Area-level excess mortality in times of COVID-19 in Switzerland: geographical, socioeconomic and political determinants.

The coronavirus disease 2019 (COVID-19)-related excess mortality in Switzerland is well documented, but no study examined mortality at the small-area level. We analysed excess mortality in 2020 for 2141 Swiss municipalities using a Bayesian spatiotemporal model fitted to 2011-19 data. Areas most affected included the Ticino, the Romandie and the Northeast. Rural areas, municipalities within cross-border labour markets, of lower socioeconomic position and with less support for control measures in the popular vote on the COVID-19 Act had greater excess mortality. Particularly vulnerable municipalities require special efforts to mitigate the impact of pandemics.

Asthma hospitalisations and heat exposure in England: a case-crossover study during 2002-2019.

BACKGROUND: Previous studies have reported an association between warm temperature and asthma hospitalisation. They have reported different sex-related and age-related vulnerabilities; nevertheless, little is known about how this effect has changed over time and how it varies in space. This study aims to evaluate the association between asthma hospitalisation and warm temperature and investigate vulnerabilities by age, sex, time and space. METHODS: We retrieved individual-level data on summer asthma hospitalisation at high temporal (daily) and spatial (postcodes) resolutions during 2002-2019 in England from the NHS Digital. Daily mean temperature at 1 km×1 km resolution was retrieved from the UK Met Office. We focused on lag 0-3 days. We employed a case-crossover study design and fitted Bayesian hierarchical Poisson models accounting for possible confounders (rainfall, relative humidity, wind speed and national holidays). RESULTS: After accounting for confounding, we found an increase of 1.11% (95% credible interval: 0.88% to 1.34%) in the asthma hospitalisation risk for every 1°C increase in the ambient summer temperature. The effect was highest for males aged 16-64 (2.10%, 1.59% to 2.61%) and during the early years of our analysis. We also found evidence of a decreasing linear trend of the effect over time. Populations in Yorkshire and the Humber and East and West Midlands were the most vulnerable. CONCLUSION: This study provides evidence of an association between warm temperature and hospital admission for asthma. The effect has decreased over time with potential explanations including temporal differences in patterns of heat exposure, adaptive mechanisms, asthma management, lifestyle, comorbidities and occupation.

Ambient heat exposure and COPD hospitalisations in England: a nationwide case-crossover study during 2007-2018.

BACKGROUND: There is emerging evidence suggesting a link between ambient heat exposure and chronic obstructive pulmonary disease (COPD) hospitalisations. Individual and contextual characteristics can affect population vulnerabilities to COPD hospitalisation due to heat exposure. This study quantifies the effect of ambient heat on COPD hospitalisations and examines population vulnerabilities by age, sex and contextual characteristics. METHODS: Individual data on COPD hospitalisation at high geographical resolution (postcodes) during 2007-2018 in England was retrieved from the small area health statistics unit. Maximum temperature at 1 km ×1 km resolution was available from the UK Met Office. We employed a case-crossover study design and fitted Bayesian conditional Poisson regression models. We adjusted for relative humidity and national holidays, and examined effect modification by age, sex, green space, average temperature, deprivation and urbanicity. RESULTS: After accounting for confounding, we found 1.47% (95% Credible Interval (CrI) 1.19% to 1.73%) increase in the hospitalisation risk for every 1°C increase in temperatures above 23.2°C (lags 0-2 days). We reported weak evidence of an effect modification by sex and age. We found a strong spatial determinant of the COPD hospitalisation risk due to heat exposure, which was alleviated when we accounted for contextual characteristics. 1851 (95% CrI 1 576 to 2 079) COPD hospitalisations were associated with temperatures above 23.2°C annually. CONCLUSION: Our study suggests that resources should be allocated to support the public health systems, for instance, through developing or expanding heat-health alerts, to challenge the increasing future heat-related COPD hospitalisation burden.

Community-level characteristics of COVID-19 vaccine hesitancy in England: A nationwide cross-sectional study.

