Time series analysis on the health effects of temperature: advancements and limitations

Temporal trends in temperature-related mortality and evidence for maladaptation to heat and cold in the Eastern Mediterranean region

The eastern Mediterranean region is characterized by rising temperature trends exceeding the corresponding global averages and is considered a climate change hot-spot. Although previous studies have thoroughly investigated the impact of extreme heat and cold on human mortality and morbidity, both for the current and future climate change scenarios, the temporal trends in temperature-related mortality or the potential historical adaptation to heat and cold extremes has never been studied in this region. This study focuses on cardiovascular mortality and assesses the temporal evolution of the Minimum Mortality Temperature (MMT), as well as the disease-specific cold- and heat-attributable fraction of mortality in three typical eastern Mediterranean environments (Athens, Thessaloniki and Cyprus). Data on daily cardiovascular mortality (ICD-10 code: I00-I99) and meteorological parameters were available between 1999 and 2019 for Athens, 1999 to 2018 for Thessaloniki and 2004 to 2019 for Cyprus. Estimation of cardiovascular MMT and mortality fractions relied on time-series Poisson regressions with distributed lag nonlinear models (DLNM) controlling for seasonal and long-term trends, performed over a series of rolling sub-periods at each site. The results indicated that in Athens, the MMT decreased from 23 °C (67.5th percentile) in 1999-2007 to 21.8 °C (62nd percentile) in 2011-2019, while in Cyprus the MMT decreased from 26.3 °C (79th percentile) in 2004-2012 to 23.9 °C (66.5th percentile) in 2011-2019. In Thessaloniki, the decrease in MMT was rather negligible. In all regions under study, the fractions of mortality attributed to both cold and heat followed an upward trend throughout the years. In conclusion, the demonstrated increase in cold attributable fraction and the decreasing temporal trend of MMT across the examined sites are suggestive of maladaptation to extreme temperatures in regions with warm climate and highlight the need for relevant public health policies and interventions.

The trends of non-accidental mortality burden attributed to compound hot-dry events in China and its provinces in a global warming world

BACKGROUND

Global warming has provoked more co-occurrence of hot extreme and dry extreme, namely compound hot-dry events (CHDEs). However, their health impacts have seldom been investigated. This study aimed to characterize CHDEs and assess its mortality burden in China from 1990 to 2100.

METHODS

CHDEs were defined as a day when daily maximum temperature > its 90th percentile and Standardized Precipitation Index < its 50th percentile. A two-stage approach, including a distributed lag nonlinear model (DLNM) and a multivariate meta-analysis, was used to estimate exposure-response associations of CHDEs with mortality in 358 counties/districts during 2006-2017 in China, which was then applied to assess the national mortality burden attributable to CHDEs from 1990 to 2100.

FINDINGS

We observed a significant increasing trend of CHDEs in China until mid-21st century, and then flatted, while the duration and intensity of CHDEs continuously increased across the 21st century. CHDEs were much riskier (ER=17.82 %, 95 %CI: 14.17 %-21.60 %) than independent hot events (ER=5.86 %,95 %CI: -0.04 %,12.45 %) or dry events (ER=0.07 %,95 %CI: -1.22 %, 1.38 %), and there was significantly additive interaction between hot events and dry events (AP=0.10,95 %CI: 0.04, 0.16). Females (ER=24.28 %, 95 %CI: 19.21 %-29.56 %), the elderly (ER=23.28 %, 95 %CI: 18.23 %-28.55 %), and people living in humid area (ER=18.98 %, 95 %CI: 15.08 %-23.02 %) had higher mortality risks than their counterparts. Mortality burden attributed to CHDEs significantly increased during historical observation and became stable since mid-21st century in China.

INTERPRETATION

CHDEs would significantly increase mortality with higher risk for females, the elderly and people living in humid areas. Mortality burden has significantly increased during historical observation and will keep relatively steady since mid-21st century.

Temperature frequency and mortality: Assessing adaptation to local temperature

Assessing the association between temperature frequency and mortality can provide insights into human adaptation to local ambient temperatures. We collected daily time-series data on mortality and temperature from 757 locations in 47 countries/regions during 1979-2020. We used a two-stage time series design to assess the association between temperature frequency and all-cause mortality. The results were pooled at the national, regional, and global levels. We observed a consistent decrease in the risk of mortality as the normalized frequency of temperature increases across the globe. The average increase in mortality risk comparing the 10th to 100th percentile of normalized frequency was 13.03% (95% CI: 12.17-13.91), with substantial regional differences (from 4.56% in Australia and New Zealand to 33.06% in South Europe). The highest increase in mortality was observed for high-income countries (13.58%, 95% CI: 12.56-14.61), followed by lower-middle-income countries (12.34%, 95% CI: 9.27-15.51). This study observed a declining risk of mortality associated with higher temperature frequency. Our findings suggest that populations can adapt to their local climate with frequent exposure, with the adapting ability varying geographically due to differences in climatic and socioeconomic characteristics.

A nationwide time-series analysis for short-term effects of ambient temperature on violent crime in South Korea

Psychological theories on heat-aggression relationship have existed for decades and recent models suggest climate change will increase violence through varying pathways. Although observational studies have examined the impact of temperature on violent crime, the evidence for associations is primarily limited to coarse temporal resolution of weather and crime (e.g., yearly/monthly) and results from a few Western communities, warranting studies based on higher temporal resolution data of modern systemic crime statistics for various regions. This observational study examined short-term temperature impacts on violent crime using national crime data for the warm months (Jun.-Sep.) across South Korea (2016-2020). Distributed lag non-linear models assessed relative risks (RRs) of daily violent crime counts at the 70th, 90th, and 99th summer temperature percentiles compared to the reference temperature (10th percentile), with adjustments for long-term trends, seasonality, weather, and air pollution. Results indicate potentially non-linear relationships between daily summer temperature (lag0-lag10) and violent crime counts. Violent crimes consistently increased from the lowest temperature and showed the highest risk at the 70th temperature (~ 28.0 °C). The RR at the 70th and 90th percentiles of daily mean temperature (lag0-lag10), compared to the reference, was 1.11 (95% CI 1.09, 1.15) and 1.04 (95% CI 1.01, 1.07), indicating significant associations. Stratified analysis showed significant increases in assault and domestic violence for increases in temperature. The lagged effects, the influences of heat on subsequent crime incidence, did not persist 21 days after the exposure, possibly due to the displacement phenomenon. We found curvilinear exposure-response relationships, which provide empirical evidence to support the psychological theories for heat and violence. Lower public safety through increased violent crime may be an additional public health harm of climate change.

Nonoptimum Temperatures Are More Closely Associated With Fatal Myocardial Infarction Than With Nonfatal Events

BACKGROUND

Ambient temperatures trigger hospitalisation, mortality, and emergency department visits for myocardial infarction (MI). However, nonoptimum temperature-related risks of fatal and nonfatal MI have not yet been compared.

METHODS

From 2007 to 2019, 416,894 MI events (233,071 fatal and 183,823 nonfatal) were identified in Beijing, China. A time-series analysis with a distributed-lag nonlinear model was used to compare the relative and population-attributable risks of fatal and nonfatal MI associated with nonoptimum temperatures.

RESULTS

The reference was the optimum temperature of 24.3°C. For single-lag effects, cold (-5.2°C) and heat (29.6°C) effects had associations that persisted for more days for fatal MI than for nonfatal MI. For cumulative-lag effects over 0 to 21 days, cold effects were higher for fatal MI (relative risk [RR] 1.99, 95% confidence interval [CI] 1.68-2.35) than for nonfatal MI (RR 1.60, 95% CI 1.32-1.94) with a P value for difference in effect sizes of 0.048. In addition, heat effects were higher for fatal MI (RR 1.33, 95% CI 1.24-1.44) than for nonfatal MI (RR 0.99, 95% CI 0.91-1.08) with a P value for difference in effect sizes of 0.002. The attributable fraction of nonoptimum temperatures was higher for fatal MI (25.6%, 95% CI 19.7%-30.6%) than for nonfatal MI (19.1%, 95% CI 12.1%-25.0%).

