Extended two-stage designs for environmental research

The Multi-Country Multi-City Collaborative Research Network: An international research consortium investigating environment, climate, and health

Research on the health risks of environmental factors and climate change requires epidemiological evidence on associated health risks at a global scale. Multi-center studies offer an excellent framework for this purpose, but they present various methodological and logistical problems. This contribution illustrates the experience of the Multi-Country Multi-City Collaborative Research Network, an international collaboration working on a global research program on the associations between environmental stressors, climate, and health in a multi-center setting. The article illustrates the collaborative scheme based on mutual contribution and data and method sharing, describes the collection of a huge multi-location database, summarizes published research findings and future plans, and discusses advantages and limitations. The Multi-Country Multi-City represents an example of a collaborative research framework that has greatly contributed to advance knowledge on the health impacts of climate change and other environmental factors and can be replicated to address other research questions across various research fields.

The impact of climate and demographic changes on future mortality in Brussels, Belgium

OBJECTIVES

City populations are particularly vulnerable to climate change, but it is difficult to reliably estimate the impact on health due to the lack of high-resolution data. We used recently developed regional climate model projections at kilometre resolution combined with demographic projections to estimate the future mortality burden associated with temperatures in the region of Brussels, Belgium.

STUDY DESIGN

The study incorporated a time-series analysis.

METHODS

Based on quasi-Poisson regression with distributed-lag non-linear models for the historical temperature-mortality relationship, we derive the mortality burden for the near (2020-2044) and mid (2045-2069) future and disaggregated the contributions of demographic and climate changes.

RESULTS

The cold-related attributable fraction of deaths is expected to decrease from 6.22% (95% empirical confidence interval: 1.76%; 10.52%) in 1994-2019 to 5.17% (1.08%; 9.09%) in 2045-2069, whereas for heat, this fraction will increase from 1.02% (0.59%; 1.47%) to 1.83% (0.82%; 2.96%), with contributions of both climate and demographic changes. In stratified analyses by age, we found that because of demographic changes, the number of cold-attributable deaths will increase for people aged above 85 years, with 6815 (95% empirical confidence interval: 1424; 12,003) deaths expected in 2045-2069 compared to 5245 (1462; 8867) deaths in 1994-2019. For people aged below 65 years, on the other hand, the number of heat-related deaths will decrease from 456 (265; 658) to 344 (154; 561) deaths.

CONCLUSIONS

Public health policies that especially target the elderly and the summer-time period are needed to limit the impact of climate change on health.

Potential impact of climatic factors on malaria in Rwanda between 2012 and 2021: a time-series analysis

BACKGROUND

Malaria remains an important public health problem, particularly in sub-Saharan Africa. In Rwanda, where malaria ranks among the leading causes of mortality and morbidity, disease transmission is influenced by climatic factors. However, there is a paucity of studies investigating the link between climate change and malaria dynamics, which hinders the development of effective national malaria response strategies. Addressing this critical gap, this study analyses how climatic factors influence malaria transmission across Rwanda, thereby informing tailored interventions and enhancing disease management frameworks.

METHODS

The study analysed the potential impact of temperature and cumulative rainfall on malaria incidence in Rwanda from 2012 to 2021 using meteorological data from the Rwanda Meteorological Agency and malaria case records from the Rwanda Health Management and Information System. The analysis was performed in two stages. First, district-specific generalized linear models with a quasi-Poisson distribution were applied, which were enhanced by distributed lag non-linear models to explore non-linear and lagged effects. Second, random effects multivariate meta-analysis was employed to pool the estimates and to refine them through best linear unbiased predictions.

RESULTS

A 1-month lag with specific temperature and rainfall thresholds influenced malaria incidence across Rwanda. Average temperature of 18.5 °C was associated with higher malaria risk, while temperature above 23.9 °C reduced the risk. Rainfall demonstrated a dual effect on malaria risk: conditions of low (below 73 mm per month) and high (above 223 mm per month) precipitation correlated with lower risk, while moderate rainfall (87 to 223 mm per month) correlated with higher risk. Seasonal patterns showed increased malaria risk during the major rainy season, while the short dry season presented lower risk.

CONCLUSION

The study underscores the influence of temperature and rainfall on malaria transmission in Rwanda and calls for tailored interventions that are specific to location and season. The findings are crucial for informing policy that enhance preparedness and contribute to malaria elimination efforts. Future research should explore additional ecological and socioeconomic factors and their differential contribution to malaria transmission.

Temporal variations in the short-term effects of ambient air pollution on cardiovascular and respiratory mortality: a pooled analysis of 380 urban areas over a 22-year period

BACKGROUND

Ambient air pollution, including particulate matter (such as PM10 and PM2·5) and nitrogen dioxide (NO2), has been linked to increases in mortality. Whether populations' vulnerability to these pollutants has changed over time is unclear, and studies on this topic do not include multicountry analysis. We evaluated whether changes in exposure to air pollutants were associated with changes in mortality effect estimates over time.