One year after the start of the COVID-19 vaccination programme in England, more than 43 million people older than 12 years old had received at least a first dose. Nevertheless, geographical differences persist, and vaccine hesitancy is still a major public health concern; understanding its determinants is crucial to managing the COVID-19 pandemic and preparing for future ones. In this cross-sectional population-based study we used cumulative data on the first dose of vaccine received by 01-01-2022 at Middle Super Output Area level in England. We used Bayesian hierarchical spatial models and investigated if the geographical differences in vaccination uptake can be explained by a range of community-level characteristics covering socio-demographics, political view, COVID-19 health risk awareness and targeting of high risk groups and accessibility. Deprivation is the covariate most strongly associated with vaccine uptake (Odds Ratio 0.55, 95%CI 0.54-0.57; most versus least deprived areas). The most ethnically diverse areas have a 38% (95%CI 36-40%) lower odds of vaccine uptake compared with those least diverse. Areas with the highest proportion of population between 12 and 24 years old had lower odds of vaccination (0.87, 95%CI 0.85-0.89). Finally increase in vaccine accessibility is associated with COVID-19 vaccine coverage (OR 1.07, 95%CI 1.03-1.12). Our results suggest that one year after the start of the vaccination programme, there is still evidence of inequalities in uptake, affecting particularly minorities and marginalised groups. Strategies including prioritising active outreach across communities and removing practical barriers and factors that make vaccines less accessible are needed to level up the differences.

Childhood cancer and residential proximity to petrol stations: a nationwide registry-based case-control study in Switzerland and an updated meta-analysis.

PURPOSE: Benzene is a known carcinogen for adult leukemia. Exposure to benzene through parental occupation and the use of household products has been associated with childhood leukemia (CL). Ambient benzene has also been associated with CL and central nervous system (CNS) tumors. We aimed to investigate whether the higher ambient levels of benzene in proximity of petrol stations are associated with a greater risk of childhood cancers, leukemia, and CNS tumors. METHODS: We identified children diagnosed with cancer at age 0-15 years during 1985-2015 from the Swiss Childhood Cancer Registry and selected 10 age and sex-matched controls per case from national censuses. We calculated the distance from children's home to the nearest petrol station using precise geocodes. We estimated odds ratios using conditional logistic regression adjusting for ambient levels of NO2, distance to highways, level of urbanization, and presence of a cantonal cancer registry. In addition, we ran a meta-analysis pooling current results for CL with those of previous studies. RESULTS: We identified 6129 cases, of which 1880 were leukemias and 1290 CNS tumors. 24 cases lived within 50 m from a petrol station. The adjusted odds ratio of a cancer diagnosis for children thus exposed compared to unexposed children (> 500 m) was 1.29 (0.84-1.98) for all cancers combined, 1.08 (0.46-2.51) for leukemia, and 1.30 (0.51-3.35) for CNS tumors. During 2000-2015, when exposure assessment was more precise, the adjusted odds ratio for any cancer diagnosis was 1.77 (1.05-2.98). The summary relative risk estimate for CL in the meta-analysis including four studies was 2.01 (1.25-3.22). CONCLUSIONS: Our study provides weak support for an increased risk of childhood cancers among children living close to petrol stations. A meta-analysis including our study suggests an increased risk for CL.

Birth characteristics and childhood leukemia in Switzerland: a register-based case-control study.