CONCLUSIONS

Fatal MI was more closely associated with nonoptimum temperatures than nonfatal MI, as evidenced by single-lag effects that have associations which persisted for more days, higher cumulative-lag effects, and higher attributable risks for fatal MI. Strategies are needed to mitigate the adverse effects of nonoptimum temperatures.

The impact of EV71 vaccination program on hand, foot and mouth disease in Zhejiang Province, China: A negative control study

Objective

To estimate the potential causal impact of Enterovirus A71 (EV71) vaccination program on the reduction of EV71-infected hand, foot, and mouth disease (HFMD) in Zhejiang Province.

Methods

We utilized the longitudinal surveillance dataset of HFMD and EV71 vaccination in Zhejiang Province during 2010-2019. We estimated vaccine efficacy using a Bayesian structured time series (BSTS) model, and employed a negative control outcome (NCO) model to detect unmeasured confounding and reveal potential causal association.

Results

We estimated that 20,132 EV71 cases (95% CI: 16,733, 23,532) were prevented by vaccination program during 2017-2019, corresponding to a reduction of 29% (95% CI: 24%, 34%). The effectiveness of vaccination increased annually, with reductions of 11% (95% CI: 6%, 16%) in 2017 and 66% (95% CI: 61%, 71%) in 2019. Children under 5 years old obtained greater benefits compared to those over 5 years. Cities with higher vaccination coverage experienced a sharper EV71 reduction compared to those with lower coverage. The NCO model detected no confounding factors in the association between vaccination and EV71 cases reduction.

Conclusions

This study suggested a potential causal effect of the EV71 vaccination, highlighting the importance of achieving higher vaccine coverage to control the HFMD.

Influence of temperature on the risk of gestational diabetes mellitus and hypertension in different pregnancy trimesters

Previous studies have proved that exposure to extreme temperature in specific windows of pregnancy could cause some complications, such as pregnancy induced hypertension (PIH) and gestational diabetes mellitus (GDM), but differences in the effect of extreme temperature on the 2 complications are rarely studied. We carried a retrospective study on the impact of temperature on GDM/PIH in different trimesters based on data from a maternal and child health center in Beijing, China. Ambient temperatures (°C) were obtained from the China Meteorological Administration from January 1st, 2013 to May 15th, 2018. We use distributed lag non-linear models (DLNMs) combined with logistic regression to calculate the lag exposure-response relationships between the temperature and GDM/PIH from 1st to 24th/20th weeks of pregnancy. In both first and second trimesters, the risk of GDM was increased in summer with high temperatures; in second trimester, the risk of GDM increased in winter with low temperatures. In first half of pregnancy, risk of PIH was decreased in winter with low temperatures. These findings can provide the guideline for preventing the GDM and PIH induced by extreme temperature during pregnancy.

Confounding by Conception Seasonality in Studies of Temperature and Preterm Birth: A Simulation Study

BACKGROUND

Seasonal patterns of conception may confound acute associations between birth outcomes and seasonally varying exposures. We aim to evaluate four epidemiologic designs (time-stratified case-crossover, time-series, pair-matched case-control, and time-to-event) commonly used to study acute associations between ambient temperature and preterm births.

METHODS

We conducted simulations assuming no effect of temperature on preterm birth. We generated pseudo-birth data from the observed seasonal patterns of birth in the United States and analyzed them in relation to observed temperatures using design-specific seasonality adjustments.

RESULTS

Using the case-crossover approach (time-stratified by calendar month), we observed a bias (among 1,000 replicates) = 0.016 (Monte-Carlo standard error 95% CI: 0.015-0.018) in the regression coefficient for every 10°C increase in mean temperature in the warm season (May-September). Unbiased estimates obtained using the time-series approach required accounting for both the pregnancies-at-risk and their weighted probability of birth. Notably, adding the daily weighted probability of birth from the time-series models to the case-crossover models corrected the bias in the case-crossover approach. In the pair-matched case-control design, where the exposure period was matched on gestational window, we observed no bias. The time-to-event approach was also unbiased but was more computationally intensive than others.

CONCLUSIONS

Most designs can be implemented in a way that yields estimates unbiased by conception seasonality. The time-stratified case-crossover design exhibited a small positive bias, which could contribute to, but not fully explain, previously reported associations.

Effects of High Temperature on COVID-19 Deaths in U.S. Counties

The United States of America (USA) was afflicted by extreme heat in the summer of 2021 and some states experienced a record-hot or top-10 hottest summer. Meanwhile, the United States was also one of the countries impacted most by the coronavirus disease 2019 (COVID-19) pandemic. Growing numbers of studies have revealed that meteorological factors such as temperature may influence the number of confirmed COVID-19 cases and deaths. However, the associations between temperature and COVID-19 severity differ in various study areas and periods, especially in periods of high temperatures. Here we choose 119 US counties with large counts of COVID-19 deaths during the summer of 2021 to examine the relationship between COVID-19 deaths and temperature by applying a two-stage epidemiological analytical approach. We also calculate the years of life lost (YLL) owing to COVID-19 and the corresponding values attributable to high temperature exposure. The daily mean temperature is approximately positively correlated with COVID-19 deaths nationwide, with a relative risk of 1.108 (95% confidence interval: 1.046, 1.173) in the 90th percentile of the mean temperature distribution compared with the median temperature. In addition, 0.02 YLL per COVID-19 death attributable to high temperature are estimated at the national level, and distinct spatial variability from -0.10 to 0.08 years is observed in different states. Our results provide new evidence on the relationship between high temperature and COVID-19 deaths, which might help us to understand the underlying modulation of the COVID-19 pandemic by meteorological variables and to develop epidemic policy response strategies.

Stronger susceptibilities to air pollutants of influenza A than B were identified in subtropical Shenzhen, China

Air pollution was indicated to be a key factor contributing to the aggressive spread of influenza viruses, whereas uncertainty still exists regarding to whether distinctions exist between influenza subtypes. Our study quantified the impact of five air pollutants on influenza subtype outbreaks in Shenzhen, China, a densely populated and highly urbanized megacity. Daily influenza outbreak data of laboratory-confirmed positive cases were obtained from the Shenzhen CDC, from May 1, 2013 to Dec 31, 2015. Concentrations of nitrogen dioxide (NO₂), sulfur dioxide (SO₂), particulate matters ≤2.5 μm (PM_2.5), particulate matters ≤10 μm (PM₁₀), and ozone (O₃), were retrieved from the 18 national monitoring stations. The generalized additive model (GAM) and distributed lag non-linear model (DLNM) were used to calculate the concentration-response relationships between environmental inducers and outbreak epidemics, respectively for influenza A (Flu-A) and B (Flu-B). There were 1687 positive specimens were confirmed during the study period. The cold season was restricted from Nov. 4th to Apr. 20th, covering all seasons other than the long-lasting summer. Relatively heavy fine particle matter (PM_2.5) and NO₂ pollution was observed in cold months, with mean concentrations of 46.06 μg/m³ and 40.03 μg/m³, respectively. Time-series analysis indicated that high concentrations of NO₂, PM_2.5, PM₁₀, and O₃ were associated with more influenza outbreaks at short lag periods (0-5 d). Although more Flu-B (679 cases) epidemics occurred than Flu-A (382 cases) in the cold season, Flu-A generally showed higher susceptibility to air pollutants. A 10 μg/m³ increment in concentrations of PM_2.5, PM₁₀, and O₃ at lag 04, was associated with a 2.103 (95%CI: 1.528-2.893), 1.618 (95%CI: 1.311-1.996), and 1.569 (95%CI: 1.214-2.028) of the relative risk (RR) of Flu-A, respectively. A 5 μg/m³ increase in NO₂ was associated with higher risk of Flu-A at lag 03 (RR = 1.646, 95%CI: 1.295-2.092) and of Flu-B at lag 04 (RR = 1.319, 95%CI: 1.095-1.588). Nevertheless, barely significant effect of particulate matters (PM_2.5, PM₁₀) on Flu-B and SO₂ on both subtypes was detected. Further, the effect estimates of NO₂ increased for both subtypes when coexisting with other pollutants. This study provides evidence that declining concentrations of main pollutants including NO₂, O₃, and particulate matters, could substantially decrease influenza risk in subtropical Shenzhen, especially for influenza A.

