Next Article in Journal / Special Issue
Correlating Remote Sensing Data with the Abundance of Pupae of the Dengue Virus Mosquito Vector, Aedes aegypti, in Central Mexico
Previous Article in Journal / Special Issue
Nexus of Health and Development: Modelling Crude Birth Rate and Maternal Mortality Ratio Using Nighttime Satellite Images

Volume 3, Issue 2

► Article Menu

Article Versions

Related Info

More by Authors

Export Article

ISPRS Int. J. Geo-Inf. 2014, 3(2), 713-731; doi:10.3390/ijgi3020713

Article
Canadian Forest Fires and the Effects of Long-Range Transboundary Air Pollution on Hospitalizations among the Elderly
1
, 2
, 3
, 2,4
, 2,†, 5
and 2,3,6,*
1
Department of Health Policy and Management, Johns Hopkins Bloomberg School of Public Health, 615 N. Wolfe Street, Baltimore, MD 21205, USA
2
Department of Environmental Health Sciences, Johns Hopkins Bloomberg School of Public Health, 615 N. Wolfe Street, Baltimore, MD 21205, USA
3
Department of Biostatistics, Johns Hopkins Bloomberg School of Public Health, 615 N. Wolfe Street, Baltimore, MD 21205, USA
4
ICF International, 9300 Lee Highway, Fairfax, VA 22031, USA
5
Yale School of Forestry & Environmental Studies, Yale University, 195 Prospect Street, New Haven, CT 06511, USA
6
Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, 615 N. Wolfe Street, Baltimore, MD 21205-2103, USA
†
Alison Geyh deceased on 20 February 2011.
*
Author to whom correspondence should be addressed.
Received: 3 March 2014; in revised form: 22 April 2014 / Accepted: 5 May 2014 / Published: 20 May 2014

Abstract

: In July 2002, lightning strikes ignited over 250 fires in Quebec, Canada, destroying over one million hectares of forest. The smoke plume generated from the fires had a major impact on air quality across the east coast of the U.S. Using data from the Medicare National Claims History File and the U.S. Environmental Protection Agency (EPA) National air pollution monitoring network, we evaluated the health impact of smoke exposure on 5.9 million elderly people (ages 65+) in the Medicare population in 81 counties in 11 northeastern and Mid-Atlantic States of the US. We estimated differences in the exposure to ambient PM2.5—airborne particulate matter with aerodynamic diameter of ≤2.5 µm—concentrations and hospitalizations for cardiovascular, pulmonary and injury outcomes, before and during the smoke episode. We found that there was an associated 49.6% (95% confidence interval (CI), 29.8, 72.3) and 64.9% (95% CI, 44.3–88.5) increase rate of hospitalization for respiratory and cardiovascular diagnoses, respectively, when the smoke plume was present compared to before the smoke plume had arrived. Our study suggests that rapid increases in PM2.5 concentrations resulting from wildfire smoke can impact the health of elderly populations thousands of kilometers removed from the fires.
Keywords:
air pollution; hospitalizations; PM2.5; forest fires; global climate change

1. Introduction

Forest fires are known to be a major source of air pollutants [1] on a local and a global scale [2,3,4,5,6]. Each year, combustion products from local and distant wildfires impact large populations worldwide [5,7,8,9,10,11,12,13]. The atmospheric pollutant that most consistently increases with biomass smoke from wildfires is suspended fine particulate matter (PM), which is commonly associated with increased mortality and morbidity [1,4,14,15,16,17]. The PM in biomass smoke consists mainly of black carbon (soot and charcoal particles), organic carbon, sulfates and/or nitrates, potassium carbonate and silica [12,18].

Short-term exposures to fine particulate matter, PM2.5 (airborne particulate matter with aerodynamic diameter of ≤2.5 µm) have been associated with increased mortality and morbidity in various communities around the world and in the United States [19,20,21,22,23,24,25,26,27,28]. Most studies addressing health impacts of short-term exposures are related to anthropogenically generated PM, which are commonly associated with automobile combustion and industrial practices. There has been a limited but growing body of literature addressing the impact of shorter-term exposure to smoke from forest and bush fires (referred to as wildfires in this paper) [8,9,11,17,29,30,31,32,33]. The majority of these studies have examined the impact on nearby local communities of exposures to wildfire aerosols. Associated health effects include increased emergency department visits and hospital admissions for chronic obstructive pulmonary disease (COPD), bronchitis, asthma and chest pain [7,13,15,34,35,36,37]. For example, the San Diego wildfire in October 2003 caused the daily 24 h average PM2.5 concentrations to exceed 150 µg/m3, and was associated with significant increases in hospital room emergency room visits for asthma, respiratory problems, eye irritation, and smoke inhalation [38], and increased eye and respiratory symptoms, medication use and physician visits in children living in the San Diego area [30]. In Canada, Moore et al. estimated that forest fire smoke in 2003 was associated with excess respiratory complaints in Kelowna (Kelowna, BC, Canada) area residents [31].

The wildfire aerosols have lifetimes on the order of many days [12], which allows transport over large distances [4,8,11,39]. While it is clear that local populations are affected by wildfire events, a growing concern is the potential health impact on geographically distant populations, specifically in susceptible groups such as the elderly. Epidemiologic research has identified the elderly, who are more likely to have pre-existing lung and heart diseases, as a population vulnerable to the effects of short-term exposures to air pollution including fine particles [24,40,41,42,43,44,45,46,47,48,49,50].

In July 2002, a dramatic increase in forest fire activity was registered in the province of Quebec, Canada [5]. Specifically, on 7 July 2002 at least 85 fires were burning out of control and destroyed approximately one million hectares of forest that month. The Canadian Forest Services indicated that the major causal factors contributing to these fires were a long period without precipitation and strong winds coming from the north. Lightning and dry conditions sparked fires on 2 July 2002 in two separate regions southeast of James Bay, which is between 200 and 400 miles north of the U.S. border.

The smoke plume generated from these forest fires had a major impact on air quality across the east coast of the United States during the first week in July 2002 [5,51]. The plume was carried by strong northerly winds from Quebec across the U.S. border covering a distance that extended from north of Montreal to northern Virginia and Maryland [5,51]. Satellite images show the plume on 7 July covering parts of New Hampshire, Vermont, and New York, and diffuse and patchy over the eastern seaboard down to Washington, DC (Figure 1: MODIS satellite image on 7 July 2002 [52]. An air quality study being conducted in Baltimore, MD at that time reported that the 24-hour PM2.5 concentrations reached 86 µg/m3 on 7 July, resulting in as much as a 30-fold increase in the daily ambient PM2.5 concentrations [5] during the peak period of haze over the city. On that day, the highest PM2.5 concentration (338 µg/m3) was reported in New Hampshire [51]. In many cities in the region (such as New York City and Philadelphia), health advisories were issued for residents with espiratory conditions [53].