PURPOSE: Initial genetic alterations in the development of childhood leukemia occur in utero or before conception; both genetic and environmental factors are suspected to play a role. We aimed to investigate the associations between childhood leukemia and perinatal characteristics including birth order, birth interval to older siblings, parental age, birth weight, and multiple birth. METHODS: We identified cases diagnosed between 1981 and 2015 and born in Switzerland between 1969 and 2015 from the Swiss Childhood Cancer Registry and randomly sampled five controls per case from national birth records matched on date of birth, sex, and municipality of residence at birth. We used conditional logistic regression to investigate associations between perinatal characteristics and leukemia at ages 0-15 and 0-4 years, and the subtypes acute lymphoblastic leukemia (ALL) and acute myeloid leukemia (AML). RESULTS: The study included 1,403 cases of leukemia. We observed increased risks associated with high birth weight (adjusted OR 1.37, 95% CI 1.12-1.69) and multiple birth (1.89, 1.24-2.86). These associations were similar for ALL and stronger for leukemia at ages 0-4 years. For AML, we observed an increased risk for higher birth order (3.08, 0.43-22.03 for fourth or later born children). We found no associations with other perinatal characteristics. CONCLUSION: This register-based case-control study adds to the existing evidence of a positive association between high birth weight and risk of childhood leukemia. Furthermore, it suggests children from multiple births are at an increased risk of leukemia.

Estimation of SARS-CoV-2 mortality during the early stages of an epidemic: A modeling study in Hubei, China, and six regions in Europe.

BACKGROUND: As of 16 May 2020, more than 4.5 million cases and more than 300,000 deaths from disease caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) have been reported. Reliable estimates of mortality from SARS-CoV-2 infection are essential for understanding clinical prognosis, planning healthcare capacity, and epidemic forecasting. The case-fatality ratio (CFR), calculated from total numbers of reported cases and reported deaths, is the most commonly reported metric, but it can be a misleading measure of overall mortality. The objectives of this study were to (1) simulate the transmission dynamics of SARS-CoV-2 using publicly available surveillance data and (2) infer estimates of SARS-CoV-2 mortality adjusted for biases and examine the CFR, the symptomatic case-fatality ratio (sCFR), and the infection-fatality ratio (IFR) in different geographic locations. METHOD AND FINDINGS: We developed an age-stratified susceptible-exposed-infected-removed (SEIR) compartmental model describing the dynamics of transmission and mortality during the SARS-CoV-2 epidemic. Our model accounts for two biases: preferential ascertainment of severe cases and right-censoring of mortality. We fitted the transmission model to surveillance data from Hubei Province, China, and applied the same model to six regions in Europe: Austria, Bavaria (Germany), Baden-Württemberg (Germany), Lombardy (Italy), Spain, and Switzerland. In Hubei, the baseline estimates were as follows: CFR 2.4% (95% credible interval [CrI] 2.1%-2.8%), sCFR 3.7% (3.2%-4.2%), and IFR 2.9% (2.4%-3.5%). Estimated measures of mortality changed over time. Across the six locations in Europe, estimates of CFR varied widely. Estimates of sCFR and IFR, adjusted for bias, were more similar to each other but still showed some degree of heterogeneity. Estimates of IFR ranged from 0.5% (95% CrI 0.4%-0.6%) in Switzerland to 1.4% (1.1%-1.6%) in Lombardy, Italy. In all locations, mortality increased with age. Among individuals 80 years or older, estimates of the IFR suggest that the proportion of all those infected with SARS-CoV-2 who will die ranges from 20% (95% CrI 16%-26%) in Switzerland to 34% (95% CrI 28%-40%) in Spain. A limitation of the model is that count data by date of onset are required, and these are not available in all countries. CONCLUSIONS: We propose a comprehensive solution to the estimation of SARS-Cov-2 mortality from surveillance data during outbreaks. The CFR is not a good predictor of overall mortality from SARS-CoV-2 and should not be used for evaluation of policy or comparison across settings. Geographic differences in IFR suggest that a single IFR should not be applied to all settings to estimate the total size of the SARS-CoV-2 epidemic in different countries. The sCFR and IFR, adjusted for right-censoring and preferential ascertainment of severe cases, are measures that can be used to improve and monitor clinical and public health strategies to reduce the deaths from SARS-CoV-2 infection.

Bayesian spatial modelling of childhood cancer incidence in Switzerland using exact point data: a nationwide study during 1985-2015.

