Air Qual Atmos Health (2013) 6:167–179 DOI 10.1007/s11869-011-0158-z
Air pollution and admissions for acute lower respiratory infections in young children of Ho Chi Minh City Sumi Mehta & Long H. Ngo & Do Van Dzung & Aaron Cohen & T. Q. Thach & Vu Xuan Dan & Nguyen Dinh Tuan & Le Truong Giang
Received: 8 April 2011 / Accepted: 16 August 2011 / Published online: 2 September 2011 # Springer Science+Business Media B.V. 2011
Abstract This study assessed the effects of exposure to air pollution on hospitalization for acute lower respiratory infection (ALRI) among children under 5 years of age in Ho Chi Minh City (HCMC) from 2003 to 2005. Casecrossover analyses with time-stratified selection of control periods were conducted using daily admissions for pneumonia and bronchiolitis and daily, citywide averages of Electronic supplementary material The online version of this article (doi:10.1007/s11869-011-0158-z) contains supplementary material, which is available to authorized users. S. Mehta (*) : A. Cohen Health Effects Institute, 101 Federal Street, Suite 500, Boston, MA 02110, USA e-mail:
[email protected] L. H. Ngo Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, MA, USA D. Van Dzung HCMC University of Medicine & Pharmacy, Ho Chi Minh City, Vietnam
PM10, NO2, SO2, and O3 (8-h maximum average) estimated from the local air quality monitoring network. Increased concentrations of NO2 and SO2 were associated with increased admissions in the dry season (November to April), with excess risks of 8.50% (95%CI 0.80–16.79) and 5.85% (95%CI 0.44–11.55), respectively. PM10 could also be associated with increased admissions in the dry season, but high correlation between PM10 and NO2 (0.78) limits our ability to distinguish between PM10 and NO2 effects. In the rainy season (May–October), negative associations between pollutants and admissions were observed. Results of this first study of the health effects of air pollution in HCMC support the presence of an association between combustion-source pollution and increased ALRI admissions. ALRI admissions were generally positively associated with ambient levels of PM10, NO2, and SO2 during the dry season, but not the rainy season. Negative results in the rainy season could be driven by residual confounding present from May to October. Preliminary exploratory analyses suggested that seasonal differences in the prevalence of viral causes of ALRI could be driving the observed differences in effects by season.
T. Q. Thach University of Hong Kong, Pokfulam, Hong Kong
Keywords Air pollution . ALRI . Children’s health . Vietnam
V. X. Dan HCMC Center for Occupational and Environmental Health, Ho Chi Minh City, Vietnam
Abbreviation ALRI CI CH1 CH2 HCMC HEPA ICD-10
N. D. Tuan HCMC Environmental Protection Agency, Ho Chi Minh City, Vietnam L. T. Giang HCMC Department of Health, Ho Chi Minh City, Vietnam
Acute lower respiratory infection Confidence interval Children’s Hospital Number 1 Children’s Hospital Number 2 Ho Chi Minh City HCMC Environmental Protection Agency International Classification of Diseases, 10th revision
168
IMCI NO2 O3 OR PM10 PM2.5 RSV SD SO2 WHO
Air Qual Atmos Health (2013) 6:167–179
WHO/UNICEF Integrated Management of Childhood Illness Nitrogen dioxide Ozone Odds ratio Particulate matter ≤10 μm in aerodynamic diameter Particulate matter ≤2.5 μm in aerodynamic diameter Respiratory syncytial virus Standard deviation Sulfur dioxide World Health Organization
Introduction Background and significance There is a growing body of epidemiological evidence that exposure to particles generated by emissions from diverse sources result in significant adverse health effects in urban populations (Cohen et al. 2004). Children that live close to heavily trafficked roads experience greater adverse respiratory episodes than children that live further away (Kim et al. 2008). In addition, a number of toxicological studies have shown that exposure to particles from traffic emissions results in inflammatory responses in vitro and in vivo studies (HEI Panel on the Health Effects of Traffic-Related Air Pollution 2009). In Asia, however, the composition of the emitted particles differs considerably from North America and Europe where the majority of these studies have been performed. Vehicle fleets in Asia are dominated by two- and three-wheeled vehicles and automobile and truck fleets are significantly older (Han and Naeher 2006). In addition, there are a number of local sources, which contribute significantly to exposures, but are not present in North America and Europe. A significant fraction of houses and roadside stalls rely on solid fuels for cooking and heating, trash is frequently burned in the street, and a large fraction of the population continues to use tobacco products in the home (HEI International Scientific Oversight Committee 2004). With extensive numbers of the world’s population living in highly polluted areas of Asia’s cities, increased effects of air pollution on the health of these populations would have significant public health impact and highly relevant policy implications. Although a recent systematic review revealed 421 studies on the health effects of air pollution in Asia (Health Effects Institute 2006), to date, no studies have been conducted in many of the poorer Southeast Asian countries, such as Laos, Cambodia, and Vietnam. The
ability to conduct such studies is currently compromised by the relative lack of reliable and easily accessible data on health outcomes, routinely collected air quality data, and collaboration between health and environment sectors. Air pollution and acute lower respiratory infection The capacity for combustion-derived air pollution to affect resistance to infection is well documented (Thomas and Zelikoff 1999). More recent studies suggest a role for fine particles (PM2.5) (Zelikoff et al. 2003). Effects on airway resistance, epithelial permeability, and macrophage function have been associated with various components of the complex mixture of air pollution generated by indoor and outdoor sources. There has also been considerable recent interest in the role of particle-associated transition metals, including iron, in producing oxidative stress in the lung (Ghio 2004; Ghio and Cohen 2005), which has been hypothesized to be a common factor in a range of adverse effects of air pollution on the cardiovascular and respiratory systems (Kelly 2003). PM-associated transition metals have also been