Ijgi 03 00713 g001 1024
Figure 1. MODIS satellite image taken on 7 July 2002, 10:35 EDT. The red dots mark areas of high forest fire activity. The black dots represent the centroids of counties used in our analysis.
Figure 1. MODIS satellite image taken on 7 July 2002, 10:35 EDT. The red dots mark areas of high forest fire activity. The black dots represent the centroids of counties used in our analysis.
Ijgi 03 00713 g001 1024

This transboundary wildfire smoke episode offered a unique opportunity to evaluate the vulnerability of a susceptible population in the United States exposed to smoke generated from a large-scale wildfires more than a thousand kilometers away. In this paper, we assess the relationship between hospital admissions for individuals 65 years and older and PM2.5 concentrations during July 2002 for U.S. states impacted by the Canadian wildfire plume. Unlike previous research that focused on health impacts on populations living near or relatively close to wildfire events, this paper presents one of the first analyses of wildfire smoke health impacts to populations living at great distances, particularly within the U.S., from fires and not previously identified to be at significant risk.

2. Methods

2.1. Study Area

The study population includes 5.9 million Medicare enrollees aged 65 or older, residing in 81 U.S. counties in the northeastern and mid-Atlantic regions of the U.S. (11 states) and in Illinois. Medicare data included daily health care hospitalization information allowing us to look at changes in hospitalization rates associated with changes in PM air pollution. The study was restricted to 81 plume affected counties that could provide PM2.5 data at least once every five days during our study period. These counties are identified in Figure 1. The state of Illinois was chosen as a reference area not affected by the plume for descriptive comparison to the forest fire induced levels of PM2.5.

2.2. Data Sources

Daily counts of hospital admissions for June–July 2002 were obtained from billing claims for the Medicare National Claims History Files. Each billing claim contains the date of admission, age, sex, place of residence, and cause of hospitalization. Coding for hospitalization is based on the ninth revision of the International Classification of Diseases (ICD-9) [54]. The daily county-specific hospitalization rates were obtained by dividing the daily county-specific admission counts by the daily county-specific number of individuals at risk defined as the number of individuals enrolled in Medicare. The number of individuals at risk, the total Medicare population, varied each day. The daily counts of each health event within each county were obtained by summing the number of hospital admissions for each of the diseases considering both primary and secondary diagnoses. This study was exempt from the Johns Hopkins Bloomberg School of Public Health’s Institutional Review Board because the data for this study did not involve individual identifiers.

The PM2.5 air pollution monitoring data were obtained from the U.S Environmental Protection Agency Air Quality System [55]. Temperature and dew point temperature data were gathered from the National Climatic Data Center (NCDC) on the Earth-Info CD database [56]. The analysis was restricted to the 11 northeastern states affected by the plume [53]. Although, carbon monoxide (CO) is a gaseous pollutant commonly associated with forest fires, we did not include it in our analysis because only a quarter of the counties of interest reported CO monitoring data.

2.3. Statistical Analysis

County average PM2.5 values for use in subsequent regression models were obtained using the geostatistical method known as block kriging. Kriging is a statistical method widely used in the environmental sciences that produces optimal spatial predictions and in the block kriging version produces statistically optimal aggregate level estimates based on point level (monitored) data [57,58]. To implement block kriging, a fine grid is placed over the study area and concentrations predicted at each grid cell using an ordinary kriging approach. Grid points within a specific county are then averaged to provide a single county measure. This allows the more sparsely point monitored PM2.5 measurements to be optimally estimated at an aggregate county level by spatially borrowing information from proximal PM2.5 measurements. To improve characterization of spatial dependence (a crucial step in the kriging process) and to address issues of edge effects in the block kriging procedure, air monitor values from all inclusive states in the affected region (Figure 1 and including Maine, Virginia, and West Virginia) were utilized. A similar approach was taken for block kriging county estimates in the chosen unaffected reference area of Illinois. Kriging was performed with the “gstat” package in the R Statistical Software [59]. The NCDC weather monitoring network is adequately dense so that county level temperature and dew point temperature estimates were obtained from averaging measurements observed at county specific monitoring stations.

In our primary analysis, we considered specific cardiovascular and respiratory outcomes that have been associated in the literature with short-term exposure to PM2.5. This selection would also allow for comparability with previous studies. Outcomes considered were grouped in three broad categories: cardiovascular outcomes (ICD-9 codes (390–459)), respiratory outcomes (460–519) and injuries (800–849) (selected as our control outcome).

In our secondary analysis we targeted specific cardiovascular and respiratory outcomes, that represent endpoints more often associated with acute impacts of short-term exposures to PM2.5 in the general population [26,60]: hypertensive disease (401–405), myocardial infarction (410), ischemic heart disease (410–414), acute pulmonary heart disease (415), acute heart disease (410–425), heart rhythm disturbances (426–427), heart failure (428), stroke (430–438), peripheral vascular disease (440–448), and asthma (493), chronic obstructive pulmonary disease (COPD) (490–492, 496), respiratory tract infections (RTI) (464–466, 480–487) and acute respiratory tract infections (460–466) (Table 1).

Satellite imagery and back-trajectory modeling showed that the smoke plume covering the northeastern states on 6–8 July 2002, a Saturday, Sunday, and Monday [5] originated from Province of Quebec. Thus we defined 6–8 July 2002 as our haze period, when air pollution was also at its highest and compared it to a control non-haze period, corresponding to the same days of the preceding week (29, 30 June and 1 July 2002). We chose not to include the same days in the following week as part of the control because of potential persisting effects in the post-haze period [19] The same days of the week were used in this fashion to minimize potential time-varying confounding effects, such as might be expected for weekday versus weekend. Within our period of interest, we observed a total of 5772 for all respiratory hospital admissions, and 18,316 hospital admissions for all cardiovascular hospital admissions.

Table
Table 1. Diagnosis Categories.
Table 1. Diagnosis Categories.
Reason for Hospital AdmissionICD-9 Codes
Respiratory Outcomes(460–519)
Chronic Obstructive Pulmonary Disease (COPD)(490–492, 496)
Asthma(493)
Respiratory Tract Infections(464–466, 480–487)
Cardiovascular Outcomes(390–492, 496)
Cerebrovascular Disease(430–438)
Heart Failure(440–448)
Stroke(410)
Peripheral Vascular Disease(410–414)
Myocardial Infarction(426–427)
Ischemic Heart Disease(426–428)
Heart Rhythm Disturbances(46–427)
Heart Disease(426–428)
Other Heart Disease(420–425, 428)
Hypertension(401–405)
Injury(800–849)

To assess the impact of exposure to PM2.5 during the haze period on hospitalization rates for the outcomes of interest, we linked billing claims from Medicare with daily concentrations of PM2.5 by county of residence for the Medicare enrollees [60,61] and employed Poisson regression. Regression inference was based on a generalized estimating equation approach (GEE) [62] to account for the possibility of residual temporal autocorrelation in the county-specific hospitalization rates. We assumed an independent working correlation structure for the county-specific daily hospitalization rates, which assumes that the correlation between daily rates decreases with time.

