Conditional Poisson Regression with Random Effects for the Analysis of Multi-site Time Series Studies.

The analysis of time series studies linking daily counts of a health indicator with environmental variables (e.g., mortality or hospital admissions with air pollution concentrations or temperature; or motor vehicle crashes with temperature) is usually conducted with Poisson regression models controlling for long-term and seasonal trends using temporal strata. When the study includes multiple zones, analysts usually apply a two-stage approach: first, each zone is analyzed separately, and the resulting zone-specific estimates are then combined using meta-analysis. This approach allows zone-specific control for trends. A one-stage approach uses spatio-temporal strata and could be seen as a particular case of the case-time series framework recently proposed. However, the number of strata can escalate very rapidly in a long time series with many zones. A computationally efficient alternative is to fit a conditional Poisson regression model, avoiding the estimation of the nuisance strata. To allow for zone-specific effects, we propose a conditional Poisson regression model with a random slope, although available frequentist software does not implement this model. Here, we implement our approach in the Bayesian paradigm, which also facilitates the inclusion of spatial patterns in the effect of interest. We also provide a possible extension to deal with overdispersed data. We first introduce the equations of the framework and then illustrate their application to data from a previously published study on the effects of temperature on the risk of motor vehicle crashes. We provide R code and a semi-synthetic dataset to reproduce all analyses presented.

Effect of walnut consumption on neuropsychological development in healthy adolescents: a multi-school randomised controlled trial.

Background: Omega-3 fatty acids are critical for neuropsychological functioning. Adolescence is increasingly believed to entail brain vulnerability to dietary intake. The potential benefit on adolescent neurodevelopment of consuming walnuts, a source of omega-3 alpha-linolenic acid (ALA), remains unclear. Methods: We conducted a 6-month multi-school-based randomised controlled nutrition intervention trial to assess whether walnut consumption has beneficial effects on the neuropsychological and behavioural development of adolescents. The study took place between 04/01/2016 and 06/30/2017 in twelve different high schools in Barcelona, Spain (ClinicalTrials.gov Identifier: NCT02590848). A total of 771 healthy teenagers aged 11-16 years were randomised into two equal groups (intervention or control). The intervention group received 30 g/day of raw walnut kernels to be incorporated into their diet for 6 months. Multiple primary endpoints concerning neuropsychological (working memory, attention, fluid intelligence, and executive function) and behavioural (socio-emotional and attention deficit hyperactivity disorder [ADHD] symptoms) development were assessed at baseline and after intervention. Red blood cell (RBC) ALA status was determined at baseline and 6 months as a measure of compliance. Main analyses were based on intention-to-treat using a linear mixed-effects model. A per-protocol effect of the intervention was analysed using inverse-probability weighting to account for post-randomisation prognostic factors (including adherence) using generalised estimating equations. Findings: In intention-to-treat analyses, at 6 months there were no statistically significant changes between the intervention and control groups for all primary endpoints. RBC ALA (%) significantly increased only in the intervention group, coefficient = 0.04 (95% Confidence Interval (CI) = 0.03, 0.06; p < 0.0001). The per-protocol (adherence-adjusted) effect on improvement in attention score (hit reaction time variability) was -11.26 ms (95% CI = -19.92, -2.60; p = 0.011) for the intervention group as compared to the control group, improvement in fluid intelligence score was 1.78 (95% CI = 0.90, 2.67; p < 0.0001), and reduction of ADHD symptom score was -2.18 (95% CI = -3.70, -0.67; p = 0.0050). Interpretation: Our study suggested that being prescribed eating walnuts for 6 months did not improve the neuropsychological function of healthy adolescents. However, improved sustained attention, fluid intelligence, and ADHD symptoms were observed in participants who better complied with the walnut intervention. This study provides a foundation for further clinical and epidemiological research on the effect of walnuts and ALA on neurodevelopment in adolescents. Funding: This study was supported by Instituto de Salud Carlos III through the projects 'CP14/00108, PI16/00261, PI21/00266' (co-funded by European Union Regional Development Fund 'A way to make Europe'). The California Walnut Commission (CWC) has given support by supplying the walnuts for free for the Walnuts Smart Snack Dietary Intervention Trial.