associated with altered host defenses in rats (Zelikoff et al. 2002). Acute lower respiratory infections, including pneumonia, bronchitis, and bronchiolitis, are the largest single cause of mortality among young children worldwide and thus account for a significant global burden of disease worldwide (Williams et al. 2002; World Health Organization 2004). According to recent estimates, these infections cause nearly one fifth of mortality in children under the age of 5 years, with 90% of acute lower respiratory infection (ALRI) deaths being directly attributable to pneumonia (World Health Organization 2004). A substantial fraction of the burden is experienced by populations in Asia and Africa; the annual incidence of lower respiratory infections is 134 million in Asia and 131 million in Africa out of an overall global annual total of 429.2 million cases for all ages. For example, and of relevance to this investigation, more than 33,000 ALRI deaths occur in Vietnam each year (World Health Organization 2004). While outdoor air pollution has been associated with increased ALRI morbidity and mortality, very few studies have been conducted in developing countries of Asia, where populations are exposed to much higher levels of air pollution, and experience the greatest burden of disease due to ALRI. Of the 42 studies reviewed by Smith et al. (2000) and Romieu et al. (2002), only three were conducted in developing countries of Asia, although the highest exposures and the greatest burden of disease due to indoor and outdoor air pollution is borne by the populations in the region (Cohen et al. 2004; HEI International Scientific Oversight Committee 2004; Smith et al. 2004). As such, the results of this study have the potential to make an important
Air Qual Atmos Health (2013) 6:167–179
contribution to the growing literature on the health effects of air pollution in Asia. Location HCMC, formerly known as Saigon, is a major industrial and commercial center of Vietnam, and home to over six million people. Rapid economic development continues to bring more migrants to the city, contributing to the traffic congestion and urban crowding. Major sources of air pollution in HCMC include transport, energy, and industry. It should be noted that the vast majority of transportation demand is met by motorbikes and/or motorcycles (56%) (Viet Nam Register 2002). Other key local sources of air pollution with the potential to result in large proportions of exposure include smoking, the use of solid fuels, such as wood and coal, for cooking (particularly at roadside food stalls), and incense burning. The city has hot and humid weather year round, with mean temperatures averaging between 23°C and 32°C (73.4° to 89.6°F). There are two seasons in HCMC—a dry season, from November to April, and a rainy season, from May to October. With daily average PM10 levels routinely ranging from 30 to over 150 μg/m3, HCMC provides a unique opportunity to evaluate the health effects of short-term changes in air pollution in a tropical climate and across a wide range of the exposure response curve.
169
Methods This study was approved by the institutional review board of the Biological and Medical Ethical Committee of HCMC Department of Health (Decision no: 2751/ SYT-NVY). Patient population We focused the study inference on the children of HCMC under 5 years of age. HCMC’s two pediatric hospitals, Children’s Hospital Number 1 (CH1) and Children’s Hospital Number 2 (CH2), cover nearly all pediatric admissions in the city. CH1, located in District 10, is a 900-bed hospital and had 1,071,756 outpatient visits and 48,854 admissions in 2004. Located in District 1, the 800bed CH2 had 587,718 outpatient visits and 54,629 hospital admissions in 2004. Diseases of the respiratory system are a leading cause of inpatient admission in both hospitals. ALRI is diagnosed on the basis of WHO’s Integrated Management of Childhood Illness (IMCI) criteria. Other than severity of illness, we have not been able to identify any other major factors that could affect the likelihood of a child’s admission. &
Specific aim Using routinely collected data on air quality and hospital admissions from January 1, 2003 to December 31, 2005, assess whether increased short-term (on the order of days) exposures to air pollution are associated with increased frequency of hospitalizations for acute lower respiratory infections among children under 5 years of age. &
&
&
Admissions for ALRI, specifically pneumonia and bronchiolitis, were extracted from computerized records of the two Children’s Hospitals of HCMC. Nearly all children admitted for respiratory illnesses in HCMC are hospitalized in one of the two pediatric hospitals. Thus, we captured nearly all children’s admissions for respiratory illness in HCMC. Daily, city-level exposure estimates of particulate matter with diameter less than 10 μm (PM10), O3, NO2, and SO2 were generated using data from the HCMC Environmental Protection Agency’s (HEPA) ambient air-quality monitoring network. Daily meteorological information including temperature, relative humidity, and rainfall were collected from KTTV NB, the Southern regional hydro-meteorological center.
& &
Hospital capacity: the hospitals adhere to the WHO/ UNICEF IMCI guidelines for the management of ALRI when determining the need for admission. Despite increasing numbers of beds over time, both hospitals operate at or beyond capacity on a regular basis. Clinicians have reassured the investigative team that admissions for ALRI are not impacted by bed capacity; when necessary, multiple patients share the same hospital bed. Family financial status: as ability to pay is determined after admission, family financial status does not influence admission. Location of residence: while a child who lives in a remote province (outside HCMC) may occasionally be admitted using less stringent criteria, due to the hardship of traveling back and forth, since we focus only on HCMC residents, this will not affect our study.
Respiratory outcome data Ideally, we would have liked to collect information on ALRI incidence. As only daily aggregate information on outpatient visits is available, however, we focused our analysis on hospital admissions. Using incidence of hospital admissions not only allowed us to identify ALRI cases severe enough to warrant hospitalization but also limited our ability to precisely ascertain the time when clinically relevant disease onset occurred.
