Generalized additive models in time-series studies of air pollution and health

On generalized additive models in time series studies of air pollution and health. A major statistical concern came into view recently about the appropriateness of the use of gams in the presence of concurvity, which is likely to be present in the data of all air pollution. On the use of generalized additive models in timeseries studies of air pollution and health and temperature and mortality in 11 cities of the eastern united states. Although air pollution is a known fundamental problem in china, few studies have investigated the associations between ambient air pollution and respiratory mortality in nonmetropolitan cities of china. Third, we introduce a conceptual framework to fully explore the sensitivity of the air pollution risk estimates to model choice. Introduction pollution of outdoor air is a public health concern throughout the world. The study aimed to investigate a potential relationship between shortterm exposure to ambient air pollutants and respiratory mortality in xian, china. This hei special report details attempts to address several questions raised by these discoveries. Here, a gam with autoregressive terms gamar is introduced to fill this gap.

Some studies generally estimated the health effect of each pollutant separately via a singlepollutant model controlling for time trends and meteorological factors 5,6,7. Improved semiparametric time series models of air pollution. Using supervised principal components analysis to assess. On the use of generalized additive models in timeseries studies of air pollution and health francesca dominici1, aidan mcdermott1, scott l. The gam is a generalization of the generalized linear model and is widely used in time series studies on health effects of air pollution because it does not expect a particular functional form of a relationship, and is flexible for modelling nonlinear associations crawley 20. Several recent studies have reported significant health effects of air pollution even at low levels of air pollutants, but in most of these studies linear nonthreshold relations were assumed. Here, a gam with autoregressive terms gamar is introduced to fill. Generalized additive models with principal component. Jan 10, 2011 in this study we applied the generalized additive model gam extended poisson regression model to quantitatively evaluate the effects of ambient air pollutants, no 2, pm 10 and so 2, on the preterm birth by analyzing the time series data of air pollution, meteorological factors, and preterm births in guangdong province in 2007. The study used an integrated spatial and temporal approach that included the kriging method and the generalized additive model gam. Shows application of gam in air pollution, climate change, and health. Generalized additive model gam with natural cubic splines ns has been commonly used as a standard analytical tool in time series studies of health effects of air pollution. The use of nonparametric smoothing in time series models of air pollution and health was suggested in schwartz 1994a, where generalized additive poisson models were used with loess smooths of time, temperature, dewpoint temperature and pm 10.

When the data to which the gam are being applied have two characteristics1 the estimated regression coefficients are small and 2. Methods in this study, a semiparametric generalized additive model gam was used to evaluate the specific influences of air pollutants pm10, so2, and no2 on hospital emergency admissions with different lag structures from 2009 to 2011, the sex and. On the use of generalized additive models in timeseries studies. A new variance estimator for parameters of semiparametric. We apply our methods to data of the national mortality morbidity air pollution study, which includes time series data from the 90 largest u. This discovery delayed the completion of the pm criteria document. Generalized additive models gam was used for data analysis with. On the use of generalized additive models in time series studies of air pollution and health and temperature and mortality in 11 cities of the eastern united states. Acute effect of daily fine particulate matter pollution on. Reanalysis of the harvard six cities study and the american cancer society study of particulate air pollution and mortality, part ii.

A temporal, multicity model to estimate the effects of. The generalized additive model allows regressions to include nonparametric smooth functions to model the potential nonlinear dependence of the daily respiratory symptoms on weather and season. Association between particulate matter and emergency room. Particulate air pollution and mortality in the united states. Generalized additive models gams with natural cubic splines ns as smoothing functions have become standard analytical tools in time series studies of health effects of air pollution. Estimating generalized semiparametric additive models. In fact, advanced statistical methods are necessary to address the challenges inherent in the detection of a relatively small. Jan, 2017 the adverse effect of air pollution on health has become a crucial research area in the last decade 1,2,3. Burnett, metaanalysis of timeseries studies of air pollution and mortality. Lately, however, it has been found that concurvitythe nonparametric analogue of multicollinearitymight lead to underestimation of standard errors of the effects of. Recently, the use of the gam has been extended from time series data to spatial data. Metaanalysis of time series studies of air pollution and mortality. The widely used generalized additive models gam method is a flexible and effective technique for conducting nonlinear regression analysis in timeseries studies of the health effects of air. The widely used generalized additive models gam method is a flexible and effective technique for conducting nonlinear regression analysis in timeseries studies of the health effects of air pollution.