We let Ijgi 03 00713 i001 and Ijgi 03 00713 i002 be the observed and expected daily number of cause-specific hospitalizations, Ijgi 03 00713 i003 be the number of people at risk on day t in county c, Ijgi 03 00713 i004 be the daily average PM2.5 for county c, Ijgi 03 00713 i005 and Ijgi 03 00713 i006 the daily average temperature and dew point temperature for county c. We fit the following log-linear regression model:

Ijgi 03 00713 i007
with Ijgi 03 00713 i008 an indicator variable representing haze versus non-haze period and Ijgi 03 00713 i009 representing the regression offset. The parameter β1 denotes the log relative risk of cause-specific hospitalization associated with one unit (µg/m3) increase in daily average PM2.5 (block kriged county estimates) adjusted for temperature and dew-point temperature. The parameter β2 denotes the increase in log relative rate of hospitalizations during the haze period compared to that during the non-haze period.

To account for potential delays in disease incidence after exposure, we also explored single lag models, where we substitute Ijgi 03 00713 i010 in the county-specific model (1) for Ijgi 03 00713 i011, where l = 0, 1, 2 days and Ijgi 03 00713 i012 is the PM2.5 concentration for county c on day t at a lag of l days [63]. A lag of 0 days corresponds to the association between PM2.5 concentrations on a given day and the risk of hospitalization on the same day. We also applied distributed lag models [64,65] to estimate the relative risk of hospitalization associated with cumulative exposure over the current day and the previous two days. The distributed lag models are obtained by substituting Ijgi 03 00713 i013 in the county-specific model (Equation (1)) for Ijgi 03 00713 i010. Models with a PM2.5 by haze period interaction Ijgi 03 00713 i014 were also considered allowing for the relative risk of hospitalizations associated with PM2.5 exposure to change during the haze period compared to the non-haze period.

3. Results

Smoke from wildfires in the Quebec region of Canada drifted over the Northeastern and Mid-Atlantic region of the United States on 6–8 July 2002 blanketing much of the area (Figure 1). Figure 2 shows the region wide average daily PM2.5 concentrations for the two months surrounding this event from both the affected area and unaffected reference area of Illinois, which are shown to have similar ambient PM2.5 concentrations. However, during the identified haze period we observe a substantial spike in PM2.5 only in the affected region. The daily averages shown Figure 2 were obtained using 10% trimmed mean to average across monitors after correcting for yearly averages for each monitor; an accepted approach for summarizing longer times series and larger geographic areas of air pollution data [55]. Missing concentrations were imputed using a natural spline interpolation method that accounts for the daily seasonality in PM2.5.

Ijgi 03 00713 g002 1024
Figure 2. Countywide daily mean PM2.5 (µg/m3) in the affected and unaffected regions from 1 June to 31 Jul 2002.
Figure 2. Countywide daily mean PM2.5 (µg/m3) in the affected and unaffected regions from 1 June to 31 Jul 2002.
Ijgi 03 00713 g002 1024

Figure 3 displays for the affected region the spatial distribution in county specific average PM2.5 concentrations (estimated via block kriging) as being substantially higher for days in the haze period compared to the same days in the preceding non-haze period. Table 2 presents summary statistics for these estimated county specific PM2.5 concentrations in the affected states and the chosen Illinois reference area during the haze and non-haze periods. The average countywide PM2.5 concentrations for 6–8 July 2002 were significantly higher (p-value < 0.001) in the haze period (mean 53.0 µg/m3, standard deviation SD = 25.0), compared to non-haze period (mean 21.5 µg/m3, SD = 10.3 for the affected states. No significant difference in average countywide PM2.5 was found in the unaffected reference area of Illinois between the same haze and non-haze periods.

Ijgi 03 00713 g003 1024
Figure 3. County level block kriged estimated mean PM2.5 (µg/m3) for the 81 affected counties in non-haze and haze periods. Top Row: Control period between 29 June–1 July designated as non-haze period. Bottom Row: Haze period between 6–8 July.
Figure 3. County level block kriged estimated mean PM2.5 (µg/m3) for the 81 affected counties in non-haze and haze periods. Top Row: Control period between 29 June–1 July designated as non-haze period. Bottom Row: Haze period between 6–8 July.
Ijgi 03 00713 g003 1024
Table
Table 2. Summary statistics for the block kriged estimated county average PM2.5 (µg/m3) for the plume affected area and Illinois reference area both stratified by the haze and non-haze periods. Data were pooled over the three days comprising both the haze and non-haze periods.
Table 2. Summary statistics for the block kriged estimated county average PM2.5 (µg/m3) for the plume affected area and Illinois reference area both stratified by the haze and non-haze periods. Data were pooled over the three days comprising both the haze and non-haze periods.
Summary StatisticsAffected RegionUnaffected Region (IL)
HazeNon-HazeHazeNon-Haze
Min17.67.012.215.2
25th %tile35.613.716.019.0
Median43.122.620.720.7
Mean53.021.522.420.8
75th %tile69.125.329.722.1
Max127.743.733.028.4
SD25.010.37.25.0

Regression results for the parameter of interest, β2 from model (1) representing the increase in log relative rate of hospitalization comparing the haze to non-haze period, are presented in Table 3. Results listed in Table 3 are for (exp(β2) − 1) × 100% representing the percent change in admissions for the three primary outcome categories, all respiratory, all cardiovascular, and the selected control outcome injury. Compared to the non-haze period, this Medicare population had a 49.55% (95% confidence interval (CI): 29.82–72.29) significantly increased rate for respiratory related hospitalizations and a 64.93% (CI: 44.30–88.51) significantly increased rate for cardiovascular related hospitalizations during the haze period compared to the non-haze period, adjusting for weather and PM2.5 on the same day (lag 0 model). For the chosen control outcome there was no significant increase in the rate of injury related hospitalizations between the haze and non-haze periods. Single and distributed lag model results show similar significant increases in respiratory and cardiovascular related hospitalizations although not as high as the lag 0 models.