170
Criteria for ALRI diagnosis and admissions The IMCI was launched in 1995 by WHO and UNICEF, in order to address five leading causes of childhood deaths in the world: pneumonia, diarrhea, measles, malaria, and malnutrition. The Initiative has three main components: improvements in case-management skills of health staff; improvements in health systems; and improvements of family and community practices (WHO-CAH 1997). All patients seen at the pediatric hospitals are diagnosed and admitted using standardized IMCI criteria for diagnosing acute respiratory illness. Both hospitals use ICD-10 codes for disease reporting and classification. Admissions data collection and management As nearly all children admitted for respiratory illnesses in HCMC are hospitalized in one of the two pediatric hospitals, we were able to capture nearly all children’s admissions for respiratory illness in HCMC. Local collaborators at the children’s hospitals informed us that they did not use objective clinical criteria for distinguishing between pneumonia and bronchiolitis. Thus, we created a single outcome category for ALRI, which includes both pneumonia and bronchiolitis. ALRI admissions in children 5 years of age and under from January 2003 to December 2005 were extracted from computerized records of Children’s Hospitals 1 and 2 (CH1 and CH2) using the following criteria:
Air Qual Atmos Health (2013) 6:167–179
A quality assurance unit conducted manual cross checks to guarantee the quality of electronic data in the hospital databases. We chose not to include a control disease in our analysis, as we were not convinced that we would be able to select an ideal “control” disease. Moreover, hospitals are operating beyond capacity; while we have been assured that this does not affect admission for lower respiratory illness, we are unclear how limited capacity may impact admission for other conditions. Definition of case period Since we focused on hospital admissions, we used an empirical induction time (Rothman and Greenland 1998), i. e. the definition of the case period took induction times for ALRI, as well as time between onset of illness and detection of disease/time of hospitalization into consideration. The induction time would likely be on the order of a few days. On the basis of information from primary health clinics, hard copy clinical records, and physicians at the Children’s hospitals, we assumed that the case period should be between 1 and 6 days. This has implications for the choice of pollutant lag times; while we explored single day lags from lag 0 (pollution on same day of admission) to lag 10 (pollution 10 days prior to admission), we focus on the average lag (1–6) days, which incorporates pollution levels 1 to 6 days prior to admission. Environmental data
1. Admission date from January 01, 2003 to December 31, 2005 2. Age at admission date less than 5 years 3. Residence of HCMC on admission date: patients residing in the five rural districts of HCMC were excluded, as their exposures are not well reflected by the air quality monitoring network 4. Discharge diagnosis includes primary diagnosis of ICD-10: J13 to J18, or J21 5. Neonatal admissions (<28 days) were excluded, since these are likely to be influenced by perinatal conditions. 6. Consistent with other studies of ALRI in young children (http://ehs.sph.berkeley.edu/guat/page.asp? id=32), we excluded all repeated visits occurring within the same 14-day period to avoid double counting of the same case. This was only possible for CH1, due to limitations in the electronic dataset for CH2. As only a handful of cases were removed in CH1, however, this is unlikely to have major implications for our study results. 7. Aside from this 14-day window restriction, all multiple visits for children within the study period were retrieved.
Air quality data The HEPA, with the technical assistance from the Norwegian Institute for Air Research (NILU), has maintained nine automatic air-quality monitoring stations to monitor PM10, O3, NO2, and SO2 around the city since 2001. Data from the four background and/or residential stations were considered potentially eligible for inclusion in our analyses. Daily average values were created for each monitoring station by taking the mean of 24 hourly values for PM10, NO2, SO2, and by generating maximum 8 h moving averages for O3. A 75% completeness criterion was applied to all hourly data. No additional constraints to the data, i.e. no thresholds, were applied. Time-series plots and inter-station correlations were carefully assessed to inform decisions about the quality and completeness of station-specific average daily pollutant concentrations and to assess spatial homogeneity. Hourly values for each monitoring site were manually reviewed to flag recurrent values. All strings of four or more repeated values, indicating a problem with the monitoring system, were deleted. Site-specific daily time-series were also
Air Qual Atmos Health (2013) 6:167–179
reviewed to identify other potential data quality concerns. Citywide estimates for each pollutant were created by taking the mean of daily average data from each eligible station.
171
season due to the observed difference in the distribution of the pollutants and/or meteorology between the dry and rainy seasons, and because ALRI incidence is known to depend on season as well.