BACKGROUND: The aetiology of most childhood cancers is largely unknown. Spatially varying environmental factors such as traffic-related air pollution, background radiation and agricultural pesticides might contribute to the development of childhood cancer. This study is the first investigation of the spatial disease mapping of childhood cancers using exact geocodes of place of residence. METHODS: We included 5947 children diagnosed with cancer in Switzerland during 1985-2015 at 0-15 years of age from the Swiss Childhood Cancer Registry. We modelled cancer risk using log-Gaussian Cox processes and indirect standardisation to adjust for age and year of diagnosis. We examined whether the spatial variation of risk can be explained by modelled ambient air concentration of NO2, modelled exposure to background ionising radiation, area-based socio-economic position (SEP), linguistic region, duration in years of general cancer registration in the canton or degree of urbanisation. RESULTS: For all childhood cancers combined, the posterior median relative risk (RR), compared to the national level, varied by location from 0.83 to 1.13 (min to max). Corresponding ranges were 0.96 to 1.09 for leukaemia, 0.90 to 1.13 for lymphoma, and 0.82 to 1.23 for central nervous system (CNS) tumours. The covariates considered explained 72% of the observed spatial variation for all cancers, 81% for leukaemia, 82% for lymphoma and 64% for CNS tumours. There was weak evidence of an association of CNS tumour incidence with modelled exposure to background ionising radiation (RR per SD difference 1.17; 0.98-1.40) and with SEP (1.6; 1.00-1.13). CONCLUSION: Of the investigated diagnostic groups, childhood CNS tumours showed the largest spatial variation. The selected covariates only partially explained the observed variation of CNS tumours suggesting that other environmental factors also play a role.

Spatial clustering of childhood cancers in Switzerland: a nationwide study.

PURPOSE: Childhood cancers are rare and little is known about their etiology. Potential risk factors include environmental exposures that might implicate spatial variation of cancer risk. Previous studies of spatial clustering have mainly focused on childhood leukemia. We investigated spatial clustering of different childhood cancers in Switzerland using exact geocodes of place of residence. METHODS: We included 6,034 cancer cases diagnosed at age 0-15 years during 1985-2015 from the Swiss Childhood Cancer Registry. Age and sex-matched controls (10 per case) were randomly sampled from the national censuses (1990, 2000, 2010). Geocodes of place of residence were available at birth and diagnosis for both cases and controls. We used the difference in k-functions and Cuzick-Edwards test to assess global clustering and Kulldorff's circular scan to detect individual clusters. We also carefully adjusted for multiple testing. RESULTS: After adjusting for multiple testing, we found no evidence of spatial clustering of childhood cancers neither at birth (p = 0.43) nor diagnosis (p = 0.13). Disregarding multiple testing, results of individual tests indicated spatial clustering of all childhood cancers combined (p < 0.01), childhood lymphoma (p = 0.01), due to Hodgkin lymphoma (HL) (p = 0.02) at diagnosis, and embryonal tumors of the central nervous system (CNS) at birth and diagnosis, respectively (p = 0.05 and p = 0.02). CONCLUSIONS: This study provides weak evidence of spatial clustering of childhood cancers. Evidence was strongest for HL and embryonal CNS tumors, adding to the current literature that these cancers cluster in space.

Spatial clustering of childhood leukaemia in Switzerland: A nationwide study.

