et al. 2016). Samoli et al. (2016), in a study in London, found that exhaust-related pollutants cause an increase in the number of adult cardiovascular hospitalisations. Further, Gan et al. (2012) reported that an increase in deaths due to coronary heart diseases is associated with the increase in black carbon levels in Vancouver. The effects may vary with exposure duration and concentration of pollutants (Sharma and Jain 2019;Kampa and Castanas 2007). The in-vehicle concentration of pollutants such as PM is higher than ambient environment and varied with the mode of transport (Rivas et al. 2017a, b;Jain 2017;Kumar and Gupta 2016;Apte et al. 2011). Thus, outdoor workers such as auto rickshaw, taxi and bus drivers are exposed to such high air pollution levels due to the nature of their occupation requiring extended working hours on-road or in congested areas (Zhang and Batterman, 2013). The exposure also varies with the type of transport (Jain 2017;Rivas et al., 2017a, b;Apte et al. 2011). Ciarrocca et al. (2012 reported that traffic police officers exposed to urban traffic emissions in Italy have higher personal exposure compared to police drivers who used cars fitted with pollen filters and air conditioners for patrolling. Similarly, Zuurbier et al. (2010) reported the highest inhaled dose of pollutants among cyclists compared to those commuting by bus or cars in Arnhem, Netherlands. Furthermore,  reported average PM 2.5 concentrations while travelling by bus and car in Asian cities to be almost two to three times than those in European and American cities. The air quality of Indian megacities is deteriorating, leading to increased morbidity and mortality among the urban dwellers (Jain, 2017;Aggarwal and Jain, 2015;Sharma et al. 2013). For e.g. various studies in Delhi have reported health impacts in terms of mortality and morbidity due to the deteriorating air quality of the city (Kumar and Mishra, 2018;Bajaj et al., 2017;Aggarwal and Jain 2015). This makes outdoor workers exposed to traffic micro-environments in Delhi a vulnerable group to severe health impacts. Bajaj et al. (2017) have reported a higher relative risk of cardiovascular and respiratory health symptoms among traffic policemen than in the control group in New Delhi. In addition, Apte et al. (2011) reported that the invehicle concentration of PM 2.5 and black carbon was higher in auto rickshaws than other closed vehicles in Delhi. Thus, the present study aims to study the health impacts of occupational exposure to urban transport microenvironment among auto rickshaw and cab drivers.

Methodology
The study used a triangular approach to assess the health impacts of occupational exposure to air pollution on auto rickshaw and cab (private taxi) drivers. It involved a crosssectional perception survey to gather responses on health information using a structured questionnaire, along with pulmonary function test and monitoring real-time in-vehicle concentration of PM 1 , PM 2.5 and PM 10 . Fig. 1 presents the research framework used in the study. The study area was National Capital Territory of Delhi (NCT-Delhi) (28.7041°N, 77.1025°E). The areas were categorised as residential, commercial and industrial areas as per the Central Pollution Control Board (CPCB) criteria (GNCTD, 2017), and sites for data collection were selected randomly from each of the strata as presented in Fig. 2.

Data collection and statistical analysis
A total of 150 perception-based surveys were conducted, which included 75 auto rickshaw drivers and 75 cab drivers that were using air-conditioned cars. The respondents were approached randomly in Delhi using a purposive approach. The data was collected during the months of April-June using a structured questionnaire to gather information about the socio-demographic information of respondents and their health perception (as explained in SI, Section S1). The health symptoms included in the questionnaire were related to respiratory, musculoskeletal, cardiovascular, ophthalmic, dermatological and general health symptoms based on literature review and discussion with medical experts. The survey data was analysed to determine the prevalence of various disease symptoms. Statistical analysis was done using Minitab 18. Since, the data was categorical, chi-squared test was used to test the following hypotheses: H 0 : The prevalence/occurrence of disease symptoms is same for auto rickshaw and cab drivers H 1 : The prevalence/occurrence of respiratory symptoms is different for auto rickshaw drivers and cab drivers H 2 : The prevalence/occurrence of ophthalmic symptoms is different for auto rickshaw drivers and cab drivers H 3 : The prevalence/occurrence of dermatological symptoms is different for auto rickshaw drivers and cab drivers H 4 : The prevalence/occurrence of cardiovascular symptoms is different for auto rickshaw drivers and cab drivers H 5 : The prevalence/occurrence of musculoskeletal symptoms is different for auto rickshaw drivers and cab drivers H 6 : The prevalence/occurrence of general health symptoms is different for auto rickshaw drivers and cab drivers t-test was used to compare the difference in the PM concentration inside autos and cabs during summers and winters. Association of respiratory impairment with factors such as smoking, tobacco consumption, age and body mass index (BMI) was determined using binary logistic regression (BLR).

