|DR. S. D. GAUR BEST PAPER AWARD ON ENVIRONMENTAL HEALTH
|Year : 2010 | Volume
| Issue : 2 | Page : 98-103
Association of air pollution and mortality in the Ludhiana city of India: A time-series study
Rajesh Kumar1, Suresh K Sharma2, JS Thakur1, P.V.M. Lakshmi1, MK Sharma2, T Singh3
1 School of Public Health, Post Graduate Institute of Medical Education and Research, Chandigarh, India
2 Punjab University, Chandigarh, India
3 Dayanand Medical College, Ludhiana, Punjab, India
|Date of Web Publication||27-Nov-2010|
School of Public Health, Post Graduate Institute of Medical Education and Research, Chandigarh, Punjab
Source of Support: None, Conflict of Interest: None
| Abstract|| |
Background : With rapid industrialization, the quality of the air is being compromised in several Indian cities. Hence, the effect of air pollution on mortality was studied in the Ludhiana city of Punjab in northern India. Materials and Methods: Air quality and meteorological and mortality data were obtained for 2002-2004. Punjab Pollution Control Board monitored air quality on specific week days at different sites. Respirable suspended particulate matter (RSPM) (equivalent of PM 10 ) was measured by the gravimetric method and NOx and SO 2 by chemical method. The estimation of the daily average RSPM level was attempted by combining 24-h average of the monitoring stations working on a particular day. Sahnewal Airport records temperature, dew point, and relative humidity at 8.30 am, 11.30 am, and 5.30 pm. Visibility of fixed landmarks is observed manually every hour from 6.30 am to 6.30 pm. Daily death records were obtained from the civil registration system. The association between visibility as proxy for RSPM and mortality was established using the generalized additive model (GAM) with natural spline smoothers at 6, 3, 3 df in R software with deaths (excluding accidents) as a dependent variable. Smoothers for day of the week, temperature, and relative humidity were also included in the model. Results: Air quality monitoring days for different monitoring stations ranged from 86 to 138 per year. The annual mean RSPM ranged from 226.7 to 269 μg/m 3 , SO 2 from 11.6 to 20.9 μg/m 3 , and NOx from 32.2 to 46.3 μg/m 3 . The mean (SD) temperature was 25.6 (7.9)°C, relative humidity was 58.1 (19.3)%, and visibility was 3398 (1418) m. Overall 28,007 deaths were registered, with an average of 25.4 deaths (SD 5.8) per day. The association between air quality as indicated by visibility (haze) and daily mortality was found to be statistically significant. For every 1 km decrease in visibility at midday, mortality due to natural causes increased by 2.4%. Conclusions: In Ludhiana, air pollution levels were quite high. The air quality (as measured by visibility) was significantly associated with mortality.
Keywords: Air quality, Visibility, Mortality, Time series
|How to cite this article:|
Kumar R, Sharma SK, Thakur J S, Lakshmi P, Sharma M K, Singh T. Association of air pollution and mortality in the Ludhiana city of India: A time-series study. Indian J Public Health 2010;54:98-103
|How to cite this URL:|
Kumar R, Sharma SK, Thakur J S, Lakshmi P, Sharma M K, Singh T. Association of air pollution and mortality in the Ludhiana city of India: A time-series study. Indian J Public Health [serial online] 2010 [cited 2020 Aug 9];54:98-103. Available from: http://www.ijph.in/text.asp?2010/54/2/98/73278
| Introduction|| |
Several time-series studies have established the association of air pollution with cardiorespiratory mortality.  Short-term exposure to outdoor air pollution has been linked to adverse health effects, including increased mortality, increased rates of hospital admissions and emergency department visits, and decreased lung function.  These acute effects are attributed to the exacerbation of preexisting cardiorespiratory conditions.  Cross-sectional studies have also reported the association of respiratory symptoms with poor ambient air quality. ,, Cohort studies have shown deterioration in lung function with air pollution. , Recent multicity analyses conducted in USA, Canada, and Europe provide further provide evidence supporting the validity and plausibility of this association. ,,,,
Most of the studies on air pollution and health have been conducted in industrialized countries of the North. The climatic conditions, characteristics of the population, and the composition of air pollutants may have important bearing on the health effects of air pollution. Not only do the physical form and properties of the air-borne particles influence their distribution in the environment, but their concentration also depends upon the wind-speed and the stability of the atmosphere which is responsible for the upward diffusion of the pollutants. Thus, studies need to be conducted in developing countries also, where the characteristic of outdoor pollution, meteorological conditions, and sociodemographic patterns differ from North America and Western Europe. 