The aetiology of childhood leukaemia remains largely unknown. Several hypotheses involve environmental exposures that could implicate spatial clustering of cases. The evidence from previous clustering studies is inconclusive. Most of them used areal data and thus had limited spatial resolution. We investigated whether childhood leukaemia tends to cluster in space using exact geocodes of place of residence both at the time of birth or diagnosis. We included 1,871 leukaemia cases diagnosed between 1985 and 2015 at age 0-15 years from the Swiss Childhood Cancer Registry. For each case, we randomly sampled 10 age and sex matched controls from national censuses closest in time. We used the difference of k-functions, Cuzick-Edwards' test and Tango's index for point data to assess spatial clustering and Kulldorff's circular scan to detect clusters. We separately investigated acute lymphoid leukaemia (ALL), acute myeloid leukaemia (AML), different age groups at diagnosis (0-4, 5-15 years) and adjusted for multiple testing. After adjusting for multiple testing, we found no evidence of spatial clustering of childhood leukaemia neither around time of birth (p = 0.52) nor diagnosis (p = 0.51). Individual tests indicated spatial clustering for leukaemia diagnosed at age 5-15 years, p k-functions = 0.05 and p Cuzick-Edwards' = 0.04 and a cluster of ALL cases diagnosed at age 0-4 years in a small rural area (p = 0.05). This study provides little evidence of spatial clustering of childhood leukaemia in Switzerland and highlights the importance of accounting for multiple testing in clustering studies.

A framework for estimating and visualising excess mortality during the COVID-19 pandemic.

COVID-19 related deaths underestimate the pandemic burden on mortality because they suffer from completeness and accuracy issues. Excess mortality is a popular alternative, as it compares observed with expected deaths based on the assumption that the pandemic did not occur. Expected deaths had the pandemic not occurred depend on population trends, temperature, and spatio-temporal patterns. In addition to this, high geographical resolution is required to examine within country trends and the effectiveness of the different public health policies. In this tutorial, we propose a framework using R to estimate and visualise excess mortality at high geographical resolution. We show a case study estimating excess deaths during 2020 in Italy. The proposed framework is fast to implement and allows combining different models and presenting the results in any age, sex, spatial and temporal aggregation desired. This makes it particularly powerful and appealing for online monitoring of the pandemic burden and timely policy making.

Direct and indirect effects of the COVID-19 pandemic on mortality in Switzerland.

The direct and indirect impact of the COVID-19 pandemic on population-level mortality is of concern to public health but challenging to quantify. Using data for 2011-2019, we applied Bayesian models to predict the expected number of deaths in Switzerland and compared them with laboratory-confirmed COVID-19 deaths from February 2020 to April 2022 (study period). We estimated that COVID-19-related mortality was underestimated by a factor of 0.72 (95% credible interval [CrI]: 0.46-0.78). After accounting for COVID-19 deaths, the observed mortality was -4% (95% CrI: -8 to 0) lower than expected. The deficit in mortality was concentrated in age groups 40-59 (-12%, 95%CrI: -19 to -5) and 60-69 (-8%, 95%CrI: -15 to -2). Although COVID-19 control measures may have negative effects, after subtracting COVID-19 deaths, there were fewer deaths in Switzerland during the pandemic than expected, suggesting that any negative effects of control measures were offset by the positive effects. These results have important implications for the ongoing debate about the appropriateness of COVID-19 control measures.

Aircraft noise and cardiovascular morbidity and mortality near Heathrow Airport: A case-crossover study.

Aircraft noise causes annoyance and sleep disturbance and there is some evidence of associations between long-term exposures and cardiovascular disease (CVD). We investigated short-term associations between previous day aircraft noise and cardiovascular events in a population of 6.3 million residing near Heathrow Airport using a case-crossover design and exposure data for different times of day and night. We included all recorded hospitalisations (n = 442,442) and deaths (n = 49,443) in 2014-2018 due to CVD. Conditional logistic regression was used to estimate the ORs and adjusted for NO2 concentration, temperature, and holidays. We estimated an increase in risk for 10 dB increment in noise during the previous evening (Leve OR = 1.007, 95% CI 0.999-1.015), particularly from 22:00-23:00 h (OR = 1.007, 95% CI 1.000-1.013), and the early morning hours 04:30-06:00 h (OR = 1.012, 95% CI 1.002-1.021) for all CVD admissions, but no significant associations with day-time noise. There was effect modification by age-sex, ethnicity, deprivation, and season, and some suggestion that high noise variability at night was associated with higher risks. Our findings are consistent with proposed mechanisms for short-term impacts of aircraft noise at night on CVD from experimental studies, including sleep disturbance, increases in blood pressure and stress hormone levels and impaired endothelial function.