Meteorological data Results Mean daily data on rainfall, temperature, and humidity were calculated from hourly data provided by KTTV NB, the Southern regional hydro-meteorological center, and have very few missing values. Statistical methods Analyses were conducted using case-crossover, timestratified design. While the majority of earlier casecrossover design-based studies have used a symmetric bidirectional design, recent publications emphasize the advantages of using a time-stratified design (D'Ippoliti et al. 2003; Janes et al. 2005a; Levy et al. 2001; Lumley and Levy 2000; Mittleman 2005). Specifically, the timestratified design has been shown to produce unbiased estimates, as it enables the removal of overlap bias resulting from time trends in the pollutant data. (Janes et al. 2005b). Control days were every 7th days from the induction date within the same month as admission. Children hospitalized on the same day share the same case and control periods. Data were analyzed using conditional logistic regression where each subject has once case period and a variable (depending on when during the month the admission occurred) number of control periods. This is equivalent to a 1 M case control study. We assume that each child is hospitalized only once or that the repeated visits occurred more than 14 days apart and can be treated as independent ALRI episodes. As the time between onset of illness and hospital admission was thought to range from 1 to 6 days, it was not possible to specify a priori a single day lag. On the basis of “typical” referral patterns and pathways to hospital admission, the relevant window of exposure was thought to be within the week prior to hospital admission, i.e. 1 to 6 days before the date of admission. We assessed results for single day lags from lag 0 to lag 10, but emphasized results which used the average 1- to 6-day lag, since this best reflects the case period. These results take pollution levels in the 1 to 6 days leading up to admission into account. All results were calculated using lag 0 for temperature; the sensitivity of results to this assumption was explored in sensitivity analyses. Before estimating the effect of PM10 controlling for the other gaseous pollutants, we examined the correlation among PM10 and the other pollutants, since collinearity could introduce an estimation problem for getting the adjusted estimate of PM10. All analyses were stratified by
Hospital admissions Table 1 summarizes the characteristics of admitted patients. There were a total of 15,717 admissions with 10,468 of these admissions occurring at CH2, nearly twice the number of admissions for CH1 (5,249). CH2 consistently has more than twice the number of admissions than CH1. Admissions tended to peak between July and August each year, during the rainy season in HCMC (Fig. 1). Over 64% of admissions were male children, and around 74% of children were under 2 years of age. ICD classification at discharge was relatively consistent across hospitals; around 58% of the ALRI admissions were discharged with the diagnosis of “pneumonia” and around 42% were discharged with a diagnosis of “bronchiolitis.” Environmental data Table 2 and Fig. 2a–d show the annual distribution of citywide pollutant concentrations across all monitoring sites for the duration of the study. The seasonal trend in the data, corresponding to the rainy and dry seasons in HCMC is evident, with pollutant levels at their highest in the dry season and lowest in the rainy season. Concentrations of all pollutants, particularly PM10, show greater variability in the dry season. With the exception of the correlation between PM10 and NO2 in the dry season, none of the correlations exceeded 0.7 (Table 3). During the dry season, inter-pollutant correlations were highest for PM10 and NO2 (r=0.78) and PM10 and O3 (r=0.66), and lowest for PM10 and SO2 (0.32). During the rainy season, inter-pollutant correlations were lower, with highest correlations observed between PM10 and O3 (r=0.60), and PM10 and SO2 (0.36), but limited correlations observed among other pollutants. Meteorological data The average daily temperature in HCMC ranges from 23°C to 32°C, and the average daily relative humidity is consistently high, ranging from 51% to 93% (Table 2). The seasonal definition used is consistent with the rainfall experienced during the study period, i.e. the observed seasonal variation corresponds to the dry and rainy seasons.
172
Air Qual Atmos Health (2013) 6:167–179
Table 1 Selected characteristics of ALRI admissions, 2003–2005
Children’s hospital 1 Sex Male Female Age 0 to 1 years of age 1 to 2 years of age 2 to 5 years of age Month January February March April May June July August September October November December Season Rainy (May–October) Dry (November–April) Year 2003 2004 2005
Children’s hospital 2
Combined
71.7% 28.3%
3761 1488
61.4% 38.6%
6424 4044
64.8% 35.2%
10185 5532
38.4% 41.0% 20.7%
2015 2150 1084
29.6% 40.5% 29.8%
3101 4243 3124
32.6% 40.7% 26.8%
5116 6393 4208
6.3% 0.0% 7.3% 7.0% 7.9% 10.3%
332 265 385 370 417 541
6.4% 5.0% 6.2% 6.3% 8.0% 10.2%
667 520 648 664 835 1064
6.4% 5.0% 6.6% 6.6% 8.0% 10.2%
999 785 1033 1034 1252 1605
11.5% 12.5% 9.1% 9.0% 6.9% 7.1%
602 654 476 474 360 373
12.4% 11.2% 8.7% 9.3% 8.5% 7.9%
1296 1173 914 975 887 825
12.1% 11.6% 8.8% 9.2% 7.9% 7.6%
1898 1827 1390 1449 1247 1198
60.3% 39.7%
3164 2085
59.8% 40.2%
6257 4211
59.9% 40.1%
9421 6269
35.0% 29.9% 35.2%
1836 1567 1846
34.9% 29.5% 35.6%
3654 3086 3728
34.9% 29.6% 35.5%
5490 4653 5574
Statistical analyses All excess relative risk estimates and confidence intervals are reported per 10 μg/m3 increase in pollutant concentrations. Main results Large seasonal differences in admission patterns and pollution levels were observed. Sixty percent of ALRI admissions occurred during the rainy season, while the highest pollutant concentrations were observed in the dry season. We initially adjusted for season to control for seasonal differences in admissions and pollution levels and checked for seasonal interaction in the singlepollutant Poisson Regression models. The significance of the seasonal interaction term for NO2 (p <0.0008) provided us with further indication that we should conduct stratified analyses to reduce the potential for confounding by season. We also conducted sensitivity analyses to assess the adequacy of our binary classification of season.
Overall and season-specific results for single-pollutant and two-pollutant models are summarized in Tables 3 and 4 and Fig. 3. Results differed markedly when analyses were stratified by (rather than simply adjusted for) season. ALRI admissions were generally positively associated with ambient levels of PM10, NO2, and SO2 during the dry season, but not the rainy season. Dry season results Exposure to PM10 was weakly associated with a 1.25% (95% CI −0.55–3.09) excess risk of ALRI admissions for every 10 μg/m3 increase in exposure. This risk remained similar after adjusting for SO2 and O3. This association was no longer observed after adjusting for NO2. Associations between exposure to O3 and ALRI admissions were not observed in the single pollutant models. O3 exposure also remained unassociated with any change in risk after adjustment for other pollutants in the two-pollutant models. There was strong evidence of an NO2 effect, with excess risk estimates ranging from 7% to 18% for every 10 μg/m3
Air Qual Atmos Health (2013) 6:167–179
173
Fig. 1 Daily ALRI admissions to HCMC children’s hospitals, 2003–2005
increase in exposure in the single-pollutant models (lag 1 to lag 10). The highest increase in risk of ALRI admissions was associated with exposure to NO2, with an excess risk of 8.50% (95% CI 0.80–16.79) for every 10 μg/m3 increase in exposure observed in the single-pollutant model. These effects were robust to adjustment for other pollutants, and became even more pronounced after adjustment for PM10. There was limited evidence of a SO2 effect, with excess risk ranging from 2% to 6% for every 10 μg/m3 increase in exposure. SO2 was associated with an increased risk of ALRI admissions in the single-pollutant model, with an excess risk of 5.85 (95% CI 0.44–11.55) for every 10 μg/ m3 increase in exposure. This association was relatively robust to adjustment for PM10 and O3, but the two-pollutant adjusted for NO2 indicated confounding of effects by NO2.