Table
Table 3. Percent change in hospital admissions for the haze period compared to the non-haze period in the affected region controlling for PM2.5, temperature, and dew point. Bolded model lag types denote significant change in hospital admissions at the 0.05 level.
Table 3. Percent change in hospital admissions for the haze period compared to the non-haze period in the affected region controlling for PM2.5, temperature, and dew point. Bolded model lag types denote significant change in hospital admissions at the 0.05 level.
Hospitalization CodesPM2.5 Model LagPercent Change95% Confidence Interval
All RespiratoryLag 049.55(29.82, 72.29)
Lag 121.14(2.69, 42.90)
Lag 223.36(10.20, 38.10)
Dlag130.20(7.66, 57.46)
Dlag26.48(−14.63, 32.81)
All CardiovascularLag 064.93(44.30, 88.51)
Lag 136.06(18.43, 56.32)
Lag 249.28(34.39, 65.81)
Dlag133.85(12.45, 59.33)
Dlag28.31(−13.89, 36.24)
InjuryLag 03.04(−21.31, 34.92)
Lag 110.97(−14.04, 43.27)
Lag 213.23(−5.64, 35.89)
Dlag10.37(−25.69, 35.55)
Dlag2−6.04(−30.50, 27.02)

The effect of the forest fire seemed to have a slightly greater impact on cardiovascular than respiratory admissions, a finding contrary to other studies [20,21], which found a higher impact on respiratory admissions than cardiovascular admissions. This may be because of over classification with the use of both primary and secondary discharge codes. However, an increase in cardiovascular and respiratory hospitalizations is consistent with the literature [8,16,24,50,66], though some literature has shown no increase in cardiovascular hospitalizations [2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21] or mortality [67].

Figure 4 displays the percent increase in hospital admissions for the haze period compared to the non-haze period in the affected region for all specific diagnoses of interest with single and distributed lag models. Percent increases in hospitalizations were significantly higher for same day models in all diagnosis groups except for respiratory tract infections, cerebrovascular disease, stroke, and myocardial infarction. For one-day lag models, COPD, heart rhythm disturbances, other heart disease, and hypertension were all significant. The largest change in hospitalization was observed on same day lag models; the magnitude of the effect decreases with increasing inclusion of lag effects. Some of the diagnostic codes, such as asthma, have a low prevalence in hospitalizations and may be more difficult to detect a change in rates of hospitalizations.

Ijgi 03 00713 g004 1024
Figure 4. Percent change and 95% CI in hospital related admissions for the haze period compared to the non-haze period in the affected region controlling for PM2.5, temperature, and dew point.
Figure 4. Percent change and 95% CI in hospital related admissions for the haze period compared to the non-haze period in the affected region controlling for PM2.5, temperature, and dew point.
Ijgi 03 00713 g004 1024

Note: Percent increase in admissions for the haze period compared to the non-haze period.

Models with the additional PM2.5 by haze period interaction term did not reveal consistent or significant results for this interaction effect across many of the outcomes considered (results not shown) suggesting the PM2.5 effect not to be the statistically different during the haze and non-haze periods. Results did however continue to reveal the strong significant increase in hospitalizations for the haze period compared to the non-haze period. The lack of evidence supporting a change in PM2.5 associated relative risk of hospitalizations during the haze period compared to the non-haze period (PM2.5 by haze period interaction) could be due to a combination of several factors including; PM2.5 from wildfire sources not any more toxic than non-wildfire sources, the fact that the haze period happened to include both a weekday and the weekend possibly confounding exposure or that there are other drivers of hospitalizations during the haze period that is not entirely explained by PM2.5 [68]. Although other studies have shown that wildfire PM is at least as toxic as urban PM [66,68]. Sensitivity analysis was performed increasing the haze period to five, seven, and nine days surrounding 7 July 2002 as well as lagging the haze period when considering models with a lagged PM2.5 exposure. Results differed quantitatively, however the overall qualitative interpretations remained consistent. As such all reported and interpreted regressions results were based on model (1) with the predefined three-day haze period.

4. Discussion

With the Medicare National Claims History, National Climatic Data Center’s weather data, and EPA’s National Monitoring Network, we conducted an opportunistic study of the effects of wildfire air pollution from the Province of Quebec on the health of the elderly population stretching between New York and the District of Columbia.

The selection of our outcome categories was informed by several considerations. Acute exposure to PM may exacerbate existing pulmonary disease [69,70,71,72]. Because COPD is a substantial risk factor for cardiovascular mortality and morbidity [73,74,75,76], air pollution exposure may also contribute to cardiovascular risk through exacerbation of COPD symptoms.

We considered several specific cardiovascular and respiratory outcomes that are impacted by inflammatory processes, such as myocardial infarction, stroke, and asthma. The short term effects of exposure to high levels of air pollution are likely to cause inflammatory responses in the lung and release of cytokines with local and systemic consequences [24,77]. Acute effects of PM exposure have also been shown to increase plasma viscosity [25,78].

The log-linear model we used to estimate associations between day-to-day variations in PM2.5 (at various lags) and day-to-day variations in the county-level hospitalization rates is typical of time-series analysis [79]. The advantage of the time-series approach is that confounding by individual-level covariates, such as smoking, is not an issue. However, factors that vary with daily pollution exposure, such as weather and co-pollutants, are likely to be confounders in these studies. Time-series analyses typically include nonlinear terms for weather and season [60,63]. One advantage of examining the effects of abrupt increases in PM2.5 concentration over a short period of time is that it is unlikely that our analyses were confounded by any seasonal pattern unaccounted for in the model.

Our study focused on a population of interest, the elderly, using a nationally available database of health claims. In interpreting the findings of our analysis, considerations need to be given to the inherent limitations of the data analyzed. Information in the Medicare database is prone to bias due to inaccuracy of claims coding for specific diagnoses [80,81,82,83]. In an attempt to reduce misclassification for outcomes of interest, we used primary and secondary diagnosis codes to identify records for inclusion.

In addition, the ambient air pollution data from administrative databases such as EPA’s National Monitoring Network, which have been created for regulatory purposes, only provide limited spatial and temporal coverage. This is an issue typical of air pollution studies relying on publicly available datasets. Also, during the short period when the plume affected the northeast U.S. (three peak days) the number of hospitalizations recorded was small compared to that observed in larger scale time-series studies. The small numbers of hospitalization counts and the limited exposure data reduces the power of our study.

In common with all studies examining the relationship between exposure to air pollution and health that depend on ambient air quality data, our study finding may be biased because of exposure misclassification. In our study all individuals were assumed to be exposed similarly to the corresponding ambient PM measured at EPA monitoring sites. However, some people may have listened to the health advisories (like the ones in New York and Pennsylvania, USA) and retreated indoors during the event. Although, as shown by Sapkota et al., some indoor environments were substantially impacted by the elevated ambient PM2.5 due to this event [5], the use air conditioning may also have ameliorated the indoor exposure [84].

Wildfires have rapid and substantial impacts on local air quality that elevate ambient PM concentrations well above the norm. The impact of this increased pollution on the health of local populations has been well documented [9,11,13,17,29,30,85,86,87]. For example, Duclos et al. showed a 40% and 30% increase in number of local emergency room visits for asthma and COPD respectively, during 1987 forest fires in northern California [13]. While composition of wildfire smoke has been shown to influence smoke related health outcomes [16], reliance on PM concentrations is common for regional studies because of the fire plume components it is most consistently elevated during smoke events, as opposed to other attributed components like carbon monoxide or nitric oxide [5,32].