Regional excess mortality during the 2020 COVID-19 pandemic in five European countries.

The impact of the COVID-19 pandemic on excess mortality from all causes in 2020 varied across and within European countries. Using data for 2015-2019, we applied Bayesian spatio-temporal models to quantify the expected weekly deaths at the regional level had the pandemic not occurred in England, Greece, Italy, Spain, and Switzerland. With around 30%, Madrid, Castile-La Mancha, Castile-Leon (Spain) and Lombardia (Italy) were the regions with the highest excess mortality. In England, Greece and Switzerland, the regions most affected were Outer London and the West Midlands (England), Eastern, Western and Central Macedonia (Greece), and Ticino (Switzerland), with 15-20% excess mortality in 2020. Our study highlights the importance of the large transportation hubs for establishing community transmission in the first stages of the pandemic. Here, we show that acting promptly to limit transmission around these hubs is essential to prevent spread to other regions and countries.

Effect of the COVID-19 pandemic on bike-sharing demand and hire time: Evidence from Santander Cycles in London.

The COVID-19 pandemic has been influencing travel behaviour in many urban areas around the world since the beginning of 2020. As a consequence, bike-sharing schemes have been affected-partly due to the change in travel demand and behaviour as well as a shift from public transit. This study estimates the varying effect of the COVID-19 pandemic on the London bike-sharing system (Santander Cycles) over the period March-December 2020. We employed a Bayesian second-order random walk time-series model to account for temporal correlation in the data. We compared the observed number of cycle hires and hire time with their respective counterfactuals (what would have been if the pandemic had not happened) to estimate the magnitude of the change caused by the pandemic. The results indicated that following a reduction in cycle hires in March and April 2020, the demand rebounded from May 2020, remaining in the expected range of what would have been if the pandemic had not occurred. This could indicate the resiliency of Santander Cycles. With respect to hire time, an important increase occurred in April, May, and June 2020, indicating that bikes were hired for longer trips, perhaps partly due to a shift from public transit.

Long-term exposure to air-pollution and COVID-19 mortality in England: A hierarchical spatial analysis.

Recent studies suggested a link between long-term exposure to air-pollution and COVID-19 mortality. However, due to their ecological design based on large spatial units, they neglect the strong localised air-pollution patterns, and potentially lead to inadequate confounding adjustment. We investigated the effect of long-term exposure to NO2 and PM2.5 on COVID-19 mortality in England using high geographical resolution. In this nationwide cross-sectional study in England, we included 38,573 COVID-19 deaths up to June 30, 2020 at the Lower Layer Super Output Area level (n = 32,844 small areas). We retrieved averaged NO2 and PM2.5 concentration during 2014-2018 from the Pollution Climate Mapping. We used Bayesian hierarchical models to quantify the effect of air-pollution while adjusting for a series of confounding and spatial autocorrelation. We find a 0.5% (95% credible interval: -0.2%, 1.2%) and 1.4% (95% CrI: -2.1%, 5.1%) increase in COVID-19 mortality risk for every 1 µg/m3 increase in NO2 and PM2.5 respectively, after adjusting for confounding and spatial autocorrelation. This corresponds to a posterior probability of a positive effect equal to 0.93 and 0.78 respectively. The spatial relative risk at LSOA level revealed strong patterns, similar for the different pollutants. This potentially captures the spread of the disease during the first wave of the epidemic. Our study provides some evidence of an effect of long-term NO2 exposure on COVID-19 mortality, while the effect of PM2.5 remains more uncertain.

Bayesian spatial modelling of terrestrial radiation in Switzerland.