Potential effect of heat adaptation on association between number of heatstroke patients transported by ambulance and wet bulb globe temperature in Japan

This study analyzed the association between heatstroke incidence and daily maximum wet bulb globe temperature (WBGT) for all 47 prefectures in Japan by age group and severity using time-series analysis, controlling for confounders, such as seasonality and long-term trends. With the obtained association, the relative risk between the reference WBGT (defined as the value at which heatstroke starts to increase) and the daily maximum WBGT at 30 °C (RR_wbgt30) of each prefecture were calculated. For the heatstroke data, the daily number of heatstroke patients transported by ambulance at the prefecture level, provided by the Fire and Disaster Management Agency, was utilized. The analysis was conducted for age groups of 7-17 y, 18-64 y, and ≥65 y, and for severity of Deceased, Severe, Moderate (combined as DSM), and Mild. The analysis period was set from May 1 to September 30, 2015-2019. Finally, the correlation between RR_wbgt30 and the average daily maximum WBGT during the analysis period (aveWBGTms) of each prefecture was analyzed to examine the regionality of heatstroke incidence. The result showed that RR_wbgt30 is negatively correlated with aveWBGTms for the age group 18-64 y and ≥65 y (except for the age group 7-17 y) and for severity. The natural logarithm of the RR_wbgt30 of all 47 prefectures ranged from 2.0 to 8.2 for the age group 7-17 y, 1.1 to 4.0 for the age group 18-64 y, 1.8 to 6.0 for the age group ≥65 y, and 1.0 to 3.6 for DSM, and 0.9 to 4.0 for Mild. This regionality can be attributed to the effects of heat adaptation, where people in hotter regions are accustomed to implementing measures against hot environments and are more heat acclimatized than people in cooler regions.

Analysis of the heat- and cold-related cardiovascular mortality in an urban mediterranean environment through various thermal indices

During the last decades the effects of thermal stress on public health have been a great concern worldwide. Thermal stress is determined by air temperature in combination with other meteorological parameters, such as relative humidity and wind speed. The present study is focused on the Mediterranean city of Thessaloniki, Greece and it aims to explore the association between thermal stress and mortality from cardiovascular diseases, using both air temperature and other thermal indices as indicators. For that, an over-dispersed Poisson regression function was used, in combination with distributed lag non-linear models, in order to capture the delayed and nonlinear effects of temperature. Our results revealed a reverse J-shaped exposure-response curve for the total population and females and a U-shaped association for males. In all cases examined, the minimum mortality temperature was identified around the 80th percentile of each distribution. It is noteworthy that despite the fact that the highest risks of cardiovascular mortality were estimated for exposure to extreme temperatures, moderate temperatures were found to cause the highest burden of mortality. On the whole, our estimations demonstrated that the population in Thessaloniki is more susceptible to cold effects and in regard with gender, females seem to be more vulnerable to ambient thermal conditions.

The Impact of Ambient Temperature on Cardiorespiratory Mortality in Northern Greece

It is well-established that exposure to non-optimum temperatures adversely affects public health, with the negative impact varying with latitude, as well as various climatic and population characteristics. This work aims to assess the relationship between ambient temperature and mortality from cardiorespiratory diseases in Eastern Macedonia and Thrace, in Northern Greece. For this, a standard time-series over-dispersed Poisson regression was fit, along with a distributed lag nonlinear model (DLNM), using a maximum lag of 21 days, to capture the non-linear and delayed temperature-related effects. A U-shaped relationship was found between temperature and cardiorespiratory mortality for the overall population and various subgroups and the minimum mortality temperature was observed around the 65th percentile of the temperature distribution. Exposure to extremely high temperatures was found to put the highest risk of cardiorespiratory mortality in all cases, except for females which were found to be more sensitive to extreme cold. It is remarkable that the highest burden of temperature-related mortality was attributed to moderate temperatures and primarily to moderate cold. The elderly were found to be particularly susceptible to both cold and hot thermal stress. These results provide new evidence on the health response of the population to low and high temperatures and could be useful to local authorities and policy-makers for developing interventions and prevention strategies for reducing the adverse impact of ambient temperature.

NICU-based stress response and preterm infant neurobehavior: exploring the critical windows for exposure

BACKGROUND

Exposure to maternal stress in utero negatively impacts cognitive and behavioral outcomes of children born at term. The neonatal intensive care unit (NICU) can be stressful for preterm infants during a developmental period corresponding to the third trimester of gestation. It is unknown whether stress in the NICU contributes to adverse neurodevelopment among NICU graduates. The aim was to examine the association between salivary cortisol and early neurodevelopment in preterm infants.

METHODS

We examined the association between cortisol levels during the NICU hospitalization and subsequent performance on the NICU Network Neurobehavioral Scales (NNNS), estimating time-specific associations and considering sex differences.

RESULTS

Eight hundred and forty salivary cortisol levels were measured from 139 infants. Average cortisol levels were inversely associated with NNNS Regulation scores for both male and female infants (β = -0.19; 95% CI: -0.44, -0.02). Critical developmental windows based on postmenstrual age were identified, with cortisol measured <30 weeks PMA positively associated with Habituation and Lethargy scores (β = 0.63-1.04). Critical developmental windows based on chronological age were identified, with cortisol measured in the first week of life inversely associated with Attention score (β = -1.01 for females; -0.93 for males).

CONCLUSIONS

Stress in the NICU at specific developmental time points may impact early preterm infant neurodevelopment.

IMPACT

Stress in the neonatal intensive care unit can impact the neurodevelopmental trajectory of premature infants. The impact of stress is different at different points in development. The impact of stress is sexually dimorphic.

Climate Change and Health: Consequences of High Temperatures among Vulnerable Groups in Finland

In this article, we examine the effects of high temperatures on hospital visits and mortality in Finland. This provides new information of the topic in a context of predominantly cool temperatures. Unique, individual-level data are used to examine the relationship at the municipality-month level over a span of 20 years. Linear regression methods alongside high-dimensional fixed effects are used to minimize confounding variation. Analysis is conducted with special emphasis on the elderly population, as well as on specific elderly risk groups identified in previous literature. We show that for an additional day per month above 25°C, monthly all-cause mortality increases by 1.5 percent (95% CI: 0.4%-2.6%) and acute hospital visits increase by 1.1 percent (95% CI: 0.7%-1.6%). We also find some evidence that these effects are elevated in selected population subgroups, the low-income elderly, and people with dementia. Hospital visits also increase among younger age groups, illustrating the importance of using multiple health indicators. Such detailed evidence is important for identifying vulnerable groups as extreme heat waves are expected to become more frequent and intense in northern countries.