In India, many cities are experiencing heavy air pollution.  An ecological study carried out recently among the residents of an industrial town of Punjab has shown an association of outdoor air pollution with chronic respiratory morbidity.  Acute effects also need to be studied so that a comprehensive policy can be evolved for the prevention of the health effects of air pollution. Therefore, a time-series study was carried out to determine the association of air quality with mortality in the Ludhiana city, which is one of the critically polluted areas in India.
| Material and Methods|| |
The study was carried out in the Ludhiana city, the largest city of Punjab in northern India with a population of 1.5 million and an area of 135 km 2 . Major sources of pollution are vehicular emissions, industries, and domestic and agricultural waste burning. This city is located at a longitude of 76º 54' E and a latitude of 30º 42' N.
Punjab Pollution Control Board (PPCB) monitors air quality for respirable suspended particulate matter (RSPM), NOx, and SO 2 using a high-volume air sampler. A gravimetric method is used for RSPM (equivalent of PM 10 ) and a chemical method for NOx and SO 2 . Eight-hour averaging is done for RSPM and 4-h average is done for NOx and SO 2 . Bureau of Indian Standard (BIS) ,, thresholds are used for monitoring these air pollutants.
Air quality was monitored intermittently at four sites. The monitors at sites 1 and 2 were located in a commercial area (Vishwakarma Chowk and Jolly Hosiery) which operated on Tuesday, Thursday, and Saturday, whereas the monitors in residential and industrial areas at sites 3 and 4, respectively, (Verka Milk Plant and Rita Sewing Machine) worked on Monday, Wednesday, and Friday. The station at site 2 stopped functioning in 2003; hence, it was excluded from the analysis. None of the stations monitored air quality on weekends and public holidays. Air sampling sites have been selected so as to be representatives of different areas of the city; however, the industrial, commercial, and residential areas in the city are not well separated. Residences are in the close proximity of the industries.
Air quality data were obtained in an electronic format from the PPCB for 3 years, i.e., 2002, 2003, and 2004 for all the criterion air pollutants, i.e., RSPM, NOx, and SO 2 . The data were checked for quality and consistency. A 100% recheck was done by comparing the electronic data with the data recorded in the site registers (log books) of the PPCB local office. Only a few data entry errors and data averaging errors were detected which were corrected. The recording of air quality for ≥ 75% of the time in a day was taken as a complete daily recording, i.e., all the three 8-h interval recordings for RSPM and at least five out of the six 4-h interval recordings for NOx and SO 2 . The city daily average was derived by averaging the daily average of each of the station available for that day from three monitors which worked till the end of the study. The air quality data were not available for more than 60% of the observations; hence, missing data were not imputed.
The Indian Meteorological Department, New Delhi, records meteorological data at the airport of the Ludhiana city. Temperature, relative humidity, dew point, wind-speed, and wind direction are recorded daily at 8.30 am, 11.30 am, and 5.30 pm. Visibility is measured (in meters) 13 times daily at 1-h intervals from 6.30 am to 6.30 pm. The visibility of fixed landmarks at distances starting from 0 to 100 m with a 10-m gap, then from 100 m to 2000 m with a 100-m gap and then from 2000 m to 6000 m with a 500 m gap is recorded manually. Meteorological data were available for all the 365 days in a year.
Municipal Corporation records mortality data at two centers which register deaths for zone A and B, and C and D of the city, respectively. Government and private hospitals report deaths that occur in their premises and heads of families register deaths which happen at home or places other than hospitals. The data available in the registers by registration number, name of the deceased, father's name, age, sex, date of death, cause of death, place of death, and address of the deceased were entered in the computer. Standard operating procedures were developed for maintaining the quality of data entry. During data entry, 1% of the entries were randomly compared with the records, each week in each of the two data entry centers. A sample of 2% records from the death register was compared with the original death forms. At the end of data entry, a 100% recheck was again done. Sex and age were not mentioned in 21 (0.1%) and 207 cases (0.7%).