METHODS

We extracted cause-specific mortality and air pollution data collected between 1995 and 2016 from the Multi-Country Multi-City (MCC) Collaborative Research Network database. We applied a two-stage approach to analyse the short-term effects of NO2, PM10, and PM2·5 on cause-specific mortality using city-specific time series regression analyses and multilevel random-effects meta-analysis. We assessed changes over time using a longitudinal meta-regression with time as a linear fixed term and explored potential sources of heterogeneity and two-pollutant models.

FINDINGS

Over 21·6 million cardiovascular and 7·7 million respiratory deaths in 380 cities across 24 countries over the study period were included in the analysis. All three air pollutants showed decreasing concentrations over time. The pooled results suggested no significant temporal change in the effect estimates per unit exposure of PM10, PM2·5, or NO2 and mortality. However, the risk of cardiovascular mortality increased from 0·37% (95% CI -0·05 to 0·80) in 1998 to 0·85% (0·55 to 1·16) in 2012 with a 10 μg/m3 increase in PM2·5. Two-pollutant models generally showed similar results to single-pollutant models for PM fractions and indicated temporal differences for NO2.

INTERPRETATION

Although air pollution levels decreased during the study period, the effect sizes per unit increase in air pollution concentration have not changed. This observation might be due to the composition, toxicity, and sources of air pollution, as well as other factors, such as socioeconomic determinants or changes in population distribution and susceptibility.

FUNDING

None.

Compilation of open access time-series datasets for studying temperature-mortality association

In this article, we present a comprehensive compilation of open access daily time-series datasets tailored to assess the temperature-mortality association. The data consists of daily mortality counts and average ambient temperature at various levels of geographic aggregation, including data from four cities, ten regions, and two counties, which have been utilised in previously published studies. These datasets are applicable for time-series regression analysis to estimate location-specific temperature-mortality associations. Additionally, the availability of data from multiple geographical locations enabled the exploration of geographical differences by pooling associations using meta-analysis. This compilation aims to serve as a valuable resource for researchers, educators, and students, facilitating their application of time-series regression modelling for research endeavours and training activities.

Association between precipitation and mortality due to diarrheal diseases by climate zone: A multi-country modeling study

Background

Precipitation could affect the transmission of diarrheal diseases. The diverse precipitation patterns across different climates might influence the degree of diarrheal risk from precipitation. This study determined the associations between precipitation and diarrheal mortality in tropical, temperate, and arid climate regions.

Methods

Daily counts of diarrheal mortality and 28-day cumulative precipitation from 1997 to 2019 were analyzed across 29 locations in eight middle-income countries (Argentina, Brazil, Costa Rica, India, Peru, the Philippines, South Africa, and Thailand). A two-stage approach was employed: the first stage is conditional Poisson regression models for each location, and the second stage is meta-analysis for pooling location-specific coefficients by climate zone.

Results

In tropical climates, higher precipitation increases the risk of diarrheal mortality. Under extremely wet conditions (95th percentile of 28-day cumulative precipitation), diarrheal mortality increased by 17.8% (95% confidence interval [CI] = 10.4%, 25.7%) compared with minimum-risk precipitation. For temperate and arid climates, diarrheal mortality increases in both dry and wet conditions. In extremely dry conditions (fifth percentile of 28-day cumulative precipitation), diarrheal mortality risk increases by 3.8% (95% CI = 1.2%, 6.5%) for temperate and 5.5% (95% CI = 1.0%, 10.2%) for arid climates. Similarly, under extremely wet conditions, diarrheal mortality risk increases by 2.5% (95% CI = -0.1%, 5.1%) for temperate and 4.1% (95% CI = 1.1%, 7.3%) for arid climates.

Conclusions

Associations between precipitation and diarrheal mortality exhibit variations across different climate zones. It is crucial to consider climate-specific variations when generating global projections of future precipitation-related diarrheal mortality.

Ambient fine particulate matter and daily mortality: a comparative analysis of observed and estimated exposure in 347 cities

BACKGROUND

Model-estimated air pollution exposure products have been widely used in epidemiological studies to assess the health risks of particulate matter with diameters of ≤2.5 µm (PM2.5). However, few studies have assessed the disparities in health effects between model-estimated and station-observed PM2.5 exposures.

METHODS

We collected daily all-cause, respiratory and cardiovascular mortality data in 347 cities across 15 countries and regions worldwide based on the Multi-City Multi-Country collaborative research network. The station-observed PM2.5 data were obtained from official monitoring stations. The model-estimated global PM2.5 product was developed using a machine-learning approach. The associations between daily exposure to PM2.5 and mortality were evaluated using a two-stage analytical approach.