In Vivo Verification of Treatment Source Dwell Times in Brachytherapy of Postoperative Endometrial Carcinoma: A Feasibility Study.

(1) Background: In brachytherapy, there are still many manual procedures that can cause adverse events which can be detected with in vivo dosimetry systems. Plastic scintillator dosimeters (PSD) have interesting properties to achieve this objective such as real-time reading, linearity, repeatability, and small size to fit inside brachytherapy catheters. The purpose of this study was to evaluate the performance of a PSD in postoperative endometrial brachytherapy in terms of source dwell time accuracy. (2) Methods: Measurements were carried out in a PMMA phantom to characterise the PSD. Patient measurements in 121 dwell positions were analysed to obtain the differences between planned and measured dwell times. (3) Results: The repeatability test showed a relative standard deviation below 1% for the measured dwell times. The relative standard deviation of the PSD sensitivity with accumulated absorbed dose was lower than 1.2%. The equipment operated linearly in total counts with respect to absorbed dose and also in count rate versus absorbed dose rate. The mean (standard deviation) of the absolute differences between planned and measured dwell times in patient treatments was 0.0 (0.2) seconds. (4) Conclusions: The PSD system is useful as a quality assurance tool for brachytherapy treatments.

Impact of road traffic noise on annoyance and preventable mortality in European cities: A health impact assessment.

BACKGROUND: Road traffic is the main source of environmental noise in European cities and one of the main environmental risks to health and wellbeing. In this study we aimed to provide an in-depth assessment of available road traffic noise data and to estimate population exposure and health impacts for cities in Europe. METHODS: We conducted the analysis for 724 cities and 25 greater cities in 25 European countries. We retrieved road traffic strategic noise maps delivered under the Environmental Noise Directive (END) or available from local sources. We assessed noise exposure using the 24 h day-evening-night noise level indicator (Lden) starting at exposure levels of 55 dB Lden - based on data availability - for the adult population aged 20 and over (n = 123,966,346). For the adults exposed to noise levels above 55 dB Lden we estimated the health impacts of compliance with the World Health Organization (WHO) recommendation of 53 dB Lden. Two primary health outcomes were assessed: high noise annoyance and Ischemic Heart Disease (IHD), using mortality from IHD causes as indicator. Exposure Response Functions (ERFs) relating road traffic noise exposure to annoyance and IHD mortality were retrieved from the literature. Uncertainties in input parameters were propagated using Monte Carlo simulations to obtain point estimates and empirical 95% Confidence Intervals (CIs). Lastly, the noise maps were categorized as high, moderate and low quality following a qualitative approach. RESULTS: Strategic noise map data was delivered in three distinct formats (i.e. raster, polygon or polyline) and had distinct noise ranges and levels of categorization. The majority of noise maps (i.e. 83.2%) were considered of moderate or low quality. Based on the data provided, almost 60 million adults were exposed to road traffic noise levels above 55 dB Lden, equating to a median of 42% (Interquartile Range (IQR): 31.8-64.8) of the adult population across the analysed cities. We estimated that approximately 11 million adults were highly annoyed by road traffic noise and that 3608 deaths from IHD (95% CI: 843-6266) could be prevented annually with compliance of the WHO recommendation. The proportion of highly annoyed adults by city had a median value of 7.6% (IQR: 5.6-11.8) across the analysed cities, while the number preventable deaths had a median of 2.2 deaths per 100,000 population (IQR: 1.4-3.1). CONCLUSIONS: Based on the provided strategic noise maps a considerable number of adults in European cities are exposed to road traffic noise levels harmful for health. Efforts to standardize the strategic noise maps and to increase noise and disease data availability at the city level are needed. These would allow for a more accurate and comprehensive assessment of the health impacts and further help local governments to address the adverse health effects of road traffic noise.

Study protocol of the European Urban Burden of Disease Project: a health impact assessment study.