Association between ambient temperature and cardiovascular disease hospitalisations among farmers in suburban northwest China

Cardiovascular disease (CVD) has become a severe public health and social issue in China. However, in northwest China, evidence on the association between ambient temperature and CVD hospitalisations in suburban farmers is somewhat limited. We collected CVD hospitalisations and meteorological data (2012-2015) in Zhangye suburbs and assessed the temperature-related risk and burden of admission by fitting a distributed lag nonlinear model to probe the relationship between ambient temperature and CVD hospitalisations among farmers in suburban northwest China. The results show that 23,921 cases of CVD admissions were recorded from 2012 to 2015. There was a "U-shaped" association between temperature and hospitalisations. Compared with the minimum admissions temperature (MAT) at 15.3 °C, the cumulative relative risk (RR) over lag 0-21 days was 1.369 (95% CI 0.980-1.911) for extreme cold temperature (1st percentile, -15 °C), 1.353 (95% CI 1.063-1.720) for moderate cold (5th percentile, -11 °C), 1.415 (95% CI 1.117-1.792) for extreme heat (99th percentile, 26 °C), and 1.241 (95% CI 1.053-1.464) for moderate heat (95th percentile, 24 °C). Female farmers were more susceptible to low and high temperatures than male farmers. Farmers aged ≥ 65 years old were more sensitive to low temperatures, while farmers aged < 65 years old were more sensitive to high temperatures. A total of 13.4% (3,208 cases) of the hospitalisation burden for CVD were attributed to temperature exposure, with the moderate range of temperatures accounting for the most significant proportion (12.2%). Ambient temperature, primarily moderate temperatures, might be an essential factor for cardiovascular-related hospitalisations among farmers in suburban northwest China.

The impact of enterovirus A71 vaccination program on hand, foot, and mouth disease in Guangdong, China:a longitudinal surveillance study

Enterovirus A71 (EV71) vaccination program was introduced in 2016 in China. Based on a longitudinal surveillance dataset from 2012 to 2019 in Guangdong, China, we estimated the impact of the EV71 vaccination program on hand, foot, and mouth disease (HFMD) incidence, by using a counterfactual prediction made from synthetic control approach integrated with a Bayesian time-series model. We observed a relative reduction of 41.4% for EV71-associated HFMD cases during the post-vaccination period of 2017-2019, corresponding to 26,226 cases averted. The reduction of EV71-associated HFMD cases raised with the elevation of EV71 vaccine coverage by year. We found an indirect effect for the children aged 6-14 years who were less likely to be vaccinated. Whereas, the EV71 vaccine may not protect against non-EV71-associated HFMD. This study provides a template for ongoing public health surveillance of EV71 vaccine effectiveness with a counterfactual study design. Our results show strong evidence of the EV71 vaccination program working on reducing EV71-associated HFMD in real-world settings. The finding will benefit policy-making of EV71 vaccination and the prevention of HFMD.

A time series analysis of ambient air pollution and low birth weight in Xuzhou, China

The objective of this study was to examine the relationship between weekly specific maternal air pollution exposures and low birth weight. We fitted a distributed lag nonlinear model (DLNM) to analyze the nonlinear exposure-response association and delayed effects of air pollutants on the risk for low birth weight. The model assumed that all live births have 40 gestational weeks.The 1st week lag was the 40th gestational week, and 40th lag week was the 1st gestational week.The study included 71,809 live births (from July 1, 2016, to June 30, 2019), of which 2,391 (3.33%) exhibited low birth weight. The results demonstrated that exposure of pregnant women to PM10 at lag 22-30 weeks was significantly associated with low birth weight risk, with the greatest impact at the lag 30 week. Exposure to SO2 at lag 29-37 weeks was significantly associated with low birth weight risk. The sensitive exposure window for NO2 began at lag 25-37 weeks of pregnancy. The lag 6-10 weeks constituted the susceptible exposure window for O3. Therefore we concluded that maternal exposures to PM10, SO2, NO2, and O3 were associated with increased risk for low birth weight.

Low ambient temperature as a novel risk factor of oral diseases: A time-series study

BACKGROUND

The seasonal variation of oral diseases has been observed in life, but the influences of oral diseases associated with non-optimal ambient temperature were unknown.

OBJECTIVE

To examine whether non-optimum ambient temperature is associated with increased risks of oral diseases.

METHODS

We conducted a time series study based on outpatient data from the Shanghai Health Information Center, containing all public hospitals in Shanghai from 2016 to 2019. Generalized additive models with distributed lagged nonlinear models were applied to fit the data.

RESULT

A total of 3,882,636 outpatient cases of oral diseases were collected. Low temperature (<7 °C) posed increased risks for oral diseases. Daily temperature above 7 °C had no effect on oral diseases. The excess risks were present on the lag 1 day and lasted till lag 7 day. Relative to referent temperatures, the cumulative risks of total oral diseases, pulpitis, periodontitis, gum pain, stomatitis, and glossitis at extreme low temperature (-3 °C, 1st percentile) over lag 0-7 day were 1.92 (95% confidence interval, CI: 1.40, 2.63), 2.40 (95% CI: 1.78, 3.25), 1.62 (95% CI: 1.15, 2.29), 1.75 (95% CI: 1.08, 2.83), 1.81 (95% CI: 1.30, 2.53), and 2.22 (95% CI: 1.23, 3.99). These associations were larger in patients who were above age 60.

CONCLUSION

This study provided novel epidemiological evidence that low ambient temperature may increase the risks of oral diseases. The temperature thresholds for eight oral diseases range from 3 to 7 °C. The excess risks could last for 7 days and were larger in older patients.

Association between satellite-detected tropospheric nitrogen dioxide and acute respiratory infections in children under age five in Senegal: spatio-temporal analysis

BACKGROUND

There is growing evidence to suggest that exposure to a high concentration of nitrogen dioxide (NO₂) can lead to a higher incidence of Acute Respiratory Infections (ARIs) in children; however, such an association remains understudied in Sub-Saharan Africa due to the limited availability of exposure data. This study explored this association by using the satellite-detected tropospheric NO₂ concentrations measured by Sentinel-5 Precursor and ARI symptoms in children under age five collected in the Demographic and Health Survey (DHS) in Senegal.

METHODS

We matched the daily tropospheric NO₂ exposure with the individual ARI symptoms according to the DHS survey clusters spatially and temporally and conducted a logistic regression analysis to estimate the association of exposure to NO₂ with ARI symptoms in two preceding weeks.

RESULTS

We observed a positive association between exposure to continuous levels of NO₂ and ARI symptoms after adjusting for confounders (OR 1.27 per 10 mol/m², 95% CI: 1.06 - 1.52). When the association was further examined by quartile exposure categories, the 4th quartile category was positively associated with symptoms of ARI after adjusting for confounders (OR 1.71, 95% CI: 1.08-2.69). This suggests that exposure to certain high levels of NO₂ is associated with the increased risk of children having symptoms of ARI in Senegal.

CONCLUSIONS

This study highlights the need for increased research on the effects of ambient NO₂ exposure in Africa as well as the need for more robust, ground-based air monitoring in the region. For a country like Senegal, where more than 90% of the population lives in areas that do not meet the national air quality standards, it is urgently required to implement air pollution prevention efforts to protect children from the health hazards of air pollution.

Effect of Cold Spells and Their Different Definitions on Mortality in Shenzhen, China

A high premium has been put on researching the effects of cold spells because of their adverse influence on people's daily lives and health. The study aimed to find the most appropriate definition of the cold spell in Shenzhen and quantify the impact of cold spells on mortality. Based on the daily mortality data in Shenzhen from 2013 to 2017 and the meteorological and pollutant data from the same period, we quantified the effect of cold spells using eight different definitions in the framework of a distributed lag non-linear model with a quasi-Poisson distribution. In Shenzhen, low temperatures increase the risk of death more significantly than high temperatures (using the optimal temperature as the cut-off value). Comparing the quasi-Akaike information criterion value, attribution fraction (b-AF), and attribution number (b-AN) for all causes of deaths and non-accidental deaths, the optimal definition of the cold spell was defined as the threshold was 3rd percentile of the daily average temperature and duration for 3 or more consecutive days (all causes: b-AF = 2.31% [1.01–3.50%], b-AN = 650; non-accidental: b-AF = 1.92% [0.57–3.17%], b-AN = 471). For cardiovascular deaths, the best definition was the temperature threshold as the 3rd percentile of the daily average temperature with a duration of 4 consecutive days (cardiovascular: b-AF = 1.37% [0.05–2.51%], b-AN = 142). Based on the best definition in the model, mortality risk increased in cold spells, with a statistically significant lag effect occurring as early as the 4th day and the effect of a single day lasting for 6 days. The maximum cumulative effect occurred on the 14th day (all-cause: RR = 1.54 [95% CI, 1.20–1.98]; non-accidental: RR = 1.43 [95% CI, 1.11–1.84]; cardiovascular: RR = 1.58 [95% CI, 1.00–2.48]). The elderly and females were more susceptible to cold spells. Cold spells and their definitions were associated with an increased risk of death. The findings of this research provide information for establishing an early warning system, developing preventive measures, and protecting susceptible populations.