RESULTS

We included 15.8 million all-cause, 1.5 million respiratory and 4.5 million cardiovascular deaths from 2000 to 2018. Short-term exposure to PM2.5 was associated with a relative risk increase (RRI) of mortality from both station-observed and model-estimated exposures. Every 10-μg/m3 increase in the 2-day moving average PM2.5 was associated with overall RRIs of 0.67% (95% CI: 0.49 to 0.85), 0.68% (95% CI: -0.03 to 1.39) and 0.45% (95% CI: 0.08 to 0.82) for all-cause, respiratory, and cardiovascular mortality based on station-observed PM2.5 and RRIs of 0.87% (95% CI: 0.68 to 1.06), 0.81% (95% CI: 0.08 to 1.55) and 0.71% (95% CI: 0.32 to 1.09) based on model-estimated exposure, respectively.

CONCLUSIONS

Mortality risks associated with daily PM2.5 exposure were consistent for both station-observed and model-estimated exposures, suggesting the reliability and potential applicability of the global PM2.5 product in epidemiological studies.

Association between PM10 pollution and the hospitalization of chronic obstructive pulmonary disease with comorbidity: evidence in 17 cities of Henan, Central China

Particulate matter (PM10) changes have been confirmed as one of the contributory factors affecting human health, the association between PM10 pollution and the hospitalization of chronic obstructive pulmonary disease (COPD) with comorbidity diseases was rarely reported. The same inpatient more than twice times admissions with COPD illness from January 1, 2016 to December 31, 2021 were identified from hospitals in the 17 cities of Henan, Central China. City-specific associations were firstly estimated using the case time series (CTS) model and then combined to obtain the regional average association. The multivariate meta-analytic model produces pooled estimates of the set of coefficients representing the PM10-COPD hospitalizations association across the 17 cities. Cause-specific hospitalization analyses were performed by COPD patients with different comorbidity combinations. A total of 34,348 elderly (age ≥ 65) subjects were analyzed and with a total of 35,122.35 person-years. These coefficients can be used to compute the linear exposure-response curve expressed as relative risk (RR) in per 10 µg/m3 increase in PM10 at lag03, which was 1.0091 (95% CI 1.0070-1.0112) for COPD with comorbidity, 1.0089 (95% CI 1.0067-1.0110) for COPD with circulatory system diseases, 1.0079 (95% CI 1.0052-1.0105) for COPD with respiratory system diseases, 1.0076 (95% CI 1.0032-1.0121) for COPD with endocrine system diseases, and 1.0087 (95% CI 1.0013-1.0162) for COPD with genitourinary system diseases, respectively. Some heterogeneity was found across cities, with estimates ranging from 1.0227 in the Puyang and Jiaozuo to 1.0053 in Henan Provance, China. The effect of higher PM10, on average, was higher in studies for northern cities, with a steeper raise in risk: per 10 µg/m3 increase in PM10, the RR from 1.0062 (95% CI 1.0030-1.0093) for the 10th percentile of latitude to 1.0124 (95% CI 1.0089-1.0160) for the 90th percentile. Our findings indicated that PM10 exposure may increase the risk of hospitalizations for COPD with comorbidity. Moreover, there might be a higher morbidity risk associated with PM10 in northern latitudes, indicating that stricter air quality standards could potentially reduce PM10-related morbidity among individuals with COPD. These findings have implications for the implementation of effective clean air interventions aligned with national climate policies.

Associations of meteorological factors and dynamics of scrub typhus incidence in South Korea: A nationwide time-series study

Scrub typhus, also known as Tsutsugamushi disease, is a climate-sensitive vector-borne disease that poses a growing public health threat. However, studies on the association between scrub typhus epidemics and meteorological factors in South Korea need to be complemented. Therefore, we aimed to analyze the association among ambient temperature, precipitation, and the incidence of scrub typhus in South Korea. First, we obtained data on the weekly number of scrub typhus cases and concurrent meteorological variables at the city-county level (Si-Gun) in South Korea between 2001 and 2019. Subsequently, a two-stage meta-regression analysis was conducted. In the first stage, we conducted time-series regression analyses using a distributed lag nonlinear model (DLNM) to investigate the association between temperature, precipitation, and scrub typhus incidence at each location. In the second stage, we employed a multivariate meta-regression model to combine the association estimates from all municipalities, considering regional indicators, such as mite species distribution, Normalized Difference Vegetation Index (NDVI), and urban-rural classification. Weekly mean temperature and weekly total precipitation exhibited a reversed U-shaped nonlinear association with the incidence of scrub typhus. The overall cumulative association with scrub typhus incidence peaked at 18.7 C° (with RRs of 9.73, 95% CI: 5.54-17.10) of ambient temperature (reference 9.7 C°) and 162.0 mm (with RRs of 1.87, 95% CI: 1.02-3.83) of precipitation (reference 2.8 mm), respectively. These findings suggest that meteorological factors contribute to scrub typhus epidemics by interacting with vectors, reservoir hosts, and human behaviors. This information serves as a reference for future public health policies and epidemiological research aimed at controlling scrub typhus infections.