INTRODUCTION: Cities have long been known to be society's predominant engine of innovation and wealth creation, yet they are also hotspots of pollution and disease partly due to current urban and transport practices. The aim of the European Urban Burden of Disease project is to evaluate the health burden and its determinants related to current and future potential urban and transport planning practices and related exposures in European cities and make this evidence available for policy and decision making for healthy and sustainable futures. METHODS AND ANALYSIS: Drawing on an established comparative risk assessment methodology (ie, Urban and Transport Planning Health Impact Assessment) tool), in nearly 1000 European cities we will (1) quantify the health impacts of current urban and transport planning related exposures (eg, air pollution, noise, excess heat, lack of green space) (2) and evaluate the relationship between current levels of exposure, health impacts and city characteristics (eg, size, density, design, mobility) (3) rank and compare the cities based on exposure levels and the health impacts, (4) in a number of selected cities assess in-depth the linkages between urban and transport planning, environment, physical activity and health, and model the health impacts of alternative and realistic urban and transport planning scenarios, and, finally, (5) construct a healthy city index and set up an effective knowledge translation hub to generate impact in society and policy. ETHICS AND DISSEMINATION: All data to be used in the project are publicly available data and do not need ethics approval. We will request consent for personal data on opinions and views and create data agreements for those providing information on current and future urban and transport planning scenarios.For dissemination and to generate impact, we will create a knowledge translation hub with information tailored to various stakeholders.

Reflection on modern methods: visualizing the effects of collinearity in distributed lag models.

Collinearity can be a problem in regression models. When examining the effects of an exposure at different time points, constrained distributed lag models can alleviate some of the problems caused by collinearity. Still, some consequences of collinearity may remain and they are often unexplored. We aimed to illustrate the effects of collinearity in the context of distributed lag models, and to provide a tool to assess whether the results of a study could be influenced by collinearity. We used simulations under different scenarios of hypothesized effects of an exposure to visualize the resulting curves of lagged effects. We analysed three real datasets: a cohort study looking for windows of vulnerability to air pollution, a time series study examining the linear association of air pollution with hospital admissions, and a time series study examining the non-linear association between temperature and mortality. We showed that collinearity could be the explanation for some unexpected results, e.g. for statistically significant associations in the opposite direction from that expected, or for wrongly suggesting that some periods are more important than others. We implemented the collin R package to explore the potential consequences of collinearity in the context of distributed lag models. Our visual tool can be a useful way to assess if the results of an analysis may be influenced by collinearity.

Green space and mortality in European cities: a health impact assessment study.

BACKGROUND: Natural outdoor environments including green spaces play an important role in preserving population health and wellbeing in cities, but the number of deaths that could be prevented by increasing green space in European cities is not known. We aimed to estimate the number of natural-cause deaths among adult residents that could be prevented in cities in 31 European countries, if the WHO recommendation for universal access to green space was achieved. METHODS: In this health impact assessment study we focused on adult residents (aged ≥20 years; n=169 134 322) in 978 cities and 49 greater cities, in 31 European countries. We used two green space proxies: normalised difference vegetation index (NDVI), and percentage of green area (%GA). The exposure was estimated at a fine grid-cell level (250 m × 250 m) and the preventable mortality burden for 2015 was estimated at the local city-level. FINDINGS: For 2015 we found that meeting the WHO recommendation of access to green space could prevent 42 968 (95% CI 32 296-64 177) deaths annually using the NDVI proxy (ie, 20% [95% CI 15-30] of deaths per 100 000 inhabitants-year), which represents 2·3% (95% CI 1·7-3·4) of the total natural-cause mortality and 245 (95% CI 184-366) years of life lost per 100 000 inhabitants-year. For the %GA proxy 17 947 (95%CI 0-35 747) deaths could be prevented annually. For %GA the number of attributable deaths were half of that of the NDVI and results were non-significant due to the exposure response function considered. The distribution of NDVI and %GA varied between cities and was not equally distributed within cities. Among European capitals, Athens, Brussels, Budapest, Copenhagen, and Riga showed some of the highest mortality burdens due to the lack of green space. The main source of uncertainty for our results was the choice of the age-structures of the population for the NDVI analysis, and exposure-response function for the %GA analysis. INTERPRETATION: A large number of premature deaths in European cities could be prevented by increasing exposure to green space, while contributing to sustainable, liveable and healthy cities. FUNDING: GoGreenRoutes, Internal ISGlobal fund, and the United States Department of Agriculture Forest Service.