A data quality assessment of the first four years of malaria reporting in the Senegal DHIS2, 2014-2017

Background

As the global burden of malaria decreases, routine health information systems (RHIS) have become invaluable for monitoring progress towards elimination. The District Health Information System, version 2 (DHIS2) has been widely adopted across countries and is expected to increase the quality of reporting of RHIS. In this study, we evaluated the quality of reporting of key indicators of childhood malaria from January 2014 through December 2017, the first 4 years of DHIS2 implementation in Senegal.

Methods

Monthly data on the number of confirmed and suspected malaria cases as well as tests done were extracted from the Senegal DHIS2. Reporting completeness was measured as the number of monthly reports received divided by the expected number of reports in a given year. Completeness of indicator data was measured as the percentage of non-missing indicator values. We used a quasi-Poisson model with natural cubic spline terms of month of reporting to impute values missing at the facility level. We used the imputed values to take into account the percentage of malaria cases that were missed due to lack of reporting. Consistency was measured as the absence of moderate and extreme outliers, internal consistency between related indicators, and consistency of indicators over time.

Results

In contrast to public facilities of which 92.7% reported data in the DHIS2 system during the study period, only 15.3% of the private facilities used the reporting system. At the national level, completeness of facility reporting increased from 84.5% in 2014 to 97.5% in 2017. The percentage of expected malaria cases reported increased from 76.5% in 2014 to 94.7% in 2017. Over the study period, the percentage of malaria cases reported across all districts was on average 7.5% higher (P < 0.01) during the rainy season relative to the dry season. Reporting completeness rates were lower among hospitals compared to health centers and health posts. The incidence of moderate and extreme outlier values was 5.2 and 2.3%, respectively. The number of confirmed malaria cases increased by 15% whereas the numbers of suspected cases and tests conducted more than doubled from 2014 to 2017 likely due to a policy shift towards universal testing of pediatric febrile cases.

Conclusions

The quality of reporting for malaria indicators in the Senegal DHIS2 has improved over time and the data are suitable for use to monitor progress in malaria programs, with an understanding of their limitations. Senegalese health authorities should maintain the focus on broader adoption of DHIS2 reporting by private facilities, the sustainability of district-level data quality reviews, facility-level supervision and feedback mechanisms at all levels of the health system.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12913-021-07364-6.

Influenza A and B outbreaks differed in their associations with climate conditions in Shenzhen, China

Under the variant climate conditions in the transitional regions between tropics and subtropics, the impacts of climate factors on influenza subtypes have rarely been evaluated. With the available influenza A (Flu-A) and influenza B (Flu-B) outbreak data in Shenzhen, China, which is an excellent example of a transitional marine climate, the associations of multiple climate variables with these outbreaks were explored in this study. Daily laboratory-confirmed influenza virus and climate data were collected from 2009 to 2015. Potential impacts of daily mean/maximum/minimum temperatures (T/T_max/T_min), relative humidity (RH), wind velocity (V), and diurnal temperature range (DTR) were analyzed using the distributed lag nonlinear model (DLNM) and generalized additive model (GAM). Under its local climate partitions, Flu-A mainly prevailed in summer months (May to June), and a second peak appeared in early winter (December to January). Flu-B outbreaks usually occurred in transitional seasons, especially in autumn. Although low temperature caused an instant increase in both Flu-A and Flu-B risks, its effect could persist for up to 10 days for Flu-B and peak at 17 C (relative risk (RR) = 14.16, 95% CI: 7.46-26.88). For both subtypes, moderate-high temperature (28 C) had a significant but delayed effect on influenza, especially for Flu-A (RR = 26.20, 95% CI: 13.22-51.20). The Flu-A virus was sensitive to RH higher than 76%, while higher Flu-B risks were observed at both low (< 65%) and high (> 83%) humidity. Flu-A was active for a short term after exposure to large DTR (e.g., DTR = 10 C, RR = 12.45, 95% CI: 6.50-23.87), whereas Flu-B mainly circulated under stable temperatures. Although the overall wind speed in Shenzhen was low, moderate wind (2-3 m/s) was found to favor the outbreaks of both subtypes. This study revealed the thresholds of various climatic variables promoting influenza outbreaks, as well as the distinctions between the flu subtypes. These data can be helpful in predicting seasonal influenza outbreaks and minimizing the impacts, based on integrated forecast systems coupled with short-term climate models.

Associations between air pollution exposure and birth defects: a time series analysis

Air pollution is a serious environmental problem in China. Birth defects are particularly vulnerable to outdoor air pollution. Our study was to evaluate the association between short-term exposure to air pollutants and the risk of birth defects. Daily data including the air pollutants, meteorological characteristics, and birth records were obtained in Hefei, China, during January 2013 to December 2016. The findings showed that PM_2.5, PM₁₀, SO₂, NO₂, and O₃ exposures were positively correlated with the risk of birth defects. Maternal exposure to PM_2.5 and SO₂ during the 4th to 13th gestational weeks was observed to have a significant association with the risk of birth defects, with the maximum effect in the 7th or 8th week for PM_2.5 and the maximum effect in the 7th week for SO₂. The positively significant exposure windows were the 4th to 14th weeks for PM₁₀, the 4th to 12th weeks for NO₂, and the 26th to 35th weeks for O₃, respectively. The strongest associations were observed in the 8th week for PM₁₀, the 7th week for NO₂, and in the 31st or 32nd week for O_3. The findings of this study demonstrate that air pollutants increase the risk of birth defects among women during pregnancy in Hefei, China, which provide evidence for improving the health of pregnant women and neonates in developing countries, and uncovered potential opportunities to reduce or prevent birth defects by proactive measures during pregnancy.

Effect of diurnal temperature change on cardiovascular risks differed under opposite temperature trends

Temperature change between neighboring days (TCN) is an important trigger for cardiovascular diseases, but the modulated effects by seasonal temperature trends have been barely taken into account. A quantified comparison between impacts of positive TCNs (temperature rise) and negative situations (temperature drop) is also needed. We evaluated the associations of TCNs with emergency room (ER) visits for coronary heart disease (CHD) and cerebral infarction (CI) in Beijing, China, from 2008 to 2012. A year was divided into two segments dominated by opposite temperature trends, quasi-Poisson regression with distributed lag nonlinear models estimating TCN-morbidity relations were employed, separately for each period. High morbidities of CHD and CI both occurred in transitional seasons accompanied by large TCNs. Under warming backgrounds, positive TCNs increased CHD risk in patients younger than 65 years, and old people showed limited sensitivity. In the cooling periods, negative TCNs induced CHD risk in females and the elderly; the highest RR showed on lag 6 d. In particular, a same diurnal temperature decrease (e.g., - 2°C) induced greater RR (RR = 1.113, 95% CIs: 1.033-1.198) on old people during warming periods than cooling counterparts (RR = 1.055, 95% CIs: 1.011-1.100). Moreover, positive TCNs elevated CI risk regardless of background temperatures, and males were particularly vulnerable. Seasonal temperature trends modify TCN-cardiovascular morbidity associations significantly, which may provide new insights into the health impact of unstable weathers.