Time-stratified case-crossover studies for aggregated data in environmental epidemiology: a tutorial

The case-crossover design is widely used in environmental epidemiology as an effective alternative to the conventional time-series regression design to estimate short-term associations of environmental exposures with a range of acute events. This tutorial illustrates the implementation of the time-stratified case-crossover design to study aggregated health outcomes and environmental exposures, such as particulate matter air pollution, focusing on adjusting covariates and investigating effect modification using conditional Poisson regression. Time-varying confounders can be adjusted directly in the conditional regression model accounting for the adequate lagged exposure-response function. Time-invariant covariates at the subpopulation level require reshaping the typical time-series data set into a long format and conditioning out the covariate in the expanded stratum set. When environmental exposure data are available at geographical units, the stratum set should combine time and spatial dimensions. Moreover, it is possible to examine effect modification using interaction models. The time-stratified case-crossover design offers a flexible framework to properly account for a wide range of covariates in environmental epidemiology studies.

A Nationwide Comparative Analysis of Temperature-Related Mortality and Morbidity in Japan

BACKGROUND

The impact of temperature on morbidity remains largely unknown. Moreover, extensive evidence indicates contrasting patterns between temperature-mortality and temperature-morbidity associations. A nationwide comparison of the impact of temperature on mortality and morbidity in more specific subgroups is necessary to strengthen understanding and help explore underlying mechanisms by identifying susceptible populations.

OBJECTIVE

We performed this study to quantify and compare the impact of temperature on mortality and morbidity in 47 prefectures in Japan.

METHODS

We applied a two-stage time-series design with distributed lag nonlinear models and mixed-effect multivariate meta-analysis to assess the association of temperature with mortality and morbidity by causes (all-cause, circulatory, and respiratory) at prefecture and country levels between 2015 and 2019. Subgroup analysis was conducted by sex, age, and regions.

RESULTS

The patterns and magnitudes of temperature impacts on morbidity and mortality differed. For all-cause outcomes, cold exhibited larger effects on mortality, and heat showed larger effects on morbidity. At specific temperature percentiles, cold (first percentile) was associated with a higher relative risk (RR) of mortality [1.45; 95% confidence interval (CI): 1.39, 1.52] than morbidity (1.33; 95% CI: 1.26, 1.40), as compared to the minimum mortality/morbidity temperature. Heat (99th percentile) was associated with a higher risk of morbidity (1.30; 95% CI: 1.28, 1.33) than mortality (1.04; 95% CI: 1.02, 1.06). For cause-specific diseases, mortality due to circulatory diseases was more susceptible to heat and cold than morbidity. However, for respiratory diseases, both cold and heat showed higher risks for morbidity than mortality. Subgroup analyses suggested varied associations depending on specific outcomes.

DISCUSSION

Distinct patterns were observed for the association of temperature with mortality and morbidity, underlying different mechanisms of temperature on different end points, and the differences in population susceptibility are possible explanations. Future mitigation policies and preventive measures against nonoptimal temperatures should be specific to disease outcomes and targeted at susceptible populations. https://doi.org/10.1289/EHP12854.

Association between environmental factors and dengue incidence in Lao People's Democratic Republic: a nationwide time-series study

BACKGROUND

Dengue fever is a vector-borne disease of global public health concern, with an increasing number of cases and a widening area of endemicity in recent years. Meteorological factors influence dengue transmission. This study aimed to estimate the association between meteorological factors (i.e., temperature and rainfall) and dengue incidence and the effect of altitude on this association in the Lao People's Democratic Republic (Lao PDR).

METHODS

We used weekly dengue incidence and meteorological data, including temperature and rainfall, from 18 jurisdictions in Lao PDR from 2015 to 2019. A two-stage distributed lag nonlinear model with a quasi-Poisson distribution was used to account for the nonlinear and delayed associations between dengue incidence and meteorological variables, adjusting for long-term time trends and autocorrelation.

RESULTS

A total of 55,561 cases were reported in Lao PDR from 2015 to 2019. The cumulative relative risk for the 90th percentile of weekly mean temperature (29 °C) over 22 weeks was estimated at 4.21 (95% confidence interval: 2.00-8.84), relative to the 25th percentile (24 °C). The cumulative relative risk for the weekly total rainfall over 12 weeks peaked at 82 mm (relative risk = 1.76, 95% confidence interval: 0.91-3.40) relative to no rain. However, the risk decreased significantly when heavy rain exceeded 200 mm. We found no evidence that altitude modified these associations.

CONCLUSIONS

We found a lagged nonlinear relationship between meteorological factors and dengue incidence in Lao PDR. These findings can be used to develop climate-based early warning systems and provide insights for improving vector control in the country.

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.

Inverted U-shaped association between bacillary dysentery and temperature: A new finding using a novel two-stage strategy in multi-region studies

BACKGROUND

Bacillary dysentery (BD) has brought a significant public health concern in China. Temperature is one of the main factors affecting BD incidence. Due to the largely different temperature ranges between regions, the classic multi-region time series studies could only explore the relative temperature-BD association and showed that BD incidence is positively associated with relative temperature (i.e., temperature percentile), which does not conform to the laboratory knowledge that both high and low temperature interfere with the survival of bacteria. The association on relative temperature scale also limits the intuition of epidemiological meanings.