In-vehicle PM monitoring
The real time in-cabin PM samples were collected using a portable optical particle sizer (OPS: TSI Model-3330) which records mass and number PM concentration in 10 channels of different size-range. The OPS measures the PM concentration in the size-range of 0.3 to 10 μm at the flow rate of 1 L/min.  The readings were taken at an interval of 6 s for all samples. Instruments kept for in-cabin air samples were placed at the breathing height, i.e. 1.5 ft above the seating height.
Monitoring of PM was done by hiring the auto rickshaws and cabs for a commute at each time of the day which includes morning, afternoon and evening hours to take representative samples during peak and non-peak hours. PM concentrations were monitored in 15 auto rickshaws and 15 cabs (different models, makes) both during summer and winter months to get a comparative account of the seasonal variation of in-vehicle PM concentration. PM concentrations and pulmonary function tests were carried out in same auto rickshaws as well as cabs.

Pulmonary function test
Spirometry was used as a tool to determine physical lung function impairment. The test was conducted using a Vitalograph alpha spirometer. Spirometry was performed for 20 auto rickshaws and 20 cab drivers based on their willingness and ability to perform the test ( Figure S1, SI Section S2).
Each subject was given a disposable mouth piece to blow at least three acceptable manoeuvres to complete the test. FEV1, FVC and FVC/FEV1 ratio were used to interpret the lung function to be normal, restrictive or obstructive as presented in Figure S2, SI, Section S2. Table 1 summarises the socio-demographic characteristics of the study population. Auto rickshaw drivers spend 47% of their time at work, while cab drivers spend 50% of the day at work. The majority of respondents from both the groups (48% and 33%, respectively) perceived the air quality at the workplace to be hazardous ( Figure S3a, b, SI, Section S3). Vehicles were considered the most significant pollution source, as mentioned by 97% of auto rickshaw drivers and 88% of cab drivers. Most auto rickshaw drivers perceived air quality in their residential areas to be unhealthy (26%), while the majority (30%) of cab drivers perceived it to be hazardous ( Figure S3c, d, SI, Section S3. Further, it is also observed that more than 85% of respondents from both groups used liquified petroleum gas (LPG) as the cooking fuel. It is important to highlight that almost 50% of the sampled group both auto rickshaw and cab drivers considered air quality had a severe impact on their health ( Figure S3e-f, SI, Section S3). In addition, more than half the auto rickshaw drivers and cab drivers found extreme summers to be the most difficult season compared to extreme winters and rains in terms of health impacts and ease of work. Auto rickshaw drivers complained of headaches (54%), rapid heartbeat (28%), muscle cramps (23%), while taxi drivers reported headaches (33%) and muscle cramps (12%) during extreme summers.

In-vehicle PM concentration
The comparative results of in-vehicle PM 10 , PM 2.5 and PM 1 concentrations monitored in cabs (taxies) and auto rickshaws at different times of the day, during summers and winters, are presented in Fig. 3. We found that both auto rickshaw and cab drivers spent almost half of the day outdoors at work. Thus, they are continuously exposed to high PM concentrations for long durations, which is likely to cause severe health impacts over time. The average PM concentration was significantly higher (p= 0.000) inside auto rickshaws as compared to cabs. Similarly, Jain (2017) Goel et al. (2015) in New Delhi reported 30% higher PM 2.5 exposures while travelling in auto rickshaws, whereas, 50% lower exposure while travelling in airconditioned cars compared with on-road ambient PM 2.5 concentrations. Also, the in-cabin PM concentration for auto rickshaws and cabs was substantially higher in winters than summers (p= 0.000). In addition, problems of episodic air pollution events in Delhi increase in winters, which necessitates that proper, timely interventions are made by the concerned authorities to ensure the safety of these vulnerable groups to avoid health inequalities among outdoor exposed groups further. Thus, auto rickshaw drivers can be considered more vulnerable to health issues due to PM exposure. Globally, there are no standards for in-vehicle indoor air quality except South Korea for public transport (PM 10 <150 μg/m 3 ). Therefore, there is a dire need to have indoor air quality standards to safeguard human beings in various micro-environments.

Prevalence of disease symptoms, relative risk and testing of hypotheses
Auto rickshaw drivers had a higher prevalence of all the disease symptoms as compared to cab drivers with breathlessness (~28%), eye redness (~44%), skin rashes (22%), joint pain (~39%), irregular heartbeat and chest discomfort (~9%) and general weakness (~44%) being the most prevalent respiratory, ophthalmic, dermatological, musculoskeletal, cardiovascular and general health symptoms, respectively ( Fig. 4a-f). The value of relative risk greater than one for all symptoms, except the common cold, further suggests that auto rickshaw drivers  are at relatively greater risk of developing health problems as compared to cab drivers who use airconditioned cars (Table 2). Overall, ophthalmic, musculoskeletal and general health symptoms were the most prevalent among both groups. However, chi-squared test (p < 0.05), suggests that auto rickshaw drivers are significantly more susceptible to health issues, which may be attributed to higher exposure to transport microenvironment for all categories of disease symptoms except cardiovascular and musculoskeletal symptoms ( Table 3). The p value >0.05 for musculoskeletal symptoms may be because these symptoms are related to ergonomic factors and long working hours and both the group of workers are involved in driving for long hours (Sekkay et al. 2018).
Further, as explained in Sect. 3.1, PM monitoring results showed that auto rickshaw drivers are exposed to significantly greater PM concentrations than cab drivers. In addition, vehicular emissions also contribute to VOCs, benzene and NO x (Nagpure et al., 2016). Such traffic emissions have been associated with chronic and acute respiratory problems like asthma, COPD, nasal and throat irritation, and cardiovascular health issues (Hansell et al. 2018;Lin et al. 2018;Kumar and Mishra 2018;Fisher et al. 2016). Kisku et al. (2013) reported the prevalence of eye watering (~51%), redness and irritation (~33%) and burning in eyes (20%) among auto rickshaw drivers in Lucknow, owing to high exposure to vehicular fumes. Further, Liu et al. (2015) in Taipei, Taiwan, reported associations between exposure to traffic-related PM 2.5 and cardiovascular health of young adults, and the effects were found to vary with the mode of commute. Exposure to PM, PAHs and black carbon is also known to aggravate various skin problems like ageing, acne and skin cancer (Kim et al. 2016).