Impact of ambient temperature on cardiovascular disease hospital admissions in farmers in China's Western suburbs

Cardiovascular disease (CVD) has been a major threat to global public health. The association between temperature and CVD has been widely studied and reported in cities in developed countries. However, information from developing countries, especially from suburbs and countryside, is quite limited. In this study, the daily time series data on CVD hospital admissions in farmers in the suburbs of Tianshui, China, and the meteorological data from 2012 to 2015, were collected; besides, a quasi-Poisson regression with a distributed-lag non-linear model (DLNM) was used to explore the impact of local daily mean temperature on CVD hospital admissions in suburban farmers. This study found that, first, from 2011 to 2015, a total of 30,611 person-times of CVD hospital admissions in farmers were recorded; second, there was a "J-shaped" relation between temperature and CVD hospital admissions, and both low and high temperature increased the risk of hospital admission, but the impact of high temperature was greater; third, compared with the minimum hospitalization temperature (MHT) at 0.3 °C, during 0 to 21 lag days, the cumulative relative risk (RR) for extreme cold and heat (1st and 99th percentile of temperature, respectively) was 1.117 (95% CI 0.941-1.325) and 1.740 (95% CI 1.302-2.327), respectively, and that of moderate cold and heat (5st and 95th percentile of temperature, respectively) was 1.029 (95% CI 0.958-1.106) and 1.572 (95% CI 1.210-2.042), respectively; fourth, compared with male and ≥ 65 years groups, the risk for low temperature was greater for female and < 65 years groups, the risk for high temperature was just the opposite; last, about 21.04% of CVD hospital admissions burden were attributed to the ambient temperature, and most of (about 19.26%) were caused by moderate heat. In Tianshui, alongside with extreme temperature, the moderate temperature might be an important risk factor for CVD hospital admissions in suburban farmers.

Preconception ambient temperature and preterm birth: a time-series study in rural Henan, China

Changes in the preconception ambient temperature (PAT) can affect the gametogenesis, disturbing the development of the embryo, but the health risks of PAT on the developing fetus are still unclear. Here, based on the National Free Preconception Health Examination Project in the rural areas of Henan Province, we evaluate the effects of PAT on preterm birth (PTB). Data of 1,231,715 records from self-reported interviews, preconception physical examination, early gestation follow-up, and postpartum follow-up were collected from 1 January 2013 to 31 December 2016. Generalized additive models were used to assess the cumulative and lag effects of PAT upon PTB. The significant cumulative effects of mean temperature within 2 weeks and 3 weeks on the risk of PTB, especially upon late PTB (34-36 weeks) (P < 0.05), were observed. Exposure to extreme heat (> 90th percentile) within 2 weeks (RR = 1.470) and 3 weeks (RR = 1.375) before conception could increase the risk of PTB. After stratifying PTB, exposure to extreme heat within 2 weeks before conception can increase the risks of early (< 34 weeks) and late PTB (P < 0.05). Besides, exposure to extreme cold (< 10th percentile) within 3 weeks or longer before conception can elevate the risk of PTB, especially late PTB. The significant lag effects of temperature changes on the risk of early PTB (lag-8 days or earlier) were observed. In conclusion, the risk of PTB was susceptible to PAT changes within 2 weeks or longer before conception. Our findings provide (i) guidance for rural couples to make pregnancy plans and (ii) scientific evidence for the government to formulate policies to prevent PTB.

A novel cross-validation strategy for artificial neural networks using distributed-lag environmental factors

In recent years, machine learning methods have been applied to various prediction scenarios in time-series data. However, some processing procedures such as cross-validation (CV) that rearrange the order of the longitudinal data might ruin the seriality and lead to a potentially biased outcome. Regarding this issue, a recent study investigated how different types of CV methods influence the predictive errors in conventional time-series data. Here, we examine a more complex distributed lag nonlinear model (DLNM), which has been widely used to assess the cumulative impacts of past exposures on the current health outcome. This research extends the DLNM into an artificial neural network (ANN) and investigates how the ANN model reacts to various CV schemes that result in different predictive biases. We also propose a newly designed permutation ratio to evaluate the performance of the CV in the ANN. This ratio mimics the concept of the R-square in conventional statistical regression models. The results show that as the complexity of the ANN increases, the predicted outcome becomes more stable, and the bias shows a decreasing trend. Among the different settings of hyperparameters, the novel strategy, Leave One Block Out Cross-Validation (LOBO-CV), demonstrated much better results, and the lowest mean square error was observed. The hyperparameters of the ANN trained by the LOBO-CV yielded the minimum number of prediction errors. The newly proposed permutation ratio indicates that LOBO-CV can contribute up to 34% of the prediction accuracy.

Meta-analysis of generalized additive models in neuroimaging studies

Analyzing data from multiple neuroimaging studies has great potential in terms of increasing statistical power, enabling detection of effects of smaller magnitude than would be possible when analyzing each study separately and also allowing to systematically investigate between-study differences. Restrictions due to privacy or proprietary data as well as more practical concerns can make it hard to share neuroimaging datasets, such that analyzing all data in a common location might be impractical or impossible. Meta-analytic methods provide a way to overcome this issue, by combining aggregated quantities like model parameters or risk ratios. Most meta-analytic tools focus on parametric statistical models, and methods for meta-analyzing semi-parametric models like generalized additive models have not been well developed. Parametric models are often not appropriate in neuroimaging, where for instance age-brain relationships may take forms that are difficult to accurately describe using such models. In this paper we introduce meta-GAM, a method for meta-analysis of generalized additive models which does not require individual participant data, and hence is suitable for increasing statistical power while upholding privacy and other regulatory concerns. We extend previous works by enabling the analysis of multiple model terms as well as multivariate smooth functions. In addition, we show how meta-analytic p-values can be computed for smooth terms. The proposed methods are shown to perform well in simulation experiments, and are demonstrated in a real data analysis on hippocampal volume and self-reported sleep quality data from the Lifebrain consortium. We argue that application of meta-GAM is especially beneficial in lifespan neuroscience and imaging genetics. The methods are implemented in an accompanying R package metagam, which is also demonstrated.

The Effects of Temperature on Accident and Emergency Department Attendances in London: A Time-Series Regression Analysis

The epidemiological research relating mortality and hospital admissions to ambient temperature is well established. However, less is known about the effect temperature has on Accident and Emergency (A&E) department attendances. Time-series regression analyses were conducted to investigate the effect of temperature for a range of cause- and age-specific attendances in Greater London (LD) between 2007 to 2012. A seasonally adjusted Poisson regression model was used to estimate the percent change in daily attendances per 1 °C increase in temperature. The risk of overall attendance increased by 1.0% (95% CI 0.8, 1.4) for all ages and 1.4% (1.2, 1.5) among 0- to 15-year-olds. A smaller but significant increase in risk was found for cardiac, respiratory, cerebrovascular and psychiatric presentations. Importantly, for fracture-related attendances, the risk rose by 1.1% (0.7, 1.5) per 1 °C increase in temperature above the identified temperature threshold of 16 °C, with the highest increase of 2.1% (1.5, 3.0) seen among 0- to 15-year-olds. There is a positive association between increasing temperatures and A&E department attendance, with the risk appearing highest in children and the most deprived areas. A&E departments are vulnerable to increased demand during hot weather and therefore need to be adequately prepared to address associated health risks posed by climate change.

Seasonal and temperature effect on serum lithium concentrations

BACKGROUND

Lithium remains the gold standard treatment for bipolar disorder. However, it has a very narrow therapeutic index (0.6-0.8 mmol/L). It has been suggested that high environmental temperature can lead to dehydration, elevated plasma lithium concentration and then lithium toxicity.

OBJECTIVES

We aimed to investigate the effect of seasonal and short-term changes in temperature on serum lithium concentrations in Sydney, Australia.