METHODS

A novel two-stage strategy was proposed to investigate the association between monthly temperature and BD incidence on the original temperature scale in 31 provinces, China. In the first stage, truncated polynomial splines, as the substitute of the common natural splines or B-splines in generalized additive models, were used to characterize the temperature-BD association on the original temperature scale in each province. In the second stage, a multivariate meta-analysis compatible with missing values was used to pool the associations. The classic strategy based on relative temperature was used as a reference.

RESULTS

The average temperature-BD association presented a U-inverted shape, but not a monotonically increasing shape obtained using the classic strategy. This inverted U-shaped association conforms more to the laboratory knowledge and the original-scale association also provided an intuitive perspective and an easily explanatory result. Another advantage is that the novel strategy can extrapolate the province-specific association outside the observed temperature ranges by utilizing information from other provinces, which is meaningful considering the frequent incidences of extreme temperatures.

CONCLUSIONS

The association between temperature and BD incidence presented a U-inverted shape. The proposed strategy can efficiently characterize the association between exposure and outcome on original scale in a multi-region study with largely different exposure ranges.

Mortality risks associated with floods in 761 communities worldwide: time series study

OBJECTIVE

To evaluate lag-response associations and effect modifications of exposure to floods with risks of all cause, cardiovascular, and respiratory mortality on a global scale.

DESIGN

Time series study.

SETTING

761 communities in 35 countries or territories with at least one flood event during the study period.

PARTICIPANTS

Multi-Country Multi-City Collaborative Research Network database, Australian Cause of Death Unit Record File, New Zealand Integrated Data Infrastructure, and the International Network for the Demographic Evaluation of Populations and their Health Network database.

MAIN OUTCOME MEASURES

The main outcome was daily counts of deaths. An estimation for the lag-response association between flood and daily mortality risk was modelled, and the relative risks over the lag period were cumulated to calculate overall effects. Attributable fractions of mortality due to floods were further calculated. A quasi-Poisson model with a distributed lag non-linear function was used to examine how daily death risk was associated with flooded days in each community, and then the community specific associations were pooled using random effects multivariate meta-analyses. Flooded days were defined as days from the start date to the end date of flood events.

RESULTS

A total of 47.6 million all cause deaths, 11.1 million cardiovascular deaths, and 4.9 million respiratory deaths were analysed. Over the 761 communities, mortality risks increased and persisted for up to 60 days (50 days for cardiovascular mortality) after a flooded day. The cumulative relative risks for all cause, cardiovascular, and respiratory mortality were 1.021 (95% confidence interval 1.006 to 1.036), 1.026 (1.005 to 1.047), and 1.049 (1.008 to 1.092), respectively. The associations varied across countries or territories and regions. The flood-mortality associations appeared to be modified by climate type and were stronger in low income countries and in populations with a low human development index or high proportion of older people. In communities impacted by flood, up to 0.10% of all cause deaths, 0.18% of cardiovascular deaths, and 0.41% of respiratory deaths were attributed to floods.

CONCLUSIONS

This study found that the risks of all cause, cardiovascular, and respiratory mortality increased for up to 60 days after exposure to flood and the associations could vary by local climate type, socioeconomic status, and older age.

Association between cannabis use and symptom dimensions in schizophrenia spectrum disorders: an individual participant data meta-analysis on 3053 individuals

Background

The association between cannabis use and positive symptoms in schizophrenia spectrum disorders is well documented, especially via meta-analyses. Yet, findings are inconsistent regarding negative symptoms, while other dimensions such as disorganization, depression, and excitement, have not been investigated. In addition, meta-analyses use aggregated data discarding important confounding variables which is a source of bias.

Methods

PubMed, ScienceDirect and PsycINFO were used to search for publications from inception to September 27, 2022. We contacted the authors of relevant studies to extract raw datasets and perform an Individual Participant Data meta-analysis (IPDMA). Inclusion criteria were: psychopathology of individuals with schizophrenia spectrum disorders assessed by the Positive and Negative Syndrome Scale (PANSS); cannabis-users had to either have a diagnosis of cannabis use disorder or use cannabis at least twice a week. The main outcomes were the PANSS subscores extracted via the 3-factor (positive, negative and general) and 5-factor (positive, negative, disorganization, depression, excitement) structures. Preregistration is accessible via Prospero: ID CRD42022329172.

Findings

Among the 1149 identified studies, 65 were eligible and 21 datasets were shared, totaling 3677 IPD and 3053 complete cases. The adjusted multivariate analysis revealed that relative to non-use, cannabis use was associated with higher severity of positive dimension (3-factor: Adjusted Mean Difference, aMD = 0.34, 95% Confidence Interval, CI = [0.03; 0.66]; 5-factor: aMD = 0.38, 95% CI = [0.08; 0.63]), lower severity of negative dimension (3-factor: aMD = -0.49, 95% CI [-0.90; -0.09]; 5-factor: aMD = -0.50, 95% CI = [-0.91; -0.08]), higher severity of excitement dimension (aMD = 0.16, 95% CI = [0.03; 0.28]). No association was found between cannabis use and disorganization (aMD = -0.13, 95% CI = [-0.42; 0.17]) or depression (aMD = -0.14, 95% CI = [-0.34; 0.06]).