Identifying Factors Influencing Attention in Adolescents with a Co-Created Questionnaire: A Citizen Science Approach with Secondary Students in Barcelona, Spain.

Studies on factors that can influence attention in healthy adolescents are recent and focus on recurrent topics. Students' contribution to public health research often revolves around collecting data but rarely around creating data collection instruments. The ATENC!Ó project reunited secondary students and scientists to create a questionnaire including factors that students thought could affect their attention. We conducted a cross-sectional study to assess whether the factors included in this questionnaire had an effect on attention in adolescents. A total of 1667 students (13-16 years old) from 28 schools in Barcelona performed a validated attention test and answered the questionnaire. The response speed consistency (attentiveness), expressed as hit reaction time standard error (HRT-SE, in ms), was used as the primary outcome. Analyses were conducted using conditional linear regression with school as strata, adjusted for sociodemographic characteristics and further stratified by gender and maternal social class. Some factors showed a negative influence on attention, including taking medication and not reading regularly. We found a significant 14.3% (95% confidence interval: 3.4%, 25.3%) higher median of HRT-SE (increase inattentiveness) among students who reported not having a good relationship with classmates. Students' input into research is relevant for advancing the knowledge production in public health.

Variability of multi-omics profiles in a population-based child cohort.

BACKGROUND: Multiple omics technologies are increasingly applied to detect early, subtle molecular responses to environmental stressors for future disease risk prevention. However, there is an urgent need for further evaluation of stability and variability of omics profiles in healthy individuals, especially during childhood. METHODS: We aimed to estimate intra-, inter-individual and cohort variability of multi-omics profiles (blood DNA methylation, gene expression, miRNA, proteins and serum and urine metabolites) measured 6 months apart in 156 healthy children from five European countries. We further performed a multi-omics network analysis to establish clusters of co-varying omics features and assessed the contribution of key variables (including biological traits and sample collection parameters) to omics variability. RESULTS: All omics displayed a large range of intra- and inter-individual variability depending on each omics feature, although all presented a highest median intra-individual variability. DNA methylation was the most stable profile (median 37.6% inter-individual variability) while gene expression was the least stable (6.6%). Among the least stable features, we identified 1% cross-omics co-variation between CpGs and metabolites (e.g. glucose and CpGs related to obesity and type 2 diabetes). Explanatory variables, including age and body mass index (BMI), explained up to 9% of serum metabolite variability. CONCLUSIONS: Methylation and targeted serum metabolomics are the most reliable omics to implement in single time-point measurements in large cross-sectional studies. In the case of metabolomics, sample collection and individual traits (e.g. BMI) are important parameters to control for improved comparability, at the study design or analysis stage. This study will be valuable for the design and interpretation of epidemiological studies that aim to link omics signatures to disease, environmental exposures, or both.

Walnuts, Long-Chain Polyunsaturated Fatty Acids, and Adolescent Brain Development: Protocol for the Walnuts Smart Snack Dietary Intervention Trial.