METHODS

We retrospectively analysed data from all patients who had serum lithium concentrations taken from the Prince of Wales and Sutherland Hospitals between 2008 and 2018. Temperature data came from the Bureau of Meteorology. We examined correlations between lithium concentrations and the preceding 5 days maximum temperatures, month and season. We also performed a longitudinal analysis of the effect of temperature and seasons within selected patients who had repeated levels.

RESULTS

A total of 11,912 serum lithium concentrations from 2493 patients were analysed. There was no significant association between higher lithium concentration and preceding higher temperatures (r = -0.008, p = 0.399). There was also no important seasonal or monthly variation, across all patients or in the smaller cohort with longitudinal data (n = 123, r = 0.008, 95% confidence interval: [-0.04, 0.06]).

CONCLUSION

There were no clinically important differences in serum lithium concentration related to seasons, months or temperatures, which suggests that patients on lithium are able to adequately maintain hydration during hot weather in Sydney.

Short-term effects of ambient temperature on non-external and cardiovascular mortality among older adults of metropolitan areas of Mexico

Multi-city studies assessing the association between acute exposure to temperature and mortality in Latin American are limited. To analyze the short-term effect of changes in temperature (increase and decrease) on daily non-external and cardiovascular mortality from 1998 to 2014, in people 65 years old and over living in 10 metropolitan areas of Mexico. Analyses were performed through Poisson regression models with distributed lag non-linear models. Statistical comparison of minimum mortality temperature (MMT) and city-specific cutoffs of 24-h temperature mean values (5th/95th and 1st/99th percentiles) were used to obtain the mortality relative Risk (RR) for cold/hot and extreme cold/extreme hot, respectively, for the same day and lags of 0-3, 0-7, and 0-21 days. A meta-analysis was conducted to synthesize the estimates (RR_pooled). Significant non-linear associations of temperature-mortality relation were found in U or inverted J shape. The best predictors of mortality associations with cold and heat were daily temperatures at lag 0-7 and lag 0-3, respectively. RR_pooled of non-external causes was 6.3% (95%CI 2.7, 10.0) for cold and 10.2% (95%CI 4.4, 16.2) for hot temperatures. The RR_pooled for cardiovascular mortality was 7.1% (95%CI 0.01, 14.7) for cold and 7.1% (95%CI 0.6, 14.0) for hot temperatures. Results suggest that, starting from the MMT, the changes in temperature are associated with an increased risk of non-external and specific causes of mortality in elderly people. Generally, heat effects on non-external and specific causes of mortality occur immediately, while cold effects occur within a few days and last longer.

The effect of ambient temperature on cardiovascular mortality in 27 Brazilian cities

BACKGROUND

There is limited evidence on the relationship between temperature and cardiovascular mortality in middle and low-income countries, particularly in Latin America. In this study, we investigated the total effect of temperature on cardiovascular mortality in 27 Brazilian cities, and the effect modification by geographic, socioeconomic, demographic and infrastructure characteristics within cities.

METHODS

In the city-specific analysis, we used time-series analyses to estimate the relationship between mean temperature and daily cardiovascular mortality using quasi-Poisson generalized linear models combined with distributed lag non-linear models. In the second stage, a meta-analysis was used to pool the effects of temperature on cardiovascular mortality for Brazil and its five regions (Central-West, North, Northeast, South, and Southeast). We used a meta-regression to examine the effect modification of city-specific geographic, socioeconomic, demographic and infrastructure-related variables.

RESULTS

The risks associated with temperature varied across the locations. Higher cardiovascular mortality was associated with low and high temperatures in most of the cities, Brazil and the Central-West, North, South, and Southeast regions. The overall relative risk (RR) for Brazil was 1.26 (95% confidence interval [CI]: 1.17-1.35) for the 1st percentile of temperature and 1.07 (95% CI: 1.01-1.13) for the 99th percentile of temperature versus the 79th percentile (27.7 °C), where RR was lowest. The temperature range was the variable that best explained the variation in effects among the cities, with greater effects in locations having a broader range.

CONCLUSIONS

The results indicate effects of low and high temperatures on the risk of cardiovascular mortality in most of Brazil's capital cities, besides a pooled effect for Brazil and the Central-West, North, South, and Southeast regions. These findings can help inform public policies addressing the health impact of temperature extremes, especially in the context of climate change.

Sex differences in the temperature dependence of kidney stone presentations: a population-based aggregated case-crossover study

Previous studies assumed a uniform relationship between heat and kidney stone presentations. Determining whether sex and other characteristics modify the temperature dependence of kidney stone presentations has implications for explaining differences in nephrolithiasis prevalence and improving projections of the effect of climate change on nephrolithiasis. We performed an aggregated case-crossover study among 132,597 children and adults who presented with nephrolithiasis to 68 emergency departments throughout South Carolina from 1997 to 2015. We used quasi-Poisson regression with distributed lag non-linear models to estimate sex differences in the cumulative exposure and lagged response between maximum daily wet-bulb temperatures and emergent kidney stone presentations, aggregated at the ZIP-code level. We also explored interactions by age, race, payer, and climate. Compared to 10 °C, daily wet-bulb temperatures at the 99th percentile were associated with a greater increased relative risk (RR) of kidney stone presentations over 10 days for males (RR 1.73; 95% CI 1.56, 1.91) than for females (RR 1.15; 95% CI 1.01, 1.32; interaction P < 0.001). The shape of the lagged response was similar for males and females, with the greatest risk estimated for the 2 days following high temperatures. There were weak differences by age, race, and climatic zone, and no differences by payer status. The estimated risk of presenting emergently with kidney stones within 10 days of high daily wet-bulb temperatures was substantially greater among men than women, and similar between patients with public and private insurance. These findings suggest that the higher risk among males may be due to sexually dimorphic physiologic responses rather than greater exposure to ambient temperatures.

Short-term impacts of air temperature on hospitalizations for mental disorders in Lisbon

BACKGROUND

Individuals with mental disorders are often susceptible to the effects of extreme ambient temperatures. The aim of this study is to assess the short-term impacts of daily mean temperature on hospitalizations for mental disorders in the Lisbon Metropolitan Area, Portugal.

METHODS

To assess the short-term impacts of daily mean temperature on hospitalizations for mental disorders (2008-2014), a quasi-Poisson generalized additive model combined with a distributed lag non-linear model was applied. The model was adjusted for day of the week, air pollution, relative humidity, time and seasonality.

RESULTS

The number of hospital admissions for mental disorder during the study period was 30,139. Hospital admissions increase significantly with high temperatures on day of exposure, at lag 0-1 and at lag 0-2. Women are more vulnerable than men, and there was no difference between the age groups studied.

CONCLUSIONS

The exposure to high temperatures should be considered a significant risk factor for mental disorders; therefore, patient management services may need to be strengthened when extreme high temperature alerts are given.

Socio-geographic disparity in cardiorespiratory mortality burden attributable to ambient temperature in the United States

Compared with relative risk, attributable fraction (AF) is more informative when assessing the mortality burden due to some environmental exposures (e.g., ambient temperature). Up to date, however, available AF-based evidence linking temperature with mortality has been very sparse regionally and nationally, even for the leading mortality types such as cardiorespiratory deaths. This study aimed to quantify national and regional burden of cardiorespiratory mortality (CRM) attributable to ambient temperature in the USA, and to explore potential socioeconomic and demographic sources of spatial heterogeneity between communities. Daily CRM and weather data during 1987-2000 for 106 urban communities across the mainland of USA were acquired from the publicly available National Morbidity, Mortality and Air Pollution Study (NMMAPS). We did the data analysis using a three-stage analytic approach. We first applied quasi-Poisson regression incorporated with distributed lag nonlinear model to estimate community-specific temperature-CRM associations, then pooled these associations at the regional and national level through a multivariate meta-analysis, and finally estimated the temperature-AF of CRM and performed subgroup analyses stratified by community-level characteristics. Both low and high temperatures increased short-term CRM risk, while temperature-CRM associations varied by regions. Nationally, the fraction of cardiorespiratory deaths caused by the total non-optimum, low, and high temperatures was 7.58% (95% empirical confidence interval, 6.68-8.31%), 7.15% (6.31-7.85%), and 0.43% (0.37-0.46%), respectively. Greater temperature-AF was identified in two northern regions (i.e., Industrial Midwest and North East) and communities with lower temperature and longitude, higher latitude, and moderate humidity. Additionally, higher vulnerability appeared in locations with higher urbanization level, more aging population, less White race, and lower socioeconomic status. Ambient temperature may be responsible for a large fraction of cardiorespiratory deaths. Also, temperature-AF of CRM varied considerably by geographical and climatological factors, as well as community-level disparity in socioeconomic status.