Interpretation

No causal relationship can be inferred from the current results. The findings could be in favor of both a detrimental and beneficial effect of cannabis on positive and negative symptoms, respectively. Longitudinal designs are needed to understand the role of cannabis is this association. The reported effect sizes are small and CIs are wide, the interpretation of findings should be taken with caution.

Funding

This research did not receive any specific grant or funding. Primary financial support for authors was provided by Le Vinatier Psychiatric Hospital.

Assessing heat effects on respiratory mortality and location characteristics as modifiers of heat effects at a small area scale in Central-Northern Europe

Background

Heat effects on respiratory mortality are known, mostly from time-series studies of city-wide data. A limited number of studies have been conducted at the national level or covering non-urban areas. Effect modification by area-level factors has not been extensively investigated. Our study assessed the heat effects on respiratory mortality at a small administrative area level in Norway, Germany, and England and Wales, in the warm period (May-September) within 1996-2018. Also, we examined possible effect modification by several area-level characteristics in the framework of the EU-Horizon2020 EXHAUSTION project.

Methods

Daily respiratory mortality counts and modeled air temperature data were collected for Norway, Germany, and England and Wales at a small administrative area level. The temperature-mortality association was assessed by small area-specific Poisson regression allowing for overdispersion, using distributed lag non-linear models. Estimates were pooled at the national level and overall using a random-effect meta-analysis. Age- and sex-specific models were also applied. A multilevel random-effects model was applied to investigate the modification of the heat effects by area-level factors.

Results

A rise in temperature from the 75th to 99th percentile was associated with a 27% (95% confidence interval [CI] = 19%, 34%) increase in respiratory mortality, with higher effects for females. Increased population density and PM2.5 concentrations were associated with stronger heat effects on mortality.

Conclusions

Our study strengthens the evidence of adverse heat effects on respiratory mortality in Northern Europe by identifying vulnerable subgroups and subregions. This may contribute to the development of targeted policies for adaptation to climate change.

Nationwide projections of heat- and cold-related mortality impacts under various climate change and population development scenarios in Switzerland

Climate change and progressive population development (i.e., ageing and changes in population size) are altering the temporal patterns of temperature-related mortality in Switzerland. However, limited evidence exists on how current trends in heat- and cold-related mortality would evolve in future decades under composite scenarios of global warming and population development. Moreover, the contribution of these drivers to future mortality impacts is not well-understood. Therefore, we aimed to project heat- and cold-related mortality in Switzerland under various combinations of emission and population development scenarios and to disentangle the contribution of each of these two drivers using high-resolution mortality and temperature data. We combined age-specific (<75 and ⩾75 years) temperature-mortality associations in each district in Switzerland (1990-2010), estimated through a two-stage time series analysis, with 2 km downscaled CMIP5 temperature data and population and mortality rate projections under two scenarios: RCP4.5/SSP2 and RCP8.5/SSP5. We derived heat and cold-related mortality for different warming targets (1.5 °C, 2.0 °C and 3.0 °C) using different emission and population development scenarios and compared this to the baseline period (1990-2010). Heat-related mortality is projected to increase from 312 (116; 510) in the 1990-2010 period to 1274 (537; 2284) annual deaths under 2.0 °C of warming (RCP4.5/SSP2) and to 1871 (791; 3284) under 3.0 °C of warming (RCP8.5/SSP5). Cold-related mortality will substantially increase from 4069 (1898; 6016) to 6558 (3223; 9589) annual deaths under 2.0 °C (RCP4.5/SSP2) and to 5997 (2951; 8759) under 3.0 °C (RCP8.5/SSP5). Moreover, while the increase in cold-related mortality is solely driven by population development, for heat, both components (i.e., changes in climate and population) have a similar contribution of around 50% to the projected heat-related mortality trends. In conclusion, our findings suggest that both heat- and cold-related mortality will substantially increase under all scenarios of climate change and population development in Switzerland. Population development will lead to an increase in cold-related mortality despite the decrease in cold temperature under warmer scenarios. Whereas the combination of the progressive warming of the climate and population development will substantially increase and exacerbate the total temperature-related mortality burden in Switzerland.

Impact of ultrafine particles and total particle number concentration on five cause-specific hospital admission endpoints in three German cities

INTRODUCTION

Numerous studies have shown associations between daily concentrations of fine particles (e.g., particulate matter with an aerodynamic diameter ≤2.5 µm; PM2.5) and morbidity. However, evidence for ultrafine particles (UFP; particles with an aerodynamic diameter of 10-100 nm) remains conflicting. Therefore, we aimed to examine the short-term associations of UFP with five cause-specific hospital admission endpoints for Leipzig, Dresden, and Augsburg, Germany.