The magnitude of the effects observed for NO2 and SO2, along with the wide confidence intervals, can be partially explained by the fact that excess risks were calculated for every 10 μg/m3 increase in pollutant concentrations, which is close to the standard deviation for these pollutants during the study period. Rainy season results Negative associations between PM10 and ALRI admissions were observed in the rainy season. No association with O3 exposure was observed in the single pollutant models, but O3 was negatively associated with ALRI admissions in the two-pollutant models. There was little evidence of an association between NO2 and ALRI admissions in the
Table 2 Distribution of mean daily air pollutants and meteorologic variables (citywide estimate), 2003–2005 Overall
PM10 (μg/m3) O3 (μg/m3) NO2 (μg/m3) SO2 (μg/m3) Temperature (Celsius) Relative Humidity (%) Rainfall (cm)
Dry season
Rainy season
N
Mean
SD
Min
Max
N
Mean
SD
Min
Max
N
Mean
SD
Min
Max
1040 1057 1022 720 1096 1096 1096
73.19 75.03 22.1 21.58 28.19 73.71 0.48
29 30 7.7 11 1.41 7.49 1.23
19 17 5 2.7 23.10 51.10 0.00
196 185 55 80 32.00 93.40 11.44
511 516 498 416 544 544 544
83.64 91.84 23.06 26.37 28.1 69.3 1.1
31 27 8.1 11 1.5 6.3 5.2
32.16 22.83 8.4 5.95 23.1 51 0
196 185 50.6 80.4 31.7 88 64.4
529 541 524 304 551 552 552
63.1 59 21.2 15 28.3 78 8.5
22.61 23.44 7.14 7.59 1.3 5.9 15.7
19 17 5 2.7 24.7 60 0
185.4 143.2 55.17 37.45 32 93.4 114.4
174
Air Qual Atmos Health (2013) 6:167–179
Fig. 2 a–d Daily average pollutant concentrations (μg/m3), citywide estimate, 2003–2005
Sensitivity analyses
rainy season. The single pollutant estimate suggested a negative association between NO2 and ALRI admissions, but this effect was no longer apparent after adjustment for other pollutants. Associations between SO2 and ALRI admissions were not observed in the rainy season. Table 3 Interpollutant correlations, by season, 2003–2005
Sensitivity analyses were conducted to examine the potential impact of individual monitoring stations, temperature lag choice, exploration of seasonal effects, and
Dry season
PM10 (μg/m3) O3 (μg/m3) NO2 (μg/m3) SO2 (μg/m3)
Rainy season
PM10
O3
NO2
SO2
PM10
O3
NO2
SO2
1.00 0.66 0.78 0.32
0.66 1.00 0.44 0.19
0.78 0.44 1.00 0.29
0.32 0.19 0.29 1.00
1.00 0.60 0.18 0.36
0.60 1.00 0.17 0.65
0.18 0.17 1.00 0.01
0.36 0.65 0.01 1.00
Air Qual Atmos Health (2013) 6:167–179
175
Table 4 Excess risk (ER%) per 10 μg/m3 increase in pollutant concentrations, overall and by season, average lag (1–6) days, single, and bipollutant models Overall
PM10 (μg/m3)
O3 (μg/m3)
NO2 (μg/m3)
SO2 (μg/m3)
Dry
Rainy
ER%
95% CI lo
95% CI hi
ER%
95% CI lo
95% CI hi
ER%
95% CI lo
95% CI hi
Single Adj SO2 Adj O3 Adj NO2 Single Adj SO2 Adj PM10 Adj NO2 Single Adj SO2
−1.10 −0.57 −0.19 −1.20 −1.96 −1.18 −1.98 −2.07 −1.08 3.40
−2.31 −2.08 −1.60 −2.60 −3.25 −2.75 −3.48 −3.46 −5.14 −2.39
0.12 0.95 1.25 0.22 −0.64 0.42 −0.45 −0.67 3.17 9.53
1.25 1.88 2.03 −0.36 −0.79 −0.63 −1.78 −1.28 8.50 12.07
−0.55 −0.15 −0.01 −3.02 −2.67 −2.78 −3.87 −3.27 0.80 2.76
3.09 3.95 4.11 2.37 1.13 1.56 0.36 0.74 16.79 22.22
−3.11 −3.62 −2.18 −2.90 −2.98 −1.01 −1.96 −2.91 −5.15 −2.71
−4.76 −5.90 −4.14 −4.67 −4.78 −3.51 −4.13 −4.85 −9.94 −9.98
−1.42 −1.28 −0.19 −1.10 −1.14 1.56 0.25 −0.92 −0.10 5.16
Adj PM10 Adj O3 Single Adj PM10 Adj O3 Adj NO2
0.95 1.11 2.61 2.77 2.95 1.84
−3.81 −3.30 −1.49 −1.35 −1.18 −2.31
5.94 5.72 6.87 7.06 7.25 6.16
9.70 10.12 5.85 5.44 5.72 3.70
−1.80 1.93 0.44 0.01 0.31 −1.76
22.55 18.97 11.55 11.15 11.43 9.47
−2.42 −2.48 −2.13 −0.81 −1.16 −1.78
−7.61 −7.73 −8.25 −7.05 −7.76 −7.99
3.07 3.06 4.41 5.86 5.92 4.85
monitor-specific effects on observed results. Using different temperature lags (0 versus average lag 1–6), including rainfall as a continuous variable, and seasonal reclassification of selected time periods had little impact on results.