Increases in wildfire activity have been linked to warmer spring and summer temperatures resulting from climate change and long-accumulated stocks of combustible vegetation [88,89]. In addition, climate models predict an increase in rising temperatures and regional drying will result in an increase in wildfire activity. Regions like the western U.S. can expect to see considerable increase in wildfire activity [90]. These wildfires may have significant impacts on future air quality and on the health of populations susceptible to the effects of air pollution even those living at a great distance. Our approach is replicable on a larger scale for estimating the health effects of large pollution events resulting from biomass burning over large distances.

5. Conclusions

This research adds to the growing body of literature demonstrating the significant impact of transboundary air pollution on public health. Short-term increases in PM2.5 concentrations due to the Canadian forest fires in the Province of Quebec show a consistent significant increase in respiratory and cardiovascular hospital admissions for the elderly across the east coast of the U.S. as far south as Washington D.C. These results highlight the public health implications of long-range transport of wildfire-related air pollutants, and raise awareness of the health costs that may be associated with exposures to pollutants generated at remote locations. Preparedness planning by public health and clinical communities is needed to ensure a rapid and adequate response to the challenges that a large wildfire event poses to the health care emergency system. This is especially important as these events are expected to increase as a consequence of global climate change. Knowledge of the potential health risks of wildfire smoke can be translated into local action and used in risk prevention decisions and risk communication such as: educating vulnerable groups of taking their prescribed medication, preparing emergency rooms for an influx of patients with respiratory and cardiac conditions, and advising affected populations on how best to limit their exposures to the smoke.

Acknowledgments

This work was in part supported by the NIEHS (grant numbers P50ES015903, P01ES018176, and P30ES03819) and funding from the U.S. Environmental Protection Agency (grant numbers RD83213901, RD83451001, and RD83241701). Although the research described in this article has been funded wholly or in part by the United States Environmental Protection Agency it has not been subjected to the Agency’s required peer and policy review and therefore does not necessarily reflect the views of the Agency and no official endorsement should be inferred. This manuscript is dedicated to Alison Geyh from the Department of Environmental Health Sciences, Johns Hopkins Bloomberg School of Public Health who passed away before this paper was completed. She was a great friend and outstanding colleague.

