Gender Disparity in Lung Function Abnormalities among a Population Exposed to Particulate Matter Concentration in Ambient Air in the National Capital Region , India

Data collected globally between 2002 and 2010 to track air pollution trends in 189 megacities shows Indian cities among the top ranked for air pollution, with the highest ambient air pollution and smog on the Earth’s surface.1 A widely used health-relevant indicator of the air pollution mixture is particulate matter (PM).2,3 The Global Burden of Disease (GBD) 2010 report estimates over 2.1 million premature deaths and 52 million years of healthy life were lost in 2010 due to ambient Background. The World Health Organization (WHO) global air quality study shows that 27 Indian cities, including New Delhi, are among the one hundred cities with the worst air quality globally. The scope of airway obstruction cases among residents in locations with critical air pollution levels like particulate matter (PM) pollutants PM2.5 and PM1 has not been addressed in the National Capital region, India. Objectives. The present cross-sectional study was undertaken to assess the forced expiratory volume in one second (FEV1) % predicted abnormalities among residents living in the National Capital Region, India with respect to their exposure to particulate matter pollutants (PM1, PM2.5) in ambient air. Methods. Eight hundred and fifty-four residents, including 433 men and 421 women ranging in age from 18–70 years, living in the National Capital Region (NCR) of India participated in the study. Particulate matter concentrations in ambient air (PM2.5 and PM1) were monitored at 10 residential locations in the National Capital Region, India (New Okhla Industrial Development Authority (NOIDA) and Gurgaon). The lung function test (FEV1) was conducted using a spirometer. Results. The Indian Air Quality Index showed either very poor or severe levels for PM2.5 at all study locations. A significant negative linear relationship was found between higher concentrations of PM1 and reduced FEV1% predicted values (r = -0.8, p < 0.05). The prevalence of airway obstruction cases (79.6%, odds ratio 1.96, confidence interval 1.42–2.71) was higher (p<0.001) among female subjects compared to their male counterparts. Even though there was a significant decline in FEV1% predicted among 80% of cases in women, only 19.24% cases were in the moderate category and 6.18% cases in the severe category. The severe category of FEV1% predicted cases showed greater respiratory symptoms than the other two categories, which denotes higher risk among those in the severe category. The present study shows that obstruction cases increased from 1.97 to 7.40% and 2.73 to 14.93% in women, with a corresponding increase in PM1 and PM2.5 from the minimum to maximum concentration. Conclusions. Since the women in this study were non-smokers, the PM in ambient air can be considered to be the major reason for the decline in lung function. The sources of PM pollutants in the study locations are large scale infrastructural development activities such as building and road construction activities. Narrowed lung airways can alter the airway caliber or resistance and flow rates proportional to the airway radius, especially in smaller airways. The present study suggests the need for policy makers and stake holders to take the necessary steps to identify PM sources and reduce the emissions of PM concentrations in ambient air. Competing Interests. The authors declare no competing financial interests.


Introduction
Data collected globally between 2002 and 2010 to track air pollution trends in 189 megacities shows Indian cities among the top ranked for air pollution, with the highest ambient air pollution and smog on the Earth's surface. 1 A widely used health-relevant indicator of the air pollution mixture is particulate matter (PM). 2 10 ) and within a preset geometric standard deviation, the number of subjects with an forced vital capacity (FVC) predicted percentage of <70% increased from 5% to 8% in Switzerland. Previous cross-sectional studies have reported that air pollution can cause both retardation of lung function growth and acceleration of lung function decline. 5, 6 The assessment of lung function is considered to be a significant predictor of respiratory diseases. 7 The World Health Organization (WHO) global air quality study shows that 27 Indian cities, including New Delhi, are among the top one hundred cities with the worst air quality. 8 The increased prevalence of respiratory symptoms and decrements in lung function among residents breathing polluted air has been previously reported. 9, 10 Rapid deterioration of lung function was observed in a cohort study in Los Angeles due to long-term ambient exposure to air pollutants. 5 The American Thoracic Society (ATS) recommends that the severity of airflow obstruction should be based on the predicted percentage of the measured forced expiratory volume in one second (FEV 1 ). 11 Recent findings revealed greater respiratory symptoms among children residing in commercial areas compared to industrial and residential areas and found evidence for a positive association with PM 10 levels in the air. 12 Earlier Indian studies on the association of air pollutants with respiratory health problems focused on the effects of vehicular pollution among traffic policemen in Patiala, India; occupational exposure to carbon monoxide pollution levels in heavy traffic zones in Chandigarh, India; and lung function decline in healthy non-smoking male transport workers in Chennai, India due to exposure to Kesavachandran  Quality assurance/quality control (QA/QC) procedures were maintained as per the instruction manual of the instrument. Manual-Zero sets the measurement baseline of the EPAM-5000 to zero µg/m 3 . The Manual-Zero check was performed prior to beginning a new set of measurements.
A flow meter was used to assess and ensure a flow rate in the instrument of 4 L/minute before air monitoring. The monitoring exercise was conducted at a height of 10 m from the ground and the instrument was placed on the roof of nearby houses. The criteria for the selection of air monitoring site at each study location was 200 m away from traffic intersections. The study was conducted during the summer season from July to September from 2008 to 2010 with the same monitor. Table 1 shows the Indian Air Quality Index (IAQI) proposed by the CPCB, India (2014) to indicate the extent to which ambient air quality has exceeded the regulatory limits. 23 No regulatory body in India has proposed an IAQI for PM 1 .
The results were interpreted using the predicted reference equation for Indians. 24 The lung function test was repeated three times in one subject for reliability and a maximum variation     Table 2 shows the mean age, height, and weight of the residents living in different locations. No significant difference in age was observed between the locations, whereas a significant difference (p < 0.001) was found for height and weight.