Discussion Main results Exposure to air pollutants was generally positively associated with air pollution during the dry season (November– April) and inversely associated with pollution in the rainy season (May–October). Results suggest that increased concentrations of NO2, SO2, and PM10 are associated with increased hospital admissions for ALRI in young children of HCMC in the dry season, although SO2 and NO2 display the most robust relationships. As will be discussed later in more detail, PM10 could also be associated with increased hospital admissions in the dry season, but high correlation between PM10 and NO2 (0.78) limits our ability to distinguish between PM10 and NO2 effects. We know of no reason to think that exposure to air pollution could reduce the risk of ALRI in the rainy season, and infer that these results could be driven by residual confounding or other bias present within the rainy season. Although we could not specifically identify these sources of bias, factors influencing the results in the rainy season could
potentially impact results observed in the dry season as well. The prevalence of respiratory illness is higher during the rainy season, when there are likely to be other risk factors which play a stronger role than pollution levels. Pollutant levels are at their lowest during the rainy season. This situation increases the potential for negative confounding within the rainy season. The results appeared robust to definition of season, inclusion of rainfall as a continuous variable, and other potential sources of error. No clear evidence of monitorspecific effects was observed; differences across monitoring stations had widely overlapping confidence intervals. Comparison with other studies Variation in disease classifications, averaging times, and seasonal definitions limits the ability to make direct comparisons with studies conducted elsewhere. Studies that have found the strongest PM effects have focused specifically on bronchiolitis (Karr et al. 2006; Segala et al. 2008), while most of the studies, which have used a more broadly grouped disease classification, have found similarly inconclusive results for PM10 (Barnett et al. 2005; Gouveia and Fletcher 2000; Hernandez-Cadena et al. 2007). Recent publications that focus on the effects of sub-chronic exposures to air pollution have used averaging times on the order of weeks or months and seem to suggest stronger effects (Karr et al. 2009). Similar to other studies focused on the effects of acute effects; however, this study used averaging times on the order of days.
176
15
20
a
PM10
O3
N2O
adj NO2
adj O3
adj PM10
single
adj O3
adj PM10
adj SO2
single
adj NO2
adj PM10
adj SO2
single
adj NO2
adj O3
adj SO2
-10
single
-5
0
5
10
Fig. 3 a–c Excess risk (ER%) per 10 μg/m3 increase in pollutant concentrations, overall and by season, average lag (1–6) days, single and bipollutant results
Air Qual Atmos Health (2013) 6:167–179
SO2
PM10
O3
N2O
adj NO2
adj O3
adj PM10
single
adj O3
adj PM10
adj SO2
single
adj NO2
adj PM10
adj SO2
single
adj NO2
adj O3
adj SO2
-10
single
-5
0
5
10
15
20
b
SO2
PM10
Although other studies have looked for potential effect modification by season, the definition of season used has varied from study to study. Seasonal differences in effects are location-dependent, in that different seasonal patterns translate into different trends in temperature, precipitation, and disease incidence. Studies
N2O
adj NO2
adj O3
adj PM10
single
adj O3
adj PM10
adj SO2
single
adj NO2
3 O3
adj PM10
adj SO2
single
adj NO2
adj O3
adj SO2
single
-10
-5
0
5
10
15
20
c
SO2
conducted in North America and Western Europe have found effects in the “winter” season, where “winter” corresponds to the season where both pollution levels and disease incidence are likely to be high (Karr et al. 2006; Segala et al. 2008). A study conducted in Australia and New Zealand (Barnett et al. 2005), with different
Air Qual Atmos Health (2013) 6:167–179
seasonal patterns of temperature and disease found stronger effects in the warm season. To the best of our knowledge, this is the first study to focus specifically on assessing differences by wet and dry seasons. HCMC has a tropical climate with little variation in temperature but distinct seasonal patterns with regard to rainfall that are correlated with distributions of respiratory infections. Thus, while the seasonal definition used here may not be directly comparable with other studies, it is appropriate for this particular study. The sensitivity of PM10 results in the rainy season to the inclusion of data on respiratory syncytial virus (RSV) prevalence was explored using methods developed to assess whether unmeasured confounders in observational studies could cause a bias large enough to reverse estimates of effect (Lin et al. 1998). Preliminary results suggest that seasonal differences in the prevalence of viral causes of ALRI could be driving the observed differences in effects by season. For more details, see the Online Reference. If ALRI of bacterial etiology is more strongly associated with air pollution exposures than ALRI caused by viruses such as RSV, it is possible that our models, which did not take RSV into account, may have induced a spurious negative association between ALRI and air pollution. In addition, with virtually no RSV incidence in the dry season, these findings also lend credibility to the notion that RSV could influence results primarily in the rainy season. Differentiating PM10 vs. NO2 effects While effects in this study appear to be driven by exposure to NO2, the high correlation between PM10 and NO2 (0.78) limits our ability to clearly distinguish PM10 and NO2 effects. Indeed, other studies focused on the association between air pollution and ALRI