Author Contributions

Conception and design of the study: Patrick Breysse, Alison Geyh, Frank Curriero, and George Le. Analysis and interpretation of data: George Le, Jesse Berman, Frank Curriero.Collection and assembly of data: Patrick Breysse, Sorina Eftim, Jesse Berman, Aidan McDermott. Drafting of the article: George Le, Frank Curriero, Patrick Breysse, Jesse Berman, Sorina. Critical revision of the article for important intellectual content: Patrick Breysse, Frank Curriero, George Le. Final approval of the article: Frank Curriero, George Le, Patrick Breysse.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Ward, D. Health Guidelines for Vegetation Fire Events, Background Papers. In Smoke from Wildland Fires; WHO: Geneva, Switzerland, 1999; pp. 70–85. [Google Scholar]
  2. Miller, D.J. Assessing boreal forest fire smoke aerosol impacts on U.S. air quality: A case study using multiple data sets. J. Geophys. Res. 2011, 116, 1–19. [Google Scholar]
  3. Mathur, R. Estimating the impact of the 2004 Alaskan forest fires on episodic particulate matter pollution over the eastern Unites States through assimilation of satellite derived aerosol optical depths in a regional air quality model. J. Geophys. Res. 2008, 113, 1–14. [Google Scholar]
  4. Jalava, P.; Salonen, R.; Halinen, A.; Penttinen, P.; Pennanen, A.; Sillanpaa, M.; Sandell, E.; Hillamo, R.; Hirvonen, M.R. In vitro inflammatory and cytotoxic effects of size-segregated particulate samples collected during long-range transport of wildfire smoke to Helsinki. Toxicol. Appl. Pharmacol. 2006, 215, 341–353. [Google Scholar] [CrossRef]
  5. Sapkota, A.; Symons, J.; Kleissl, J.; Wang, L.; Parlange, M.; Ondov, J.; Breysse, P.N.; Diette, G.B.; Eggleston, P.A.; Buckley, T.J. Impact of the 2002 Canadian forest fires on particulate matter air quality in Baltimore city. Environ. Sci. Technol. 2005, 39, 24–32. [Google Scholar] [CrossRef]
  6. Morris, G.A.; Hershey, S.; Thompson, A.M.; Stohl, A.; Colarco, P.R.; McMillan, W.W.; Warner, J.; Johnson, B.J.; Witte, J.C.; Kucsera, T.L.; et al. Alaskan and Canadian forest fires exacerbate ozone pollution in Houston, Texas, on 19 and 20 July. J. Geophys. Res. 2004, 111, 1–10. [Google Scholar]
  7. Ovadnevaite, J.; Kvietkus, K.; Marsalka, A. 2002 summer fires in Lithuania: Impact on the Vilnius city air quality and the inhabitants health. Sci. Total Environ. 2006, 356, 11–21. [Google Scholar] [CrossRef]
  8. Mott, J.; Mannino, D.M.; Alverson, C.; Kiyu, A.; Hashim, J.; Lee, T.; Falter, K.; Redd, S.C. Cardiorespiratory hospitalizations associated with smoke exposure during the 1997, southeast Asian forest fires. Int. J. Hyg. Environ. Health 2005, 208, 75–85. [Google Scholar] [CrossRef]
  9. Heil, A.; Goldammer, J. Smoke-haze pollution: A review of the 1997 episode in southeast Asia. Reg. Environ. Chang. 2001, 2, 24–37. [Google Scholar] [CrossRef]
  10. Wotawa, G.; Trainer, M. The influence of Canadian forest fires on pollutant concentrations in the United States. Science 2000, 288, 324–328. [Google Scholar] [CrossRef]
  11. Health Guidelines for Vegetation Fire Events. In Technical Report; World Health Organization: Geneva, Switzerland, 1999.
  12. Andreae, M. Biomass burning in the tropics: Impact on environmental quality and global climate. Popul. Dev. Rev. 1990, 16, 268–291. [Google Scholar] [CrossRef]
  13. Duclos, P.; Sanderson, L.M.; Lipsett, M. The 1987 forest fire disaster in California: Assessment of emergency room visits. Arch. Environ. Health 1990, 45, 53–58. [Google Scholar] [CrossRef]
  14. Emmanuel, S.C. Impact to lung health of haze from forest fires: The Singapore experience. Respirology 2000, 5, 175–182. [Google Scholar] [CrossRef]
  15. Johnston, F.H.; Webby, R.J.; Pilotto, L.S.; Bailie, R.S.; Parry, D.L.; Halpin, S.J. Vegetation fires, particulate air pollution and asthma: A panel study in the Australian monsoon tropics. Int. J. Environ. Health Res. 2006, 16, 391–404. [Google Scholar] [CrossRef]
  16. Rappold, A.G.; Stone, S.L.; Cascio, W.E.; Neas, L.M.; Kilaru, V.J.; Carraway, M.S.; Szykman, J.J.; Ising, A.; Cleve, W.E.; Meredith, J.T.; et al. Peat bog wildfire smoke exposure in rural north Carolina is associated with cardiopulmonary emergency department visits assessed through syndromic surveillance. Environ. Health Perspect. 2011, 119, 1415–1420. [Google Scholar] [CrossRef]
  17. Sastry, N. Forest fires, air pollution and mortality in southeast Asia. Demography 2002, 39, 1–23. [Google Scholar] [CrossRef]
  18. Larson, T.V.; Koenig, J.Q. Wood smoke: Emissions and noncancer respiratory effects. Annu. Rev. Public Health 1994, 15, 133–156. [Google Scholar] [CrossRef]
  19. Delfino, R.J.; Brummel, S.; Wu, J.; Stern, H.; Ostro, B.; Lipsett, M.; Winer, A.; Street, D.H.; Zhang, L.; Tjoa, T.; et al. The relationship of respiratory and cardiovascular hospital admissions to the southern California wildfires of 2003. Occup. Environ. Med. 2009, 66, 189–197. [Google Scholar] [CrossRef]
  20. Henderson, S.B.; Brauer, M.; MacNab, Y.C.; Kennedy, S.M. Three measures of forest fire smoke exposure and their associations with respiratory and cardiovascular health outcomes in a population-based cohort. Environ. Health Perspect. 2011, 119, 1266–1271. [Google Scholar] [CrossRef]
  21. Johnston, F.H.; Bailie, R.S.; Pilotto, L.S.; Hanigan, I.C. Ambient biomass smoke and cardio-respiratory hospital admissions in Darwin, Australia. BMC Public Health 2007, 7. [Google Scholar] [CrossRef]
  22. Kunii, O.; Kanagawa, S.; Yajima, I.; Hisamatsu, Y.; Yamamura, S.; Amagai, T.; Ismail, I.T. The 1997 haze disaster in Indonesia: Its air quality and health effects. Arch. Environ. Health 2002, 174, 16–22. [Google Scholar]
  23. Ostro, B.; Broadwin, R.; Green, S.; Feng, W.; Lipsett, M. Fine particulate air pollution and mortality in nine California counties: Results from Calfine. Environ. Health Perspect. 2006, 114, 29–33. [Google Scholar]
  24. Pope, C.A., III; Dockery, D.W. Health effects of particulate air pollution: Lines that connect. J. Air Waste Manag. Assoc. 2006, 56, 709–742. [Google Scholar] [CrossRef]
  25. Rückerl, R.; Schneider, A.; Breitner, S.; Cyrys, J.; Peters, A. Health effects of particulate air pollution: A review of epidemiological evidence. Inhal. Toxicol. 2011, 23, 555–592. [Google Scholar] [CrossRef]
  26. Samet, J.M.; Domnici, F.; Curriero, F.; Coursac, I.; Zeger, S. Fine particulate air pollution and mortality in 20 U.S. cities: 1987–1994. N. Engl. J. Med. 2000, 342, 1742–1757. [Google Scholar] [CrossRef]
  27. Schwartz, J.; Dockery, D.W.; Neas, L. Is daily mortality associated specifically with fine particles? J. Air Waste Manag. Assoc. 1996, 46, 927–939. [Google Scholar] [CrossRef]
  28. Zanobetti, A.; Schwartz, J. Air pollution and emergency admissions in Boston, MA, USA. J. Epidemiol. Community Health 2006, 60, 890–896. [Google Scholar] [CrossRef]
  29. Aditama, T. Impact of haze from forest fire to respiratory health: Indonesian experience. Respirology 2000, 5, 169–174. [Google Scholar] [CrossRef]