Background: Adolescence, when the most complex behaviors are refined to adult sophistication, represents a major window of opportunity and vulnerability for neuropsychological development. To support and protect this complex and active brain growth, different nutritional components considered essential need to be acquired from the diet. For instance, omega-3 fatty acids are mainly obtained from seafood, seeds, and walnuts. Known for their rich lipid profile, walnuts contain sizable amounts of an essential fatty acid, alpha-linolenic acid (ALA), the vegetable omega-3 fatty acid that is the precursor of two longer-chain omega-3 polyunsaturated fatty acids (omega-3 PUFA): docosahexaenoic (DHA) and eicosapentaenoic (EPA) acids. While there is growing evidence of neuropsychological improvements in the young developing brain associated with omega-3 PUFA intake, few studies have examined whether consuming walnuts during adolescence entails similar beneficial effects. There is a need to further explore the ways in which walnuts influence youthful brain function, particularly for the long-term. Thus, we designed the WALNUTs study (WSS), a population-based randomized controlled trial conducted in adolescents in Barcelona, Spain. We hypothesize that walnut intake will increase omega-3 PUFA tissue availability (particularly ALA) to a level that enhances the neuropsychological development during adolescence. Methodology/Design: We conducted a 6-month population-based randomized controlled trial in teenagers (n = 800) and we aimed to determine the effectiveness of the intervention (four walnuts per day, or 30 kernel g, ~1.5g of ALA) in enhancing brain neuropsychological and socio-emotional development compared to a control group with no walnut intervention. Before randomization, different neuropsychological tests were recorded for all participants, and blood samples (in a subsample of participants) were collected to measure omega-3 PUFA levels at baseline, and all again, after randomization and the intervention. The data is now collected and we will conduct linear regression models to assess the effect of the intervention. Discussion: The WALNUTs (WSS) study results will allow us to better understand the role of plant-based omega-3 PUFA intake from regular walnut consumption on neuropsychological development during adolescence. Results could be translated into nutritional public health recommendations targeting teenagers. Trial Registration: ClinicalTrials.gov, U.S. National Library of Medicine, National Institutes of Health # NCT02590848. Retrospectively registered 29/10/2015.

Short-term effect of air pollution on attention function in adolescents (ATENC!Ó): A randomized controlled trial in high schools in Barcelona, Spain.

BACKGROUND: The recent evidence of the short-term impact of air pollution on youth cognitive functions is based primarily on observational studies. OBJECTIVES: We conducted a randomized controlled trial to assess whether purifying the air of the classrooms produced short-term changes in attention processes of adolescents. METHODS: We recruited a total of 2,123 adolescents (13-16 years old) in 33 high schools in Barcelona metropolitan area (Spain). In each school, adolescents from each class were randomly split into two equal-sized groups and assigned to two different classrooms. A set of two air cleaner devices with the same appearance (one recirculating and filtrating the air and the other only recirculating the air) was used. Each one of the devices was placed at random at one of the two classrooms. Students were masked to intervention allocation and had to complete several computerized activities for 1.5 h, including an attention test (Flanker task) to be performed at baseline and at the end of the intervention. The response speed consistency, expressed as hit reaction time standard error (HRT-SE, in ms), was measured as the primary outcome. Analyses were conducted using conditional linear regressions with classroom as strata, adjusted for variables that may differ from one class to another such as temperature, humidity and carbon dioxide concentration. RESULTS: Average levels of PM2.5 and black carbon throughout the 1.5 h of experiment were 89% and 87%, respectively, lower in the classrooms with air cleaner than in the control classrooms. No differences were found in the median of HRT-SE between classrooms with cleaned air and normal air (percent change: 1.37%, 95% confidence interval: -2.81%, 5.56%). Sensitivity analyses with secondary attention outcomes resulted in similar findings. CONCLUSIONS: Cleaning the air of a classroom to reduce exposure to air pollutants for 1.5 h did not have an impact on the attention function of adolescents. Still, in light of previous evidence suggesting an association between air pollution and attention, further experimental studies should explore other short-term timescales of exposure and age ranges.

Does surrounding greenness moderate the relationship between apparent temperature and physical activity? Findings from the PHENOTYPE project.