The data were examined for quality and consistency. Descriptive statistics, i.e., mean, standard deviation, median, and range were computed. A generalized additive model (GAM) was fitted in R statistical software with the quasi-Poisson function for mortality due to all natural causes (excluding accidents) as the dependent variable and daily average visibility at midday (at 12.30, 1.30, and 2.30 pm) as the independent variable with penalized or natural spline smoothers for meteorological parameters (i.e., temperature and relative humidity). The model also takes into account the effect of the day of the week (i.e., dichotomous variables for each day of the week from Monday through Friday), temperature and relative humidity, using 3-12 degrees of freedom (df).
The adopted model was
Log (E(mortality))= air quality (visibility) +s(day effect)+ s(temperature)+s(relative humidity).
After the base core model was developed, sensitivity analysis was also conducted. The generalized cross-validation (GCV) score did not vary much between the natural spline at 6, 3, 3 and 8, 4, 4 df and between penalized spline at 6, 3, 3 and 8, 4, 4 df. As there were large numbers of missing observations for RSPM, hence, the analysis was done using visibility as a proxy measure for RSPM. Separate analysis was carried out for deaths in the 65+ year age group. Small number of deaths in the 0-4 year age group did not allow age-specific analysis for the 0-4 year age group.
| Results|| |
The air quality monitoring days for different monitoring sites ranged from 86 to 138 days in a year. Most (80-90%) of the monitoring days had all three 8-h interval observations for RSPM and at least five 4-h interval observations for NOx and SO 2 . The city daily average was derived by averaging the daily average of each of the station available for that day. The daily average of RSPM, SO 2 , and NOx were higher in 2004 as compared to 2003 and 2002 [Table 1]. This may be due to an increase in industries and number of vehicles during the course of the study. The RSPM was lowest in 2002 and highest in 2004. Before computing the city average, the correlation between various sites was examined. The correlation between sites 1 and 2 which operated on the same week days ranged from 0.1 to 0.6 and between sites 3 and 4 the range of correlations was from 0.1 to 0.7. RSPM and NOx had a significant correlation at all sites except site 4, whereas between RSPM and SO 2 , a significant correlation was observed only at site 1. SO 2 was correlated with NOx at sites 1 and 4. The summary of statistics of meteorological parameters is presented in [Table 2]. Large seasonal variations are seen in all meteorological parameters.
Overall in the 3- year period, 28,007 deaths were registered, with an average of 25.4 deaths per day. The age and sex distribution of the deaths is shown in [Table 3]. There were more deaths among males (65%) and in the above 45 years age group (67%). Only 787 (2.8%) of the deaths were due to accidents and these were excluded from analysis. Most of the deaths (71.2%) fall under the category of "symptoms, signs, and abnormalities classified elsewhere" due to improper certification of the cause of death.
The association between RSPM and mortality could not be statistically tested as large numbers of values were missing in the air pollutant data. Visibility was taken as a surrogate measure for studying the association between air pollution and daily mortality as visibility was shown to have a correlation with RSPM in many studies. ,, A significant inverse relationship was observed between visibility and mortality [Figure 1]. The relative risk of deaths due to natural causes was significantly higher at lag 0-3 [Table 4]. For every 1-km decrease in midday visibility (i.e., higher air pollution), the mortality increased by 2.4%.
|Table 4 :Effect of air quality (visibility) on mortality, in Ludhiana, 2002– 2004|
Click here to view
| Discussion|| |
The levels of RSPM in Ludhiana were quite high as compared to the standards advocated by the World Health Organization. However, the levels of NOx and SO 2 (11.9 and 5.8 μg/m 3 ) obtained in the study were lower than the levels observed in China (66.6 and 44.7 μg/m 3 ).  A similar pattern has been observed in other Indian cities which also report quite low levels of NOx and SO 2 . We could not test the association of RSPM and mortality due the fact that there were large numbers of missing observations for air quality monitoring. It was difficult to construct the daily average of RSPM for the city as the monitoring stations located in different areas of the city functioned on different days, and the daily average of different monitors was very different. Hence, the daily RSPM average of the city calculated from the 24-h average of all the monitors working on different days was lower on the days when the monitor at site 1 located in a commercial area was not working which had a consistently high recording of RSPM. The classification of the sites into residential, commercial, and industrial areas was also not very satisfactory as the RSPM levels and other air quality parameters in a commercial area were higher than the levels found in the industrial area.