The geographic variation of terrestrial radiation can be exploited in epidemiological studies of the health effects of protracted low-dose exposure. Various methods have been applied to derive maps of this variation. We aimed to construct a map of terrestrial radiation for Switzerland. We used airborne γ-spectrometry measurements to model the ambient dose rates from terrestrial radiation through a Bayesian mixed-effects model and conducted inference using Integrated Nested Laplace Approximation (INLA). We predicted higher levels of ambient dose rates in the alpine regions and Ticino compared with the western and northern parts of Switzerland. We provide a map that can be used for exposure assessment in epidemiological studies and as a baseline map for assessing potential contamination.

Protein isoelectric point distribution in the interactomes across the domains of life.

The distribution of the protein isoelectric point (pI) in the protein-protein interaction (PPI) networks across the domains of life has not been investigated yet. This work attempts to correlate the pI with the number of direct interacting partners in the experimentally supported networks involving 226.085 PPIs from 14 various organisms including human, mouse, yeast, bacteria, viruses and 53.606 virus-host interactions. The results showed that the acidic proteins (pI<3) have the highest average number of interactions in eukaryotes, while in bacteria more neutral proteins. On the contrary, the basic proteins (pI>11) have the lowest average number of interactions in human, mouse, yeast, bacteria and human-viral interactomes and the highest average in intraviral interactomes. We examined the correlation of the pI of the interacting partners by calculating the assortativity index of various PPI networks. We found that the interactions between the acidic, neutral and basic proteins have a fairly random mix, implying weak if any association between the acidic and basic proteins. Furthermore, protein features such as biological function, structurally order and disorder, subcellular localization, and homodimerization were classified according to pI in prokaryote and eukaryote proteomes.

Discrete versus continuous domain models for disease mapping.

The main goal of disease mapping is to estimate disease risk and identify high-risk areas. Such analyses are hampered by the limited geographical resolution of the available data. Typically the available data are counts per spatial unit and the common approach is the Besag-York-Mollié (BYM) model. When precise geocodes are available, it is more natural to use Log-Gaussian Cox processes (LGCPs). In a simulation study mimicking childhood leukaemia incidence using actual residential locations of all children in the canton of Zürich, Switzerland, we compare the ability of these models to recover risk surfaces and identify high-risk areas. We then apply both approaches to actual data on childhood leukaemia incidence in the canton of Zürich during 1985-2015. We found that LGCPs outperform BYM models in almost all scenarios considered. Our findings suggest that there are important gains to be made from the use of LGCPs in spatial epidemiology.

Long-term exposure to air-pollution and COVID-19 mortality in England: a hierarchical spatial analysis.

BACKGROUND: Recent studies suggested a link between long-term exposure to air-pollution and COVID-19 mortality. However, due to their ecological design, based on large spatial units, they neglect the strong localised air-pollution patterns, and potentially lead to inadequate confounding adjustment. We investigated the effect of long-term exposure to NO2 and PM2.5 on COVID-19 deaths up to June 30, 2020 in England using high geographical resolution. METHODS: We included 38 573 COVID-19 deaths up to June 30, 2020 at the Lower Layer Super Output Area level in England (n=32 844 small areas). We retrieved averaged NO2 and PM2.5 concentration during 2014-2018 from the Pollution Climate Mapping. We used Bayesian hierarchical models to quantify the effect of air-pollution while adjusting for a series of confounding and spatial autocorrelation. FINDINGS: We find a 0.5% (95% credible interval: -0.2%-1.2%) and 1.4% (-2.1%-5.1%) increase in COVID-19 mortality rate for every 1µg/m3 increase in NO2 and PM2.5 respectively, after adjusting for confounding and spatial autocorrelation. This corresponds to a posterior probability of a positive effect of 0.93 and 0.78 respectively. The spatial relative risk at LSOA level revealed strong patterns, similar for the different pollutants. This potentially captures the spread of the disease during the first wave of the epidemic. INTERPRETATION: Our study provides some evidence of an effect of long-term NO2 exposure on COVID-19 mortality, while the effect of PM2.5 remains more uncertain. FUNDING: Medical Research Council, Wellcome Trust, Environmental Protection Agency and National Institutes of Health.