Air Quality at Study Locations and its Associated FEV 1 % Predicted Outcomes in Residents
The temperature varied from 24-40ºC, with an average of 31ºC, while average humidity and wind speed were found to be 58% and 12 km/hr, respectively. On the other hand, an average temperature of 30ºC was observed in the Gurgaon study locations without any marked differences in average humidity (55%) and wind speed (12 km/hr). No rainfall was observed during the study period at either location.
All study locations showed concentrations exceeding the National Ambient Air Quality Standards Research among residents at all study locations was observed at Dhankot ( Table 3). The present study shows that the prevalence of FEV 1 % <80% predicted was higher among residents living at locations with higher PM concentrations, especially at Bhaktiyarpur (78.7%), Garhi (75.6%), Sadhrana (75.5%) and Chandu (75%) ( Table 3).

Respiratory Symptoms with Respect to FEV 1 % Predicted
A higher prevalence of cough symptoms was observed in Bhangal and Sadhrana. Cold symptoms were more or less the same at all locations, while more complaints of breathlessness were observed in Bakhtiyarpur and Garhi; productive cough was more common among residents of Chandu and Maharishi Nagar, and the prevalence of combined symptoms such as cough and cold/cough with breathlessness were higher among residents of Dhankot ( Table 4). The overall prevalence of respiratory symptoms among residents was higher for residents of Bakhtiyarpur (24.1%), followed by Bhangal (23.6%), and Garhi (23.6%). All of the above three locations fell into the severe IAQI category ( Table 4).
In the FEV 1 % predicted >80% and 50-79% categories, the prevalence of respiratory symptoms was comparatively lower than the prevalence of residents without respiratory ailments (Table 5). An increase in the frequency of prevalence with respect to respiratory illnesses was also observed among residents in the 30-49% and <29% categories of

Relationship of PM Concentration and FEV 1 % Predicted
A significant negative linear relationship was observed between higher PM 1 concentrations and FEV 1 % < 80% predicted ( Figure 1). However, the correlation between PM 2.5 concentrations and FEV 1 % <80% predicted did not attain statistical significance at the 5% level ( Figure 1). Table 6 shows adjusted FEV 1 % predicted for age, sex, smoking status and respiratory symptoms among residents of the study locations. A higher prevalence of obstructive respiratory abnormalities was