A moderate to strong correlation (r 0.7) has been observed between visibility and air quality. ,,,, A study from New Delhi has also reported a good correlation between visibility and RSPM (Uma Rangarajan, personal communication). Hence, the association of visibility and mortality was explored by taking visibility as the proxy measure for air quality. A statistically significant association between daily mortality and average midday visibility at 0-3 day lag was observed in this study [Table 4]. The model was found to be robust as the findings with both natural spline and penalized spline smoothers at 6, 3, 3 and 8, 4, 4 df were similar. Several studies conducted in various parts of the world showed a significant association between air quality levels and daily mortality. ,
Studies based on retrospective data face several difficulties, and this study is no exception. Time-series studies require daily data on exposures, outcomes, and confounders. Data for air pollution in Ludhiana were not available for a large number of days, though meteorological data were complete. On the basis of the age-specific mortality rates and age distribution of the population of urban India, it seems that about 5-23% of the deaths may not have been registered in various age groups; however, there is no suggestion that this proportion would vary by day.
To conclude, though in this study we could not directly study the association of RSPM with mortality, a significant association was observed between visibility, a marker for poor air quality, and mortality in northern India. There is a need for the monitoring of air quality daily at all sites in various Indian cities to study the association between air quality and mortality. The certification of causes of deaths also needs to be improved so that the effect of air pollution on cause-specific mortality can also be studied.
| Acknowledgment|| |
We are especially grateful to Dr. M. L. Garg , Dr. G. P. I. Singh, Dr. H. K. Parwana, and Dr. S. K. Jindal for their valuable technical support and Mr. Sanjay Kumar for assistance in data analysis. We are thankful to Punjab Pollution Control Board, Patiala; Birth and Death Registration Office, Ludhiana Municipal Corporation; and Meteorological Department, New Delhi, for sharing their data and Health Effects Institute, Boston, USA, for providing financial support.
| References|| |
|1.||Thurston GD. A critical review of PM10-mortality time-series studies. J Expo Anal Environ Epidemiol 1996;6:3-21. |
|2.||Goldberg MS, Burnett RT, Steib D. A review of time series studies used to evaluate the short term effects of air pollution on human health. Rev Environ Health 2003;18:269-303. |
|3.||Pope CA 3 rd . What do epidemiologic findings tell us about health effects of environmental aerosols? J Aerosol Med 2000;13:335-54. |
|4.||Zemp E, Elsasser S, Schindler C, Künzli N, Perruchoud AP, Domenighetti G, et al. Long-term ambient air pollution and respiratory symptoms in adults (SAPSLDIA study). Am J Respir Crit Care Med 1999;159:1257-66. |
|5.||Wong CM, Hu ZG, Lam TH, Hedley AJ, Peters J. Effects of ambient air pollution and environmental tobacco smoke on respiratory health of non-smoking women in Hong Kong. Int J Epidemiol 1999;28:859-64. |
|6.||Chhabra SK, Chhabra P, Rajpal S, Gupta SK. Ambient air pollution and chronic respiratory morbidity in Delhi. Arch Environ Health 2001;56:58-64. |
|7.||Detels R, Tashkin DP, Sayre JW, Rokaw SN, Massey FJ Jr, Coulson AH, et al. The UCLA population studies of COPD: X. A cohort study changes in respiratory function associated with chronic exposure to SOx, NOx, and hydrocarbons. Am J Public Health 1991;81:350-9. |
|8.||Dockery DW, Pope CA 3rd, Xu X, Spengler JD, Ware JH, Fay ME, et al. An association between air pollution and mortality in six US cities. N Engl J Med 1993;24:1753-9. |
|9.||Burnett RT, Brook J, Dann T, Delocla C, Philips O, Cakmak S, et al. Association between particulate matter and gas phase components of urban air pollution and daily mortality in eight Canadian counties. Inhal Toxicol 2000;12:15-39. |
|10.||Katasouyanni K, Schwartz J, Spix C. Short term effects of air pollution on health: An European approach using epidemiological time series data: The APHEA protocol. J Epidemol Community Health 1996;50:212-8. |