MATERIAL AND METHODS

We obtained daily counts of (cause-specific) cardiorespiratory hospital admissions between 2010 and 2017. Daily average concentrations of UFP, total particle number (PNC; 10-800 nm), and black carbon (BC) were measured at six sites; PM2.5 and nitrogen dioxide (NO2) were obtained from monitoring networks. We assessed immediate (lag 0-1), delayed (lag 2-4, lag 5-7), and cumulative (lag 0-7) effects by applying station-specific confounder-adjusted Poisson regression models. We then used a novel multi-level meta-analytical method to obtain pooled risk estimates. Finally, we performed two-pollutant models to investigate interdependencies between pollutants and examined possible effect modification by age, sex, and season.

RESULTS

UFP showed a delayed (lag 2-4) increase in respiratory hospital admissions of 0.69% [95% confidence interval (CI): -0.28%; 1.67%]. For other hospital admission endpoints, we found only suggestive results. Larger particle size fractions, such as accumulation mode particles (particles with an aerodynamic diameter of 100-800 nm), generally showed stronger effects (respiratory hospital admissions & lag 2-4: 1.55% [95% CI: 0.86%; 2.25%]). PM2.5 showed the most consistent associations for (cardio-)respiratory hospital admissions, whereas NO2 did not show any associations. Two-pollutant models showed independent effects of PM2.5 and BC. Moreover, higher risks have been observed for children.

CONCLUSIONS

We observed clear associations with PM2.5 but UFP or PNC did not show a clear association across different exposure windows and cause-specific hospital admissions. Further multi-center studies are needed using harmonized UFP measurements to draw definite conclusions on the health effects of UFP.

Exploring vulnerability to heat and cold across urban and rural populations in Switzerland

Heat- and cold-related mortality risks are highly variable across different geographies, suggesting a differential distribution of vulnerability factors between and within countries, which could partly be driven by urban-to-rural disparities. Identifying these drivers of risk is crucial to characterize local vulnerability and design tailored public health interventions to improve adaptation of populations to climate change. We aimed to assess how heat- and cold-mortality risks change across urban, peri-urban and rural areas in Switzerland and to identify and compare the factors associated with increased vulnerability within and between different area typologies. We estimated the heat- and cold-related mortality association using the case time-series design and distributed lag non-linear models over daily mean temperature and all-cause mortality series between 1990-2017 in each municipality in Switzerland. Then, through multivariate meta-regression, we derived pooled heat and cold-mortality associations by typology (i.e. urban/rural/peri-urban) and assessed potential vulnerability factors among a wealth of demographic, socioeconomic, topographic, climatic, land use and other environmental data. Urban clusters reported larger pooled heat-related mortality risk (at 99th percentile, vs. temperature of minimum mortality (MMT)) (relative risk=1.17(95%CI:1.10;1.24, vs peri-urban 1.03(1.00;1.06), and rural 1.03 (0.99;1.08)), but similar cold-mortality risk (at 1st percentile, vs. MMT) (1.35(1.28;1.43), vs rural 1.28(1.14;1.44) and peri-urban 1.39 (1.27-1.53)) clusters. We found different sets of vulnerability factors explaining the differential risk patterns across typologies. In urban clusters, mainly environmental factors (i.e. PM2.5) drove differences in heat-mortality association, while for peri-urban/rural clusters socio-economic variables were also important. For cold, socio-economic variables drove changes in vulnerability across all typologies, while environmental factors and ageing were other important drivers of larger vulnerability in peri-urban/rural clusters, with heterogeneity in the direction of the association. Our findings suggest that urban populations in Switzerland may be more vulnerable to heat, compared to rural locations, and different sets of vulnerability factors may drive these associations in each typology. Thus, future public health adaptation strategies should consider local and more tailored interventions rather than a one-size fits all approach. size fits all approach.

Impact of Ambient Ultrafine Particles on Cause-Specific Mortality in Three German Cities

Rationale: Exposure to ambient air pollution has been associated with adverse effects on morbidity and mortality. However, the evidence for ultrafine particles (UFPs; 10-100 nm) based on epidemiological studies remains scarce and inconsistent. Objectives: We examined associations between short-term exposures to UFPs and total particle number concentrations (PNCs; 10-800 nm) and cause-specific mortality in three German cities: Dresden, Leipzig, and Augsburg. Methods: We obtained daily counts of natural, cardiovascular, and respiratory mortality between 2010 and 2017. UFPs and PNCs were measured at six sites, and measurements of fine particulate matter (PM2.5; ⩽2.5 μm in aerodynamic diameter) and nitrogen dioxide were collected from routine monitoring. We applied station-specific confounder-adjusted Poisson regression models. We investigated air pollutant effects at aggregated lags (0-1, 2-4, 5-7, and 0-7 d after UFP exposure) and used a novel multilevel meta-analytical method to pool the results. Additionally, we assessed interdependencies between pollutants using two-pollutant models. Measurements and Main Results: For respiratory mortality, we found a delayed increase in relative risk of 4.46% (95% confidence interval, 1.52 to 7.48%) per 3,223-particles/cm3 increment 5-7 days after UFP exposure. Effects for PNCs showed smaller but comparable estimates consistent with the observation that the smallest UFP fractions showed the largest effects. No clear associations were found for cardiovascular or natural mortality. UFP effects were independent of PM2.5 in two-pollutant models. Conclusions: We found delayed effects for respiratory mortality within 1 week after exposure to UFPs and PNCs but no associations for natural or cardiovascular mortality. This finding adds to the evidence on the independent health effects of UFPs.