BACKGROUND: Physical activity can be affected by both meteorological conditions and surrounding greenness, but few studies have evaluated the effects of these environmental factors on physical activity simultaneously. This multi-city comparative study aimed to assess the synergetic effects of apparent temperature and surrounding greenness on physical activity in four European cities. Specifically, we aimed to identify an interaction between surrounding greenness and apparent temperature in the effects on physical activity. METHODS: Data were collected from 352 adult residents of Barcelona (Spain), Stoke-on-Trent (United Kingdom), Doetinchem (The Netherlands), and Kaunas (Lithuania) as part of the PHENOTYPE study. Participants wore a smartphone for seven consecutive days between May-December 2013 and provided additional sociodemographic survey data. Hourly average physical activity (Metabolic Equivalent of Task (MET)) and surrounding greenness (NDVI) were derived from the Calfit mobile application collecting accelerometer and location data. Hourly apparent temperature was calculated from temperature and relative humidity, which were obtained from local meteorological stations along with other meteorological covariates (rainfall, windspeed, and sky darkness). We assessed the interaction effects of apparent temperature and surrounding greenness on hourly physical activity for each city using linear mixed models, while adjusting for meteorological, demographic, and time-related variables. RESULTS: We found significant interactions between apparent temperature and surrounding greenness on hourly physical activity in three of four cities, aside from the coastal city of Barcelona. Significant quadratic effects of apparent temperature were found in the highest level of surrounding greenness for Stoke-on-Trent and Doetinchem, with 4% decrease in median MET observed for a 10°C departure from optimal temperature (15.2°C and 14.6°C, respectively). Significant linear effects were found for higher levels of surrounding greenness in Kaunas, whereby an increase of 10°C was associated with ∼4% increase in median MET. CONCLUSION: Apparent temperature and surrounding greenness interacted in the effect on hourly physical activity across three of four European cities, with varying effect between cities. While quadratic effects of temperature suggest diminishing levels of physical activity in the highest greenness levels in cities of temperate climates, the variation in surrounding greenness between cities could be further explored, particularly by looking at indoor-outdoor locations. The study findings support the need for evidence-based physical activity promotion and urban design.

Ultraviolet radiation as a predictor of sex hormone levels in postmenopausal women: A European multi-center study (ECRHS).

BACKGROUND: Solar ultraviolet radiation (UVR) affects the body through pathways that exhibit positive as well as negative health effects such as immunoregulation and vitamin D production. Different vitamin D metabolites are associated with higher or lower concentrations of estrogens and may thus alter the female sex hormone balance. OBJECTIVE: To study whether exposure to UVR, as a modifiable lifestyle factor, is associated with levels of sex hormones (17ß-estradiol, estrone, estrone 3-sulfate, testosterone, dehydroepiandrosterone sulfate), gonadotropins (follicle stimulating hormone, luteinizing hormone) as well as sex hormone binding globulin in postmenopausal women, and thus investigate whether managing UVR exposure can influence the hormone balance, with potential benefits for the biological aging process. METHODS: The study included 580 postmenopausal women from six European countries, participating in the European Community Respiratory Health Survey (2010-2014). Average UVR exposure during the month before blood sampling was estimated based on personal sun behavior and ambient levels. Hormone concentrations were measured in serum using state-of-the-art methods. Subsequently we applied linear mixed-effects models, including center as random intercept, hormone concentrations (one at a time) as outcome and UVR, age, skin type, body mass index, vitamin D from dietary sources, smoking, age at completed full-time education and season of blood sampling as fixed-effect predictors. RESULTS: One interquartile range increase in UVR exposure was associated with decreased levels of 17ß-estradiol (-15.6 pmol/L, 95 % Confidence Interval (CI): -27.69, -3.51) and estrone (-13.36 pmol/L, 95 % CI: -26.04, -0.68) and increased levels of follicle stimulating hormone (9.34IU/L, 95 % CI: 2.91, 15.77) and luteinizing hormone (13.86 IU/daL, 95 % CI: 2.48, 25.25). CONCLUSIONS: Exposure to UVR is associated with decreased estrogens and increased gonadotropins in postmenopausal women, a status associated with osteoporosis, lung function decline and other adverse health effects. This study indicates that managing UVR exposure has potential to influence the hormone balance and counteract adverse health conditions after menopause.

Premature mortality due to air pollution in European cities: a health impact assessment.