Burnett healthy environments and consumer safety branch, health canada, ottawa, ontario, canada in an earlier paper, we reported the results of a meta. Mathematical models for air pollution health effects by. Bootstrap model averaging in time series studies of. Statistical issues and a novel model fitting approach shui he, phd university of pittsburgh, 2004 the generalized additive model gam has been used as a standard tool for epidemiologic analysis exploring the e. Mortality projections using generalized additive models with. Mathematical models for air pollution health effects. The generalized additive models represent a method of fittin g a smooth. Health revised analyses of effects timeseries studies of. Studies mentioned in table 1 used different models in gam and time series models, including link functions and various splines. Key outcomes from each study are also discussed, along with observations made. Generalized additive models for data with concurvity. Hence there is clearly a need for mixed additive models that allow assessment of heterogeneity while addressing nonlinear relationships.

It does not require the shape of the response curve as a priori knowledge, and allows for nonparametric adjustments for nonlinear confounding effects. Using a generalized additive model with autoregressive terms. On the use of generalized additive models in timeseries. Objective to investigate the association between ambient air pollution and hospital emergency admissions in beijing. Update in relation to the use of generalized additive models david m. A bootstrap method to avoid the effect of concurvity in. Generalized additive models in environmental health. To investigate the no2 mortality doseresponse association in nine cities participating in the aphea2 project using two different methods.

On the use of generalized additive models in timeseries studies of air pollution and health, am. The effect of concurvity in generalized additive models link. Extensive clinical, epidemiological, and toxicological studies have provided evidence of. The relation between air pollution and respiratory deaths in tehran. A measurement error model for timeseries studies of air. In recent years a great number of studies have applied generalised additive models gams to time series data to estimate the short term health effects of air pollution. Investigating the doseresponse relation between air. On the use of generalized additive models in time series studies of air pollution and health. Preponderance methods used by various researchers include timeseries analysis with generalized linear models glm or generalized additive models gam using parametricnonparametric splines for estimating the long and small interval health effects of atmospheric parameters along with air pollution ramsay et al. A comparative study of the use of gam and glm in air pollution. In statistics, a generalized additive model gam is a generalized linear model in which the linear predictor depends linearly on unknown smooth functions of some predictor variables, and interest focuses on inference about these smooth functions. Table 1 depicts the comprehensive details of various studies in the area of air pollution and human health.

Generalized additive models gams with age, period and cohort as possible covariates are used to predict future mortality improvements for the irish population. Statistically significant associations were identified between pm10 and pm2. However, gam assumes that errors are mutually independent, while time series can be correlated in adjacent time points. Ijerph free fulltext effects of particulate matter.

Dec 01, 2006 risk models for particulate air pollution. Association between ambient air pollution and hospital. The widely used generalized additive models gam method is a flexible and effective technique for conducting nonlinear regression analysis in time series studies of the health effects of air. The adverse effect of air pollution on health has become a crucial research area in the last decade 1,2,3. In this study we applied the generalized additive model gam extended poisson regression model to quantitatively evaluate the effects of ambient air pollutants, no 2, pm 10 and so 2, on the preterm birth by analyzing the timeseries data of air pollution, meteorological factors, and preterm births in guangdong province in 2007. Ambient air pollution and respiratory mortality in xian. Model choice in time series studies of air pollution and. However, gam assumes that errors are mutually independent, while time series can be. The effect of concurvity in generalized additive models. Exploring bias in a generalized additive model for spatial. Generalized additive models gams have been used as a standard analytic tool in time.

Spatiotemporal analysis of air pollution and asthma. The first section addresses the impact of the issues on the heifunded national morbidity, mortality, and air pollution study nmmaps. Generalized linear mixed models in time series studies of. Using a generalized additive model with autoregressive. Aug 17, 2005 the consensus from time series studies that have investigated the mortality effects of particulate matter air pollution pm is that increases in pm are associated with increases in daily mortality. Aug 01, 2003 dominici f, mcdermott a, zeger s, samet j. Generalized additive model gam is a flexible and effective technique for estimating the unknown nonlinear relationship between health effects and air pollution 15, 16. In the last decade, many epidemiological studies have shown an association between measurements of ambient concentrations of. The aim of this study was to investigate the effect of air pollution on mortality from respiratory diseases in tehran, iran. Annotated bibliography of recent studies on the health. For comparison with many papers in the published literature and to address the impact of recently identified problems with generalized additive models gam in estimating air pollution health. If there is no linear part, then this model reduces to the familiar generalized additive model hastie and tibshirani 1990.