Short-Term Association between Sulfur Dioxide and Mortality: A Multicountry Analysis in 399 Cities

BACKGROUND

Epidemiological evidence on the health risks of sulfur dioxide (SO 2 ) is more limited compared with other pollutants, and doubts remain on several aspects, such as the form of the exposure-response relationship, the potential role of copollutants, as well as the actual risk at low concentrations and possible temporal variation in risks.

OBJECTIVES

Our aim was to assess the short-term association between exposure to SO 2 and daily mortality in a large multilocation data set, using advanced study designs and statistical techniques.

METHODS

The analysis included 43,729,018 deaths that occurred in 399 cities within 23 countries between 1980 and 2018. A two-stage design was applied to assess the association between the daily concentration of SO 2 and mortality counts, including first-stage time-series regressions and second-stage multilevel random-effect meta-analyses. Secondary analyses assessed the exposure-response shape and the lag structure using spline terms and distributed lag models, respectively, and temporal variations in risk using a longitudinal meta-regression. Bi-pollutant models were applied to examine confounding effects of particulate matter with an aerodynamic diameter of ≤ 10 μ m (PM 10 ) and 2.5 μ m (PM 2.5 ), ozone, nitrogen dioxide, and carbon monoxide. Associations were reported as relative risks (RRs) and fractions of excess deaths.

RESULTS

The average daily concentration of SO 2 across the 399 cities was 11 . 7  μ g / m 3 , with 4.7% of days above the World Health Organization (WHO) guideline limit (40   μ g / m 3 , 24-h average), although the exceedances occurred predominantly in specific locations. Exposure levels decreased considerably during the study period, from an average concentration of 19.0   μ g / m 3 in 1980-1989 to 6.3   μ g / m 3 in 2010-2018. For all locations combined, a 10 - μ g / m 3 increase in daily SO 2 was associated with an RR of mortality of 1.0045 [95% confidence interval (CI): 1.0019, 1.0070], with the risk being stable over time but with substantial between-country heterogeneity. Short-term exposure to SO 2 was associated with an excess mortality fraction of 0.50% [95% empirical CI (eCI): 0.42%, 0.57%] in the 399 cities, although decreasing from 0.74% (0.61%, 0.85%) in 1980-1989 to 0.37% (0.27%, 0.47%) in 2010-2018. There was some evidence of nonlinearity, with a steep exposure-response relationship at low concentrations and the risk attenuating at higher levels. The relevant lag window was 0-3 d. Significant positive associations remained after controlling for other pollutants.

DISCUSSION

The analysis revealed independent mortality risks associated with short-term exposure to SO 2 , with no evidence of a threshold. Levels below the current WHO guidelines for 24-h averages were still associated with substantial excess mortality, indicating the potential benefits of stricter air quality standards. https://doi.org/10.1289/EHP11112.

Temporal variation of the temperature-mortality association in Spain: a nationwide analysis

BACKGROUND

Although adaptation to continuously rising ambient temperatures is an emerging topic and has been widely studied at a global scale, detailed analysis of the joint indicators for long-term adaptation in Spain are scarce. This study aims to explore temporal variations of the minimum mortality temperature and mortality burden from heat and cold between 1979 and 2018.

METHODS

We collected individual all-cause mortality and climate reanalysis data for 4 decades at a daily time step. To estimate the temperature-mortality association for each decade, we fitted a quasi-Poisson time-series regression model using a distributed lag non-linear model with 21 days of lag, controlling for trends and day of the week. We also calculated attributable mortality fractions by age and sex for heat and cold, defined as temperatures above and below the optimum temperature, which corresponds to the minimum mortality in each period.

RESULTS

We analysed over 14 million deaths registered in Spain between 1979 and 2018. The optimum temperature estimated at a nationwide scale declined from 21 °C in 1979-1988 to 16 °C in 1999-2008, and raised to 18 °C in 2009-2018. The mortality burden from moderate cold showed a 3-fold reduction down to 2.4% in 2009-2018. Since 1988-1999, the mortality risk attributable to moderate (extreme) heat reduced from 0.9% (0.8%) to 0.6% (0.5%). The mortality risk due to heat in women was almost 2 times larger than in men, and did not decrease over time.

CONCLUSION

Despite the progressively warmer temperatures in Spain, we observed a persistent flattening of the exposure-response curves, which marked an expansion of the uncertainty range of the optimal temperatures. Adaptation has been produced to some extent in a non-uniform manner with a substantial decrease in cold-related mortality, while for heat it became more apparent in the most recent decade only.