in young children have noted the challenges of adequately distinguishing between PM10 and NO2 effects (Gouveia and Fletcher 2000; Barnett et al. 2005; Braga et al. 2001). One study (Saldiva et al. 1994) that found a strong positive association between NOx and child respiratory mortality, while PM10 effects were not observed. PM10 is a complex mixture of components that, like NO2, serves as an imperfect indicator of combustion-related pollutants, but also represents non-combustion/crustal sources of pollution that are prominent in HCMC, such as construction. The much lower correlation between NO2 and PM10 during the rainy season provides further evidence that these indicator pollutants may not be accurately characterizing exposures to air pollution from combustion processes in the rainy season. PM2.5 data, which would serve as a better indicator of combustion processes, are not routinely available in HCMC, and differences in PM composition by season also remain unknown. Nevertheless, taken as a whole, results suggest that increased risks of ALRI
177
admissions in young children are associated with increases in combustion-related pollution (including, but not exclusive to traffic pollution). A recent systematic review of the literature found that each 10 μg/m3 increase in long term ambient PM2.5 concentrations is associated with around a 12% increased risk of ALRI incidence, with the results of short-term studies reviewed indicating that the health effects of air pollution continue to be observed at higher concentrations, and across a range of geographic locations (Mehta et al. 2011). In addition, exposure to air pollution from indoor combustion of solid fuels has also been consistently associated with increased incidence and mortality risk in 14 studies in developing countries (Smith et al. 2004), and children’s exposure to second-hand smoke, defined as having one or both parents who smoke indoors, has also been associated with increased incidence of ALRI infections and hospital admissions (U.S. Department of Health and Human Services 2006). Thus, while the results of this study would be difficult to interpret in the absence of other evidence, given the strong effect modification by season observed, results support the presence of an association between combustion-source pollution and increased ALRI admissions in HCMC which is consistent with studies conducted elsewhere. This study is, to the best of our knowledge, the first study of ALRI admissions in young children to be conducted in an Asian city, and the first study of the health effects of air pollution to be conducted in Ho Chi Minh City, Vietnam. Ambient pollution levels in HCMC are certainly much higher than those experienced in the existing air pollution and ALRI morbidity literature, but remain somewhat lower than levels in other Asian megacities. The results of this study may inform global health impact assessments at the mid-range of the exposureresponse curve. In addition, the study contributes to the growing literature on the health effects of air pollution in Asia, particularly given the lack of studies in Southeast Asia (HEI International Scientific Oversight Committee 2010). Acknowledgements This Technical Assistance was supported with funds from the Health Effects Institute’s Public Health and Air Pollution in Asia (PAPA) Program, the Poverty Reduction Cooperation Fund of the Asian Development Bank, (Technical Assistance 4714-VIE) and in-kind support from the Government of Viet Nam. The Working Group is grateful to the Clean Air Initiative for Asian Cities (CAI-Asia), for initiating communications between HEI, ADB, and the Government of Viet Nam, the local steering committee for the project, and the (PAPA) program’s International Scientific Oversight Committee for providing technical guidance and suggestions throughout the process, especially Drs. Michael Brauer, Ross Anderson, Kirk Smith, and Frank Speizer. We are grateful for useful comments provided by HEI’s Review Committee and external quality assurance consultant David Bush, and for administrative assistance provided by Tiffany North and Morgan Younkin.
178
References Barnett AG, Williams GM, Schwartz J, Neller AH, Best TL, Petroeschevsky AL, Simpson RW (2005) Air pollution and child respiratory health: a case-crossover study in Australia and New Zealand. Am J Respir Crit Care Med 171(11):1272–1278 Braga AL, Saldiva PH, Pereira LA, Menezes JJ, Conceicao GM, Lin CA, Zanobetti A, Schwartz J, Dockery DW (2001) Health effects of air pollution exposure on children and adolescents in Sao Paulo, Brazil. Pediatr Pulmonol 31(2):106–113 Cohen A, Anderson HR, Ostro B, Pandey KD, Krzyzanowski M, Kuenzli N, Gutschmidt K, Pope CA, Romieu I, Samet JM, Smith KR (2004) Mortality impacts of urban air pollution. In: Ezzati MLA, Rodgers A, Murray CJL (eds) Comparative quantification of health risks: global and regional burden of disease due to selected major risk factors vol 2. World Health Organization, Geneva D’Ippoliti D, Forastiere F, Ancona C, Agabiti N, Fusco D, Michelozzi P, Perucci CA (2003) Air pollution and myocardial infarction in Rome—a case-crossover analysis. Epidemiology 14(5):528–535 Ghio AJ (2004) Biological effects of Utah Valley ambient air particles in humans: a review. J Aerosol Med Depos Clear Effects Lung 17 (2):157–164 Ghio AJ, Cohen MD (2005) Disruption of iron homeostasis as a mechanism of biologic effect by ambient air pollution particles. Inhal Toxicol 17(13):709–716 Gouveia N, Fletcher T (2000) Respiratory diseases in children and outdoor air pollution in Sao Paulo, Brazil: a time series analysis. Occup Environ Med 57(7):477–483 Han X, Naeher LP (2006) A review of traffic-related air pollution exposure assessment studies in the developing