Update in relation to the use of generalized additive models. In this ecological study, air pollution data was inquired from the tehran province environmental protection agency and the tehran air. Annotated bibliography of recent studies on the health effects of air pollution in the last year, hundreds of scientific papers on the health effects of air. Generalized linear mixed models in time series studies of air.

Different time series methods have been used in these studies, i. Spatial regression models in epidemiological studies. On the use of generalized additive models in timeseries studies of. Mar 20, 2018 dominici f, mcdermott a, zeger sl, samet jm. Metaanalysis of timeseries studies of air pollution and. Lately, however, it has been found that concurvitythe nonparametric analogue of multicollinearitymight lead to underestimation of standard errors of the effects of independent variables. The relation between air pollution and respiratory deaths. In this ecological study, air pollution data was inquired from the tehran province environmental protection agency and the tehran air quality control company. The relation between air pollution and respiratory deaths in. A time series study was conducted using poisson regression with generalized additive models adjusted for confounders. Generalized additive model gam was used for multivariate analysis. The widely used generalized additive models gam method is a flexible and effective technique for conducting nonlinear regression analysis in time series studies of the health effects of air pollution. Health revised analyses of effects timeseries studies.

The impact of outdoor air pollutants on outpatient visits. One important application of gsams involves timeseries analysis of the acute effect of ambient air pollution on public health ramsay 2005. Generalized additive models the gam see hastie and tibshirani 1990 with a poisson marginal distribution is typically used to relate a discrete outcome variable with a set of covariates in the epidemiological area, for example, to quantify the association between health problems and air pollution concentrations. Generalized additive models gams have been used as a standard analytic tool in timeseries studies of air pollution and health during the last. Method for mapping populationbased casecontrol studies. The widely used generalized additive models gam method is a flexible and effective. The shortterm effects of air pollutants on respiratory. Exploring bias in a generalized additive model for. Revised analyses of timeseries studies of air pollution and health. Gams were originally developed by trevor hastie and robert tibshirani to blend properties of generalized linear models with additive models. Aug 01, 2003 during the past few years, the generalized additive model gam has become a standard tool for epidemiologic analysis exploring the effect of air pollution on population health.

This approach can be thought of as regressing residuals from the smoothed dependent variable on. In the gam models used by researchers to explore the effect of air pollution on population health, a poisson model is typically used with time series data to explain log counts of an outcome such as mortality or hospital admissions as a function of air pollution plus a sum of functions of other confounding variables. Extensive clinical, epidemiological, and toxicological studies. We analyzed daily outpatient and emergency visit data from the taiwan bureau of national health insurance and air pollution data from the taiwan environmental protection administration during 20002002. Studies of air pollution and human health have evolved from descriptive studies of the early phenomena of large increases in adverse health effects following extreme air pollution episodes to time series analyses based on the use of sophisticated regression models. An increasing number of studies have demonstrated the association between daily ambient air pollution levels and adverse health outcomes worldwide 1,2,3,4. Generalized additive models with principal component analysis. Generalized additive model gam is a flexible and effective technique for estimating the unknown nonlinear relationship between health effects and air pollution. Environmental project, 1005 time series study of air. On the use of generalized additive models in timeseries studies of air pollution and health. Time series study of air pollution health effects in copsac children. Oct 30, 2012 generalized additive model gam provides a flexible and effective technique for modelling nonlinear time series in studies of the health effects of environmental factors.

Samet2 1 department of biostatistics, bloomberg school of public health, johns hopkins university, baltimore, md. Metaanalysis of timeseries studies of air pollution and mortality. Improved semiparametric time series models of air pollution and mortality. During the past few years, the generalized additive model gam has become a standard tool for epidemiologic analysis exploring the effect of air pollution on population health. Introduction epidemiologic time series studies conducted in cities around the world have consistently found associations between daily levels of airborne particulate matter smaller than 10 microns. Some epidemiological evidence has shown a relation between ambient air pollution and adverse health outcomes.

Shortterm effects of ambient air pollution on chronic. Generalized additive model gam provides a flexible and effective technique for modelling nonlinear timeseries in studies of the health effects of environmental factors. A comparative study of the use of gam and glm in air. Generalized additive and generalized linear models, presentation on variance of gam estimators, environmental protection agency workshop on gamrelated statistical issues in pm epidemiology, november 46, 2002, durham, nc.