Spirometry
The results of pulmonary function tests indicated that 48% of respondents among auto rickshaw drivers as well as 33% of cab drivers had restrictive lung function, while obstructive lung function was reported only for 6% of the auto rickshaw drivers and none of the cab drivers (Figure 5a, b). This could be because the auto rickshaw drivers had higher exposure to PM than cab drivers. Obstructive lung function is associated with exposure to pollutants like NO 2 and PM 2.5 and diseases such as asthma, chronic obstructive pulmonary disease and cystic fibrosis (Hansell et al. 2018;Lin et al. 2018;Fisher et al. 2016;Caronia, 2017). The restrictive pulmonary defect is associated with systemic inflammation, pulmonary fibrosis, interstitial lung disease (Johnson and Threuer 2014). Rice et al. (2015) reported significant association of chronic exposure to PM 2.5 and other traffic emissions with a decline in FVC and FEV1 values for those residing nearby major roads. Binary logistic regression (BLR) was used to test the association of lung function impairment with various factors such as

Use of protective measures
Nearly 54% of the auto rickshaw drivers use cloth/ handkerchiefs to cover their face as a remedial measure, while only 5% of cab drivers do so (Figure 5a, b, SI, Section S4). These are not formal measures such as protective face masks specifically designed to prevent PM inhalation and comply with standards such as the European standards, National Institute for Occupational Safety and Health standards (BS EN 529, 2005; NIOSH Guide to the Selection and Use of Particulate Respirators, 1996). Thus, such measures do not offer ample protection against PM (Noomnual and Shendell 2017;Chaschin et al. 2014;Jung et al. 2014;Langrish et al. 2012). Further, 25% of auto rickshaw drivers were unaware of which mask to use to prevent PM exposure, from a wide range of masks such as surgical masks, quarantine masks, available masks, etc. Further, nearly 19% could not afford proper masks due to high cost, and 10% had discomfort wearing the masks while driving. This suggests the need for government interventions to make appropriate safety masks economically viable and available to all. This would also help reduce health inequalities. The lesser number of cab drivers covering their face while driving was because of their perception that they are not exposed to air pollution within their vehicles.

Conclusion
It was found that auto rickshaw drivers were exposed to significantly higher PM concentrations and had a greater risk of developing various respiratory, ophthalmic, dermatological and other health problems (relative risk >1, p< 0.05). The prevalence of restrictive lung impairment was higher among auto rickshaw drivers, and obstruction was reported only among auto rickshaw drivers (6%). This suggests causality of health symptoms with occupational exposure to a traffic environment. However, the current study did not verify the symptoms with qualified doctors, which can be done in the future for better understanding. Further, it was observed that though auto rickshaw drivers are more exposed to vehicular emissions, they do not use proper/standard protective measures like masks due to unawareness and discomfort in use. Thus, it is suggested that the governing bodies like Ministry of Health and Family Welfare, Government of India, the Central Pollution Control Board and Ministry of Environment, Forest and Climate Change should intervene to provide occupational health safety to these vulnerable groups through various means such as creating awareness about the health impacts of ambient air pollution and the various protective measures. This suggests that there is a need for interventions like the development of guidelines on the use of standard protective measures by outdoor workers and commuters and making protective equipment affordable and available for all. These suggestions would also help in achieving the sustainable development goal on good health and well-being (SDG 3) by reducing morbidity arising from occupational exposure to air pollution.
Acknowledgements The authors would like to acknowledge the contribution of Manasvi Singha, Lady Irwin College, Delhi University for their help during data collection and compilation. We would also like to thank Mr. Ved Prakash Sharma, TERI and Mr. Govind Marwari, Maulana Azad Medical College, for their technical assistance in spirometry.
Author contribution SJ-research concept, methodology, data analysis and manuscript writing. VB-literature review, field work, data analysis and manuscript writing. Data Availability The data used in this manuscript is provided in the supplementary information. In case, further information required will be provided on request.

Declarations
Ethics approval and consent to participate Not applicable.
Consent for publication Not applicable.
Competing interests The authors declare no competing interests.