world. Environ Int 32(1):106–120 Health Effects Institute (2006) PAPA-SAN Database. www.healtheffects. org HEI International Scientific Oversight Committee (2004) Health effects of outdoor air pollution in developing countries of Asia: a literature review. Health Effects Institute, Boston HEI International Scientific Oversight Committee (2010) Outdoor air pollution and health in the developing countries of Asia: a comprehensive review. Special report 18. Health Effects Institute, Boston HEI Panel on the Health Effects of Traffic-Related Air Pollution (2009) Traffic-related air pollution: a critical review of the literature on emissions, exposure, and health effects. HEI special report 17. Health Effects Institute, Boston Hernandez-Cadena L, Barraza-Villarreal A, Ramirez-Aguilar M, Moreno-Macias H, Miller P, Carbajal-Arroyo LA, Romieu I (2007) Infant morbidity caused by respiratory diseases and its relation with the air pollution in Juarez City, Chihuahua, Mexico. Salud Publica Mex 49(1):27–36 Janes H, Sheppard L, Lumley T (2005a) Case-crossover analyses of air pollution exposure data: referent selection strategies and their implications for bias. Epidemiology 16(6):717–726 Janes H, Sheppard L, Lumley T (2005b) Overlap bias in the casecrossover design, with application to air pollution exposures. Stat Med 24(2):285–300 Karr C, Lumley T, Shepherd K, Davis R, Larson T, Ritz B, Kaufman J (2006) A case-crossover study of wintertime ambient air pollution and infant bronchiolitis. Environ Heal Perspect 114(2):277–281 Karr CJ, Rudra CB, Miller KA, Gould TR, Larson T, Sathyanarayana S, Koenig JQ (2009) Infant exposure to fine particulate matter and traffic and risk of hospitalization for RSV bronchiolitis in a region with lower ambient air pollution. Environ Res 109 (3):321–327
Air Qual Atmos Health (2013) 6:167–179 Kelly FJ (2003) Oxidative stress: its role in air pollution and adverse health effects. Occup Environ Med 60(8):612–616 Kim JJ, Huen K, Adams S, Smorodinsky S, Hoats A, Malig B, Lipsett M, Ostro B (2008) Residential traffic and children’s respiratory health. Environ Health Perspect 116(9):1274–1279 Levy D, Lumley T, Sheppard L, Kaufman J, Checkoway H (2001) Referent selection in case-crossover analyses of acute health effects of air pollution. Epidemiology 12(2):186–192 Lin DY, Psaty BM, Kronmal RA (1998) Assessing the sensitivity of regression results to unmeasured confounders in observational studies. Biometrics 54(3):948–963 Lumley T, Levy D (2000) Bias in the case-crossover design: implications for studies of air pollution. Environmetrics 11(6):689–704 Mehta S, Shin H, Burnett R, North T, Cohen AJ (2011) Ambient particulate air pollution and acute lower respiratory infections: a systematic review and implications for estimating the global burden of disease. Air Quality Atmos Health. doi:10.1007/ s11869-011-0146-3 Mittleman MA (2005) Optimal referent selection strategies in casecrossover studies—a settled issue. Epidemiology 16(6):715–716 Romieu I, Samet JM, Smith KR, Bruce N (2002) Outdoor air pollution and acute respiratory infections among children in developing countries. J Occup Environ Med 44(7):640–649 Rothman KJ, Greenland S (1998) Modern epidemiology. LippencottRaven, Philadelphia Saldiva PH, Lichtenfels AJ, Paiva PS, Barone IA, Martins MA, Massad E, Pereira JC, Xavier VP, Singer JM, Bohm GM (1994) Association between air pollution and mortality due to respiratory diseases in children in Sao Paulo, Brazil: a preliminary report. Environ Res 65(2):218–225 Segala C, Poizeau D, Mesbah M, Willems S, Maidenberg M (2008) Winter air pollution and infant bronchiolitis in Paris. Environ Res 106(1):96–100 Smith KR, Samet JM, Romieu I, Bruce N (2000) Indoor air pollution in developing countries and acute lower respiratory infections in children. Thorax 55(6):518–532 Smith K, Mehta S, Maeusezahl-Feuz M (2004) Indoor air pollution from household use of solid fuels. In: LA Ezzati M, Rogers A, Murray CJL (eds) Global and regional burden of disease attributable to selected major risk factors, vol 2. World Health Organization, Geneva Thomas P, Zelikoff J (1999) Air pollutants: modulators of pulmonary host resistance against infection. In: Holgate ST, Samet JM, Koren HS, Maynard RL (eds) Air pollution and health. Academic, San Diego U.S. Department of Health and Human Services (2006) The health consequences of involuntary exposure to tobacco smoke: a report of the Surgeon General. U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, Coordinating Center for Health Promotion, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health, Atlanta, GA Viet Nam Register (2002) Integrated action plan to reduce vehicle emissions in VietNam. ADB. http://www.adb.org/Vehicle-Emis sions/actionviet.asp WHO-CAH (1997) The management of childhood illness in developing countries: rationale for an integrated strategy. WHO, Department of Child and Adolescent Health and Development, Geneva Williams B, Gouws E, Boschi-Pinto C, Bryce J, Dye C (2002) Estimates of world-wide distribution of child deaths from acute respiratory infections. Lancet 2:25–32 World Health Organization (2004) The global burden of disease: 2004 update. WHO, Department of Health Statistics and Informatics in the Information, Evidence and Research Cluster Geneva, Switzerland
Air Qual Atmos Health (2013) 6:167–179 Zelikoff JT, Schermerhorn KR, Fang KJ, Cohen MD, Schlesinger RB (2002) A role for associated transition metals in the immunotoxicity of inhaled ambient particulate matter. Environ Heal Perspect 110:871–875
179 Zelikoff JT, Chen LC, Cohen MD, Fang KJ, Gordon T, Li Y, Nadziejko C, Schlesinger RB (2003) Effects of inhaled ambient particulate matter on pulmonary antimicrobial immune defense. Inhal Toxicol 15(2):131–150