Discussion
In the present study, we analyzed air quality and lung function of residents in different parts of the NCR, India. FEV 1 % predicted was lower among residents living in poor air quality, which indicates cases of obstructive lung function, especially among severe cases. The severe category of FEV 1 % predicted also showed more respiratory morbidity cases compared to the other two categories of mild and moderate respiratory obstruction. Particulate matter (PM 2.5 , PM 1 ) at a higher concentration can lead to poor ambient air quality in residential areas as evidenced from the present study. The present study proposes that higher PM concentrations (especially PM 1 ) can lead to considerable variation in the prevalence of FEV 1 % <80% predicted among residents living in locations with a poor air quality index. Females are at a higher risk of obstructive lung function abnormalities compared to males. The confounding effect of age, sex, and smoking status on the risk of FEV 1 % predicted among residents was also demonstrated in the present study. Lung function (FEV 1 and peak expiratory flow rate (PEFR)) is influenced by age, sex, smoking habits, height, and weight, etc. 28,29 Therefore, understanding the role of these physical characteristics on lung function is crucial. The mean ages of study subjects across locations were similar. Women showed a higher risk of obstruction-related lung function abnormality due to PM exposure than men. The study also found that the prevalence rate of obstruction cases in women increased with a corresponding increase in PM 1 and PM 2.5 from minimum to maximum concentration. Even though there was a higher risk of lower FEV 1 % predicted among females, only 19% were in the moderate category and 6% were in the severe category of respiratory problems. Mild category (50%) cases are mostly reversible and can be considered to be temporary.
Spirometric lung function parameters including FEV 1 % predicted are used as an early diagnostic tool and to monitor therapy effectiveness or the course of respiratory disease. 30 Women showed a higher prevalence of FEV 1 % <80% predicted than their male counterparts. The genderrelated prevalence of respiratory health problems in females may be attributed to biomass fuel exposure during cooking, which generates Research higher PM in indoor air. 31 Female subjects in all study locations in the present study cooked food using liquified petroleum gas (LPG). LPG fuel is considered to be a better fuel compared to biomass fuels, as reported earlier, and all were non-smokers. 31 Since the PM in indoor air was not measured in any household at any of the study locations, the role of indoor air pollutants in the reduction of lung function among women is unclear. However, an earlier study reported a reduction in lung function among women during cooking activities. 19 Age and height are considered to be confounders for FEV 1 . 32 In this study, age and height were similar across locations, and hence the confounding effect on FEV 1 % predicted was nullified. Thus, the higher prevalence rate of obstructive lung functions among females can be considered to be an effect of PM, especially at fractions of 2.5 and 1 microns. Predicted values of lung functions should be reliable and representative of the reference cohort. 30 In the present study, we used the representative reference values proposed for Indian male and female populations. This also makes our assumption of women being at higher risk due to PM pollution more reliable, since the reference equation used was for an Indian population. The decline in FEV 1 % predicted can also be attributed to the narrowing of the lower airways. In these conditions, the airway caliber or resistance and flow rates are proportional to the airway radius, especially in smaller airways. 33 The level of FEV 1 in relation to a predicted value is an indicator of the presence and severity of airflow impairment. The excessive rate of FEV 1 decline with respect to a predicted value is an indicator of ongoing respiratory health effects such as chronic obstructive pulmonary disease (COPD) progression. 6 COPD is more commonly seen in people living in polluted areas, as well as those working in a dusty environment. 34 Although COPD is multi-factorial by nature, it is generally caused by inhaled environmental and occupational exposures that may progress to chronic bronchitis, small airway disease or emphysema, and airway obstruction after prolonged and persistent injury to the airways. 6,34 A recent study by our group has shown that poor ambient air quality can lead to a reduction in lung function. 35 Hence, the effect of PM 2.5 and PM 1 in ambient air can be proposed as one reason for the higher prevalence of FEV 1 < 80% predicted among residents. The risk of respiratory symptoms among study subjects was associated with higher concentrations of PM 2.5 and PM 1 , although this was not significant in the present study. Higher rates of hospitalization or admission to emergency departments due to COPD were observed on days with elevated pollution in an earlier report.

Study Strengths and Limitations
The strength of the study is the large sample size of 854 subjects (433 males and 421 females). To date, this is the largest study sample size in an Indian population after recommended guidelines were introduced for ambient air PM 2.5 . Although this study has many strengths, it is not without limitations. The PiKo-1 spirometer used in the present study can measure PEF and FEV 1 . However, the device cannot be used to measure forced vital capacity or ratios used to categorize the patients with restriction-related lung function abnormalities. Further study is warranted to correlate the sources close to an individual participant's residence (different fuel types, environment tobacco smoke and other indoor air sources) and around their work place (occupational exposures) with their respiratory health status. Personal dosimeter analysis was not used in this study due to feasibility and practical limitations. Even though the data set was five years old, large scale infrastructural development activities such as building and road construction activities observed in these areas continue to generate large amounts of dust in the ambient air in the study locations. Hence, we presume that the study data is still relevant. Due to the limited sample size in each study location, a statistical correlation analysis of lung functions with respiratory problems was not conducted for each study location. Household indoor air monitoring was also not conducted, which may be considered another limitation of the present study.

Conclusions
Higher concentrations of particulate Research matter (PM 2.5 , PM 1 ) have led to poor ambient air quality in the study locations, as evidenced from the present study. The present study shows that higher PM concentrations (especially PM 1 ) can lead to mild obstruction-related lung function abnormalities among residents, especially among women living in these locations. A reduction of PM emissions through strict interventions and regulations at the source or point of emission should be considered by policy makers and stake holders in these locations to reduce the burden of PM concentrations in ambient air. This can further reduce PM-related respiratory health problems. Similar studies should be conducted in other parts of the country to validate the observations of the present study. Exposure-response analysis for PM with respiratory health problems should also focus on children and the elderly population living in different geographical locations in India with poor air quality to understand the national scope of this problem.