BACKGROUND: Ambient air pollution is a major environmental cause of morbidity and mortality worldwide. Cities are generally hotspots for air pollution and disease. However, the exact extent of the health effects of air pollution at the city level is still largely unknown. We aimed to estimate the proportion of annual preventable deaths due to air pollution in almost 1000 cities in Europe. METHODS: We did a quantitative health impact assessment for the year 2015 to estimate the effect of air pollution exposure (PM2·5 and NO2) on natural-cause mortality for adult residents (aged ≥20 years) in 969 cities and 47 greater cities in Europe. We retrieved the cities and greater cities from the Urban Audit 2018 dataset and did the analysis at a 250 m grid cell level for 2015 data based on the global human settlement layer residential population. We estimated the annual premature mortality burden preventable if the WHO recommended values (ie, 10 µg/m3 for PM2·5 and 40 µg/m3 for NO2) were achieved and if air pollution concentrations were reduced to the lowest values measured in 2015 in European cities (ie, 3·7 µg/m3 for PM2·5 and 3·5 µg/m3 for NO2). We clustered and ranked the cities on the basis of population and age-standardised mortality burden associated with air pollution exposure. In addition, we did several uncertainty and sensitivity analyses to test the robustness of our estimates. FINDINGS: Compliance with WHO air pollution guidelines could prevent 51 213 (95% CI 34 036-68 682) deaths per year for PM2·5 exposure and 900 (0-2476) deaths per year for NO2 exposure. The reduction of air pollution to the lowest measured concentrations could prevent 124 729 (83 332-166 535) deaths per year for PM2·5 exposure and 79 435 (0-215 165) deaths per year for NO2 exposure. A great variability in the preventable mortality burden was observed by city, ranging from 0 to 202 deaths per 100 000 population for PM2·5 and from 0 to 73 deaths for NO2 per 100 000 population when the lowest measured concentrations were considered. The highest PM2·5 mortality burden was estimated for cities in the Po Valley (northern Italy), Poland, and Czech Republic. The highest NO2 mortality burden was estimated for large cities and capital cities in western and southern Europe. Sensitivity analyses showed that the results were particularly sensitive to the choice of the exposure response function, but less so to the choice of baseline mortality values and exposure assessment method. INTERPRETATION: A considerable proportion of premature deaths in European cities could be avoided annually by lowering air pollution concentrations, particularly below WHO guidelines. The mortality burden varied considerably between European cities, indicating where policy actions are more urgently needed to reduce air pollution and achieve sustainable, liveable, and healthy communities. Current guidelines should be revised and air pollution concentrations should be reduced further to achieve greater protection of health in cities. FUNDING: Spanish Ministry of Science and Innovation, Internal ISGlobal fund.

Public Transport Strikes and Their Relationships With Air Pollution, Mortality, and Hospital Admissions.

There is limited suggestive evidence of relationships between public transport strikes and either increased air pollution or worse population health. In this study we aimed to assess whether public transport strikes were associated with increases in health events (overall, cardiovascular and respiratory mortality, and cardiovascular and respiratory hospitalizations). We also explored whether air pollution mediated those associations. We used data from the city of Barcelona (Spain) for the period 2005-2016 on strikes, health events, and ambient air pollution (nitrogen dioxide, nitrogen monoxide, particulate matter (PM) with an aerodynamic diameter ≤10 µm, PM with an aerodynamic diameter ≤2.5µm, PM with an aerodynamic diameter ≤1µm, number of particles with a diameter greater than 5 nm per cm3 (particle number concentration), and black carbon). We used linear and quasi-Poisson regression models to explore the associations between air pollution and public transport strikes and between public transport strikes and health outcomes. We also investigated potential causal mediation by air pollution. Overall, this study suggested that public transport strikes are associated with increased overall mortality, respiratory mortality, and respiratory hospitalizations. However, our findings suggest that such increases are not mediated by the increase in air pollution. Our results indicate the need to further investigate these relationships and potential mechanisms.

Ambient air pollution and overweight and obesity in school-aged children in Barcelona, Spain.