  30. Künzli, N.; Avol, E.; Wu, J.; Gauderman, W.J.; Rappaport, E.; Millstein, J.; Bennion, J.; McConnell, R.; Gilliland, F.D.; Berhane, K.; et al. Health effects of the 2003 Southern California wildfire on children. Amer. J. Respir. Crit. Care Med. 2006, 174, 1221–1228. [Google Scholar] [CrossRef]
  31. Moore, D.; Copes, R.; Fisk, R.; Joy, R.; Chan, K.; Brauer, M. Population health effects of air quality changes due to forest fires in British Columbia in 2003: Estimates from physician-visit billing data. Can. J. Public Health 2006, 97, 105–108. [Google Scholar]
  32. Naeher, L.P.; Brauer, M.; Lipsett, M.; Zelikoff, J.T.; Simpson, C.D.; Koenig, J.Q.; Smith, K.R. Woodsmoke health effects: A review. Inhal. Toxicol. 2007, 19, 67–106. [Google Scholar]
  33. Vedal, S. Ambient particles and health: Lines that divide. J. Air Waste Manag. Assoc. 1997, 37, 551–581. [Google Scholar] [CrossRef]
  34. Mott, J.; Meyer, P.; Mannino, D.M.; Redd, S.; Smith, E.; Gotway-Crawford, C.; Chase, E. Wildland forest fire smoke: Health effects and intervention evaluation, Hoopa, California, 1999. West. J. Med. 2002, 176, 157–162. [Google Scholar] [CrossRef]
  35. Schreuder, A.; Larson, T.; Sheppard, L.; Claiborn, C. Ambient woodsmoke and associated respiratory emergency department visits in Spokane, Washington. Int. J. Occup. Environ. Health 2006, 12, 147–153. [Google Scholar] [CrossRef]
  36. Shusterman, D.; Kaplan, J.; Canabarro, C. Immediate health effects of an urban wildfire. West. J. Med. 1993, 158, 133–138. [Google Scholar]
  37. Sutherland, E.; Make, B.; Vedal, S.; Zhang, L.; Dutton, S.; Murphy, J.; Silkoff, P.E. Wildfire smoke and respiratory symptoms in patients with chronic obstructive pulmonary disease. J. Allergy Clin. Immunol. 2005, 115, 420–422. [Google Scholar] [CrossRef]
  38. Viswanathan, S.; Eria, L.; Diunugala, N.; Johnson, J.; McClean, C. An analysis of effects of San Diego wildfire on ambient air quality. J. Air Waste Manag. Assoc. 2006, 56, 56–67. [Google Scholar] [CrossRef]
  39. Report of the Biregional Workshop on the Health Impacts of Haze Related Air Pollution. In Technical Report RS/98/GE/17(MAA); Regional Office for the Western Pacific, World Health Organization: Manila, Philippines, 1998.
  40. American Lung Association. State of the Air: 2004. In Technical Report; American Lung Association: New York, NY, USA, 2004. [Google Scholar]
  41. Anderson, H.; Atkinson, R.; Peacock, J.; Marston, L.; Konstantinou, K. Meta-Analysis of Time-Series Studies and Panel Studies of Particulate Matter (PM) and Ozone (O3). In Technical Report; World Health Organization (WHO): Geneva, Switzerland, 2004. [Google Scholar]
  42. Annesi-Maesano, I.; Agabiti, N.; Pistelli, R.; Couilliot, M.; Forastiere, F. Subpopulations at increased risk of adverse health outcomes from air pollution. Eur. Respir. J.Suppl. 2003, 40, S57–S63. [Google Scholar]
  43. Barnett, A.; Williams, G.; Schwartz, J.; Best, T.; Neller, A.; Petroeschevsky, A.; Simpson, R. The effects of air pollution on hospitalizations for cardiovascular disease in elderly people in Australian and New Zealand cities. Environ. Health Perspect. 2006, 114, 1018–1023. [Google Scholar] [CrossRef]
  44. Boezen, H.; Vonk, J.; van der Zee, S.; Gerritsen, J.; Hoek, G.; Brunekreef, B.; Schouten, J.P.; Postma, D.S. Susceptibility to air pollution in elderly males and females. Eur. Respir. J. 2005, 25, 1018–1024. [Google Scholar] [CrossRef]
  45. Brook, R.; Franklin, B.; Cascio, W.; Hong, Y.; Howard, G.; Lipsett, M.; Luepker, R.; Mittleman, M.; Samet, J.; Smith, S.C., Jr.; et al. Air pollution and cardiovascular disease: A statement for healthcare professionals from the Expert Panel on Population and Prevention Science of the American Heart Association. Circulation 2004, 109, 2655–2671. [Google Scholar] [CrossRef]
  46. Chen, Y.; Yang, Q.; Krewski, D.; Burnett, R.; Shi, Y.; McGrail, K. The effect of coarse ambient particulate matter on first, second, and overall hospital admissions for respiratory disease among the elderly. Inhal. Toxicol. 2005, 17, 649–655. [Google Scholar] [CrossRef]
  47. Goldberg, M.S.; Burnett, R.; Bailar, J., III; Brook, J.; Bonvalot, Y.; Tamblyn, R.; Singh, R.; Valois, M. Identifying Subgroups of 19 the General Population that may be Susceptible to Short-Term Increases in Particulate Air Pollution: A Time-Series Study in Montreal, Quebec. In Research Report 097; Health Effects Institute: Cambridge, MA, USA, 2000. [Google Scholar]
  48. Pope, C.A., III. Epidemiology of fine particulate air pollution and human health: Biologic mechanisms and who’s at risk? Environ. Health Perspect. 2000, 108, 713–723. [Google Scholar] [CrossRef]
  49. Pope, C.A., III; Burnett, R.; Thun, M.; Calle, E.; Krewski, D.; Ito, K.; Thurston, G.D. Lung cancer, cardiopulmonary mortality and long-term exposure to particulate air pollution. JAMA 2002, 287, 1132–1141. [Google Scholar] [CrossRef]
  50. Villeneuve, P.; Chen, L.; Stieb, D.; Rowe, B. Associations between outdoor air pollution and emergency department visits for stroke in Edmonton, Canada. Eur. J. Epidemiol. 2006, 21, 689–700. [Google Scholar] [CrossRef]
  51. DeBell, L.J.; Talbot, R.; Dibb, J. A major regional air pollution event in the northeastern United States caused by extensive forest fires in Quebec, Canada. J. Geophys. Res. 2004, 109, D19305. [Google Scholar] [CrossRef]
  52. NASA: Visible Earth. Wildfires in Quebec and Smoke over United States East Coast. Available online: http://visibleearth.nasa.gov/view.php?id=60610 (accessed on 1 February 2011).
  53. GFMC (Global Fire Monitoring Center)/Fire Ecology Research Group. 2007 UN International Strategy for Disaster Reduction (ISDR). Available online: http://www.fire.uni-freiburg.de/ (accessed on 3 December 2010).
  54. International Classification of Diseases, Ninth Revision (ICD-9); World Health Organization: Geneva, Switzerland, 1977.
  55. U.S. Environmental Protection Agency. Air Quality System Data Mart (Internet Database). Available online: http://www.epa.gov/ttn/airs/aqsdatamart (accessed on 1 February 2011).
  56. National Climatic Data Center. NCDC Surface Airways, CD Database; EarthInfo Inc.: Boulder, CO, USA, 2011. [Google Scholar]
  57. Cressie, N.A.; Cassie, N.A. Statistics for Spatial Data; Wiley: New York, NY, USA, 1993. [Google Scholar]
  58. Young, L.J.; Gotway, C.A.; Yang, J.; Kearney, G.; DuClos, C. Linking health and environmental data in geographical analysis: It’s so much more than centroids. Spat. Spatiotemporal Epidemiol. 2009, 1, 73–84. [Google Scholar] [CrossRef]
  59. R Core Team. R: A Language and Environment for Statistical Computing, 2014. Available online: Available online: www.r-project.org (accessed on 1 February 2014).
  60. Dominici, F.; Peng, R.; Bell, M.; Pham, L.; McDermott, A.; Zeger, S.; Samet, J.M. Fine particulate air pollution and hospital admission for cardiovascular and respiratory diseases. JAMA 2006, 295, 1127–1134. [Google Scholar] [CrossRef]
  61. Eftim, S.E.; Janes, H.; McDermott, A.; Samet, J.; Dominici, F. A comparative analysis of the chronic effects of fine particulate matter. Epidemiology 2008, 19, 209–216. [Google Scholar] [CrossRef]