|11.||Katsouyanni K, Touloumi G, Samoli E, Gryparis A, Le Tertre A, Monopolis Y, et al. Confounding and effect modifications in short term effects of ambient particles on total mortality: Results from 29 European countries within APHEA 2 project. Epidemiology 2001;12:521-3. |
|12.||Katsouyanni K, Touloumi G, Spix C, Schwartz J, Balducci F, Medina S, et al. Short term effects of ambient Sulphur dioxide and particulate matter on mortality in 12 European cities: Results from time series data from APHEA project. Air Pollution and Health An European Approach. BMJ 1997;314:1658-63. |
|13.||Samet JM, Dominici F, Curriero FC, Cousac J, Zeger SL. Fine particulate air pollution and mortality in 20 US cities 1987 - 1994. N Engl J Med. 2000;343:1242-9. |
|14.||Health Effects Institute. Health effects of outdoor air pollution in developing countries of Asia: A literature review. Boston: Health Effects Institute; 2004. |
|15.||Air quality in seven major cities during 2001 and 2002. Central Pollution Control Board, New Delhi. Available from: http://www.envfor.nic.in/cpcb/ Accessed on 5 December 2003. |
|16.||Kumar R, Sharma M, Srivastva A, Thakur JS, Jindal SK, Parwana HK. Association of Outdoor Air Pollution with Chronic Respiratory Morbidity in an Industrial Town in Northern India. Archives of Environmental Health 2004;59:471-7. |
|17.||Indian Standard methods for measurement of air pollution (Nitrogen oxides). Bureau of Indian Standard, 5182 (6) Manak Bhavan, 9 Bahadur Shah Zafar Marg, New Delhi, 1998. |
|18.||Indian Standard methods for measurement of air pollution (Sulphur dioxide). Bureau of Indian Standard, IS 5182 (2) Manak Bhavan, 9 Bahadur Shah Zafar Marg, New Delhi, 2001. |
|19.||Indian standard methods for measurement of air pollution (Suspended Particulate Matter). Bureau of Indian Standard, 5182 (4) Manak Bhavan, 9 Bahadur Shah Zafar Marg, New Delhi, 1999. |
|20.||Chen G, Song G, Jiang L, Zhang Y, Zhao N, Chen B, Kan H. Short-term effects of ambient gaseous pollutants and particulate matter on daily mortality in Shanghai, China. J Occup Health 2008;50:41-7. |
|21.||Thach TQ, Chung RY, Hedley AJ, Chan E, Chau P, Wong CM. Effects of Air pollution Measured by Visibility in Hong Kong. Available from: http://www.cleanairnet.org/baq2006/1757/docs/SP5_3.ppt [last accessed on 2009 Aug 31]. |
|22.||Ostro B. Fine particulate air pollution and mortality in two Southern California Counties. Environ Res 1995;70:98-104. |
|23.||Vajanapoom N, Shy CM, Neas LM, Loomis D. Associations of particulate matter and daily mortality in Bangkok, Thailand. Southeast Asian J Trop Med Public Health 2002;33:389-99. |
|24.||Huang W, Tan J, Kan H, Zhao N, Song W, Song G, et al. Visibility, air quality and daily mortality in Shanghai, China. The Science of the Total Environment 2009;407:3295-300. |
|25.||Samoli E, Peng R, Ramsay T, Pipikou M, Touloumi G, Dominici F, et al. Acute effects of ambient particulate matter on mortality in Europe and North America: Results from the APHENA study. Environ Health Perspect 2008;116:1480-6. |
|26.||Wong CM, Vichit-Vadakan N, Kan H, Qian Z. Public Health and Air Pollution in Asia (PAPA): A multicity study of short-term effects of air pollution on mortality. Environ Health Perspect 2008;116:1195-20. |
[Table 1], [Table 2], [Table 3], [Table 4]
|This article has been cited by|
||Increased cardiovascular risk due to systemic inflammatory changes and enhanced oxidative stress in urban Indian women
| ||Dutta, A. and Ray, M.R. |
| ||Air Quality, Atmosphere and Health. 2013; 6(2): 501-508 |
||Applied Mixed Generalized Additive Model to Assess the Effect of Temperature on the Incidence of Bacillary Dysentery and Its Forecast
| ||Ma, W. and Sun, X. and Song, Y. and Tao, F. and Feng, W. and He, Y. and Zhao, N. and Yuan, Z. |
| ||PLoS ONE. 2013; 8(4) |
||Mortality trend and pattern in tertiary care hospital of Solapur in Maharashtra
| ||Godale, L. and Mulaje, S. |
| ||Indian Journal of Community Medicine. 2013; 38(1): 49-52 |
||A spatially disaggregated time-series analysis of the short-term effects of particulate matter exposure on mortality in Chennai, India
| ||Balakrishnan, K. and Ganguli, B. and Ghosh, S. and Sambandam, S. and Roy, S.S. and Chatterjee, A. |
| ||Air Quality, Atmosphere and Health. 2013; 6(1): 111-121 |