BACKGROUND: Ambient air pollution may increase the risk of overweight and obesity in children. However, available evidence is still scarce and has mainly focused on ambient air pollution exposure occurring at home without considering the school environment. The aim of this study is to assess whether exposure to ambient air pollution at home and school is associated with overweight and obesity in primary school children. METHODS: We studied 2660 children aged 7-10 years during 2012 in Barcelona. Child weight and height were measured and age- and sex-specific z-scores for body mass index (zBMI) were calculated using the WHO growth reference 2007. Overweight and obesity were defined using the same reference. Land use regression models were used to estimate levels of nitrogen dioxide (NO2), particulate matter <2.5 µm (PM2.5), <10 µm (PM10) and coarse (PMcoarse) at home. Outdoor levels of NO2, PM2.5, elemental carbon (EC), and ultrafine particles (UFP) were measured in the schoolyard. Multilevel mixed linear and ordered logistic models were used to assess the association between ambient air pollution (continuous per interquartile range (IQR) increase and categorical with tertile cutoffs) and zBMI (continuous and ordinal: normal, overweight, obese), after adjusting for socio-demographic characteristics. RESULTS: An IQR increase in PM10-home (5.6 µg/m3) was associated with a 10% increase in the odds of being overweight or obese (odds ratio (OR) = 1.10; 95% CI = 1.00, 1.22). Children exposed to the highest tertile of UFP-school (>27,346 particles/cm3) had a 30% higher odds of being overweight or obese (OR = 1.30; 95%CI = 1.03, 1.64) compared to the lowest tertile of UFP exposure. We also observed that exposure to NO2, PM2.5 or EC at schools was associated with higher odds of overweight or obese at medium compared to low levels of exposure. Home and school exposures did not show any significant associations with zBMI (except PM2.5-school comparing tertile 2 vs tertile 1) but were similar in direction. CONCLUSIONS: This study suggests that exposure to ambient air pollution, especially at school, is associated with childhood risk for overweight and obesity. A cautious interpretation is warranted because associations were not always linear and because school and home air pollution measurements were not directly comparable.

Analysis of multicentre epidemiological studies: contrasting fixed or random effects modelling and meta-analysis.

Multicentre studies are common in epidemiological research aiming at identifying disease risk factors. A major advantage of multicentre over single-centre studies is that, by including a larger number of participants, they allow consideration of rare outcomes and exposures. Their multicentric nature introduces some complexities at the step of data analysis, in particular when it comes to controlling for confounding by centre, which is the focus of this tutorial. Commonly, epidemiologists use one of the following options: pooling all centre-specific data and adjusting for centre using fixed effects; adjusting for centre using random effects; or fitting centre-specific models and combining the results in a meta-analysis. Here, we illustrate the similarities of and differences between these three modelling approaches, explain the reasons why they may provide different conclusions and offer advice on which model to choose depending on the characteristics of the study. Two key issues to examine during the analyses are to distinguish within-centre from between-centre associations, and the possible heterogeneity of the effects (of exposure and/or confounders) by centre. A real epidemiological study is used to illustrate a situation in which these various options yield different results. A synthetic dataset and R and Stata codes are provided to reproduce the results.

A systematic comparison of statistical methods to detect interactions in exposome-health associations.

BACKGROUND: There is growing interest in examining the simultaneous effects of multiple exposures and, more generally, the effects of mixtures of exposures, as part of the exposome concept (being defined as the totality of human environmental exposures from conception onwards). Uncovering such combined effects is challenging owing to the large number of exposures, several of them being highly correlated. We performed a simulation study in an exposome context to compare the performance of several statistical methods that have been proposed to detect statistical interactions. METHODS: Simulations were based on an exposome including 237 exposures with a realistic correlation structure. We considered several statistical regression-based methods, including two-step Environment-Wide Association Study (EWAS2), the Deletion/Substitution/Addition (DSA) algorithm, the Least Absolute Shrinkage and Selection Operator (LASSO), Group-Lasso INTERaction-NET (GLINTERNET), a three-step method based on regression trees and finally Boosted Regression Trees (BRT). We assessed the performance of each method in terms of model size, predictive ability, sensitivity and false discovery rate. RESULTS: GLINTERNET and DSA had better overall performance than the other methods, with GLINTERNET having better properties in terms of selecting the true predictors (sensitivity) and of predictive ability, while DSA had a lower number of false positives. In terms of ability to capture interaction terms, GLINTERNET and DSA had again the best performances, with the same trade-off between sensitivity and false discovery proportion. When GLINTERNET and DSA failed to select an exposure truly associated with the outcome, they tended to select a highly correlated one. When interactions were not present in the data, using variable selection methods that allowed for interactions had only slight costs in performance compared to methods that only searched for main effects. CONCLUSIONS: GLINTERNET and DSA provided better performance in detecting two-way interactions, compared to other existing methods.