  62. Liang, K.-Y.; Zeger, S. Longitudinal data analysis using generalized linear models. Biometrika 1986, 73, 13–22. [Google Scholar] [CrossRef]
  63. Peng, D.; Dominici, F. Statistical Methods for Environmental Epidemiology with R; Springer Science, Business Media, LLC: New York, NY, USA, 2008. [Google Scholar]
  64. Schwartz, J. The distributed lag between air pollution and daily deaths. Epidemiology 2000, 1, 320–326. [Google Scholar] [CrossRef]
  65. Welty, L.; Zeger, S. Are the acute effects of particulate matter on mortality in the national morbidity, mortality, and air pollution study the result of inadequate control for weather and season? A sensitivity analysis using flexible distributed lag models. Amer. J. Epidemiol. 2005, 162, 80–88. [Google Scholar] [CrossRef]
  66. Morgan, G.; Sheppard, V.; Khalaj, B.; Ayyar, A.; Lincoln, D.; Jalaludin, B.; Beard, J.; Corbett, S.; Lumley, T. Effects of brushfire smoke on daily mortality and hospital admissions in Sidney, Australia. Epidemiology 2010, 21, 47–55. [Google Scholar] [CrossRef]
  67. Vedal, S.; Steven, J.D. Wildfire air pollution and daily mortality in a large urban area. Environ. Res. 2006, 102, 29–35. [Google Scholar] [CrossRef]
  68. Dennekamp, M.; Abramson, M.J. The effects of bushfire smoke on respiratory health. Respirology 2011, 16, 198–209. [Google Scholar] [CrossRef]
  69. Dockery, D.; Ware, J.; Ferris, B.J.; Speizer, F.; Cook, N.; Herman, S. Change in pulmonary function in children associated with air pollution episodes. J. Air Pollut. Control Assoc. 1982, 32, 937–942. [Google Scholar] [CrossRef]
  70. Roemer, W.; Hoek, G.; Brunekreef, B. Effect of ambient winter air pollution on respiratory health of children with chronic respiratory symptoms. Amer. Rev. Resp. Dis. 1993, 147, 118–124. [Google Scholar] [CrossRef]
  71. Hoek, G.; Brunekreef, B. Effect of photochemical air pollution on acute respiratory symptoms in children. Amer. J. Resp. Crit. Care Med. 1995, 151, 27–32. [Google Scholar] [CrossRef]
  72. Hoek, G.; Dockery, D.; Pope, C.A.; Neas, L.; Roemer, W.; Brunekreef, B. Association between PM10 and decrements in peak expiratory flow rates in children: Reanalysis of data from five panel studies. Eur. Resp. J. 1998, 11, 1307–1311. [Google Scholar] [CrossRef]
  73. Brook, R.D.; Rajagopalan, S.; Pope, C.A.; Brook, J.R.; Bhatnagar, A.; Diez-Roux, A.V.; Holguin, F.; Hong, Y.; Luepker, R.V.; Mittleman, M.A.; et al. Particulate matter air pollution and cardiovascular disease: An update to the scientific statement from the American Heart Association. Circulation 2010, 121, 2331–2378. [Google Scholar] [CrossRef]
  74. Sin, D.; Man, S. Chronic obstructive pulmonary disease as a risk factor for cardiovascular morbidity and mortality. Proc. Amer. Thorac. Soc. 2005, 2, 8–11. [Google Scholar] [CrossRef]
  75. Sin, D.; Wu, L.; Man, S. The relationship between reduced lung function and cardiovascular mortality: A population-based study and a systematic review of the literature. Chest 2005, 127, 1952–1959. [Google Scholar] [CrossRef]
  76. Van Eeden, S.; Yeung, A.; Quinlam, K.; Hogg, J. Systemic response to ambient particulate matter: Relevance to chronic obstructive pulmonary disease. Proc. Amer. Thorac. Soc. 2005, 2, 61–67. [Google Scholar] [CrossRef]
  77. Swiston, J.R.; Davidson, W.; Attridge, S.; Li, G.T.; Brauer, M.; van Eeden, S.F. Wood smoke exposure induces a pulmonary and systemic inflammatory response in fire fighters. Eur. Resp. J. 2008, 32, 129–138. [Google Scholar] [CrossRef]
  78. Peters, A.; Döring, A.; Wichmann, H.E.; Koenig, W. Increased plasma viscosity during an air pollution episode: A link to mortality? Lancet 1997, 349, 1582–1587. [Google Scholar] [CrossRef]
  79. Dominici, F.; Sheppard, L.; Clyde, M. Statistical review of air pollution. Int. Statist. Rev. 2003, 71, 243–276. [Google Scholar] [CrossRef]
  80. Baicker, K.; Chandra, A.; Skinner, J.S.; Wennberg, J.E. Who you are and where you live: How race and geography affect the treatment of Medicare beneficiaries. Health Affair. 2004, 23. [Google Scholar] [CrossRef]
  81. Fisher, E.; Whaley, F.; Krushat, W.; Malenka, D.; Fleming, C.; Baron, J.A.; Hsia, D.C. The accuracy of Medicare’s hospital claims data: Progress has been made, but problems remain. Amer. J. Public Health 1992, 82, 243–248. [Google Scholar] [CrossRef]
  82. Havranek, E.; Wolfe, P.; Masoudi, F.; Rathore, S.; Krumholz, H.; Ordin, D. Provider and hospital characteristics associated with geographic variation in the evaluation and management of elderly patients with heart failure. Arch. Intern. Med. 2004, 164, 1186–1191. [Google Scholar]
  83. Kiyota, Y.; Schneeweiss, S.; Glynn, R.; Cannuscio, C.; Avorn, J.; Solomon, D. Accuracy of Medicare claims-based diagnosis of acute myocardial infarction: Estimating positive predictive value on the basis of review of hospital records. Amer. Heart J. 2004, 148, 99–104. [Google Scholar] [CrossRef]
  84. Janssen, N.A.H.; Schwartz, J.; Zanobetti, A.; Suh, H.H. Air conditioning and source-specific particles as modifiers of the effect of PM10 on hospital admissions for heart and lung disease. Environ. Health 2002, 110, 43–49. [Google Scholar]
  85. Analitis, A.; Georgiadis, I.; Katsouyanni, K. Forest fires are associated with elevated mortality in a dense urban setting. Occup. Environ. Med. 2012, 69, 158–162. [Google Scholar]
  86. Hanigan, I.C.; Johnston, F.H.; Morgan, G.G. Vegetation fire smoke, indigenous status and cardio-respiratory hospital admissions in Darwin, Australia, 1996–2005: A time-series study. Environ. Health 2008, 7. [Google Scholar] [CrossRef]
  87. See, S.; Balasubramanian, R.; Rianawati, E.; Karthikevyan, S.; Streets, D. Characterization and source apportionment of particulate matter ≤2.5 μm in Sumatra, Indonesia, during a recent peat fire episode. Environ. Sci. Technol. 2007, 41, 3488–3494. [Google Scholar] [CrossRef]
  88. IPCC 2007: Climate Change: Synthesis Report. Contribution of Working Groups I, II and III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change; Core Writing Team, Pachauri, R.K., Reisinger, A., Eds.; IPCC: Geneva, Switzerland, 2007; p. 104.
  89. Westerling, A.L.; Hidalgo, H.G.; Cayan, D.R.; Swetnam, T.W. Warming and earlier spring increases western U.S. forest wildfire activity. Science 2006, 313, 940–943. [Google Scholar] [CrossRef]
  90. Pechony, O.; Shindell, D.T. Driving forces of global wildfires over the past millennium and the forthcoming century. Proc. Nat. Acad. Sci. USA 2010, 107, 19167–19170. [Google Scholar] [CrossRef]
ISPRS Int. J. Geo-Inf. EISSN 2220-9964 Published by MDPI AG, Basel, Switzerland RSS E-Mail Table of Contents Alert

Further Information

Article Processing Charges Pay an Invoice Open Access Policy Terms of Use Terms and Conditions Privacy Policy Contact MDPI Jobs at MDPI

Guidelines

For Authors For Reviewers For Editors For Librarians For Publishers For Societies

MDPI Initiatives

Institutional Open Access Program (IOAP) Sciforum Preprints Scilit MDPI Books Encyclopedia MDPI Blog

Follow MDPI

LinkedIn Facebook Twitter
Back to Top