The effect of ambient temperature on the activity of influenza and influenza like illness in Jiangsu Province, China

OBJECTIVE

We aimed to evaluate and quantify the association between ambient temperature and activity of influenza like illness (ILI) and influenza in Jiangsu Province, China.

METHOD

Daily data of meteorology, influenza-like illness and detected influenza virus from 1 April 2013 to 27 March 2016 were collected. Distributed lag non-linear model (DLNM) was used to quantify the exposure-lag-response of ILI and influenza activity to daily average temperature.

RESULT

Influenza A virus (Flu-A) circulated throughout the year with two peaks at -4 °C and 28 °C respectively, while influenza B (Flu-B) viruses were usually tested positive in winter or early spring and peaked at 5 °C. The lag-response curves revealed that the RR of ILI increased with time and peaked 1 day later at low temperature (3 °C), however, the maximum RR of ILI caused by high temperature (26 °C) appeared immediately on day 0, the similar phenomena of immediate effect to ILI at high temperature were also observed in the lag-response curve for Flu-A or Flu-B.

CONCLUSION

ILI and Flu-A experienced two peaks of circulates at both low and high temperature in Jiangsu. The influenza viruses activity did drive up the rising of ILI%, particularly the activity of Flu-A which circulated throughout the year played a crucial role. Regional homogeneity was the relatively mainstream in aspects of cumulative association between influenza activity and temperature in Jiangsu Province.

Differences of hemorrhagic and ischemic strokes in age spectra and responses to climatic thermal conditions

The risks of emergency room (ER) visits for cerebral infarction (CI) and intracerebral hemorrhage (ICH) is found to differ in different age groups under different climatic thermal environments. Based on CI and ICH related ER-visit records from three major hospitals in Beijing, China, from 2008 to 2012, the advanced Universal Thermal Climate Index (UTCI), was adopted in this study to assess the climatic thermal environment. Particularly, daily mean UTCI was used as a predictor for the risk of ER visits for CI and ICH. A generalized quasi-Poisson additive model combined with a distributed lag non-linear model was performed to quantify their association. The results indicated that (i) the highest growth rate of ER visits for ICH occurred in age 38 to 48, whereas an increasing ER admissions for CI maintained at age 38 to 78. (ii) The frequency distribution of UTCI in Beijing peaked at -8 and 30 °C, corresponding to moderate cold stress and moderate heat stress, respectively. (iii) Correlation analysis indicated that ICH morbidity was negatively correlated with UTCI, whereas occurrence of CI showed no significant association with UTCI. (iv) The estimated relative risk of ER visits corresponding to 1 °C change in UTCI, which was then stratified by age and gender, indicated that all sub-groups of ICH patients responded similarly to thermal stress. Namely, there is an immediate ICH risk (UTCI = -13 °C, RR = 1.35, 95% CIs: 1.11-1.63) from cold stress on the onset day, but non-significant impact from heat stress. As for CI occurrences, no effect from cold stress was identified, except for only those aged 45 to 65 were threatened by heat stress (UTCI = 38 °C, RR = 1.64, 95% CIs: 1.10-2.44) on lag 0-2 d.

Dentine biomarkers of prenatal and early childhood exposure to manganese, zinc and lead and childhood behavior

BACKGROUND

Metal exposure alters neurodevelopmental outcomes; little is known about critical windows of susceptibility when exposure exerts the strongest effect.

OBJECTIVE

To examine associations between dentine biomarkers of manganese (Mn), zinc (Zn) and lead (Pb) and later childhood behaviors.

METHODS

Subjects enrolled in a longitudinal birth cohort study in Mexico City provided naturally shed deciduous teeth. We estimated weekly prenatal and postnatal dentine Mn, Zn and Pb concentrations in teeth using laser ablation-inductively coupled plasma-mass spectrometry (LA-ICP-MS) and measured behavior at ages 8-11 years of age using the Behavior Assessment System for Children, 2nd edition (BASC-2). We used distributed lag models and lagged weighted quantile sum regression to identify the role of individual and combined dentine biomarkers of Mn, Zn and Pb on behavioral outcomes controlling for maternal education and gestational age.

RESULTS

Among the 133 subjects included in this study, prenatal and early postnatal dentine Mn appeared protective against childhood behavioral problems, specifically hyperactivity and attention. Postnatal dentine Mn was associated with increased reporting of internalizing problems, specifically anxiety. At 6 months, a 1-unit increase (unit = 1 SD of log concentration) in Mn was associated with a 0.18-unit (unit = 1 SD of BASC-2 score) increase in internalizing symptoms score and a 0.25-unit increase in anxiety. Postnatal Pb was associated with increasing anxiety symptoms; at 12 months, a 1-unit increase in Pb was associated with a 0.4 unit increase in anxiety symptoms. When examined as a metal mixture, we observed two potential windows of susceptibility to increased anxiety symptoms: the first window (0-8 months) appeared driven by Mn, the second window (8-12 months) was driven by the metal mixture and dominated by Pb. A 1-unit increase in the mixture index was associated with a 0.7-unit increase in SD of anxiety symptoms.

CONCLUSIONS

Childhood behaviors may demonstrate postnatal windows of susceptibility to individual and mixed metal concentrations measured in deciduous teeth. Prenatal dentine Mn may be protective, while excessive early postnatal Mn may increase risk for adverse behaviors. In combination, higher concentrations of Mn, Zn and Pb may have an adverse impact on behavior.

Health Impacts of Exposure to Gaseous Pollutants and Particulate Matter in Beijing-A Non-Linear Analysis Based on the New Evidence

This paper aimed to estimate health risks focusing on respiratory diseases from exposure to gaseous multi-pollutants based on new data and revealed new evidence after the most stringent air pollution control plan in Beijing which was carried out in 2013. It used daily respiratory diseases outpatient data from a hospital located in Beijing with daily meteorological data and monitor data of air pollutants from local authorities. All data were collected from 2014 to 2016. Distributed lag non-linear model was employed. Results indicated that NO₂ and CO had positive association with outpatients number on the day of the exposure (1.045 (95% confidence interval (CI): 1.003, 1.089) for CO and 1.022 (95% CI: 1.008, 1.036) for NO₂) (and on the day after the exposure (1.026 (95% CI: 1.005, 1.048) for CO and 1.013 (95% CI: 1.005, 1.021) for NO₂). Relative risk (RR) generally declines with the number of lags; ozone produces significant effects on the first day (RR = 0.993 (95% CI: 0.989, 0.998)) as well as second day (RR = 0.995 (95% CI: 0.991, 0.999)) after the exposure, while particulate pollutants did not produce significant effects. Effects from the short-term exposure to gaseous pollutants were robust after controlling for particulate matters. Our results contribute to a comprehensive understanding of the dependencies between the change of air pollutants concentration and their health effects in Beijing after the implementation of promising air regulations in 2013. Results of the study can be used to develop relevant measures minimizing the adverse health consequences of air pollutants and supporting sustainable development of Beijing as well as other rapidly growing Asian cities.

Future temperature-related years of life lost projections for cardiovascular disease in Tianjin, China

It is widely accepted that temperatures is associated with cardiovascular mortality, however, few studies have explored the effects of temperature on years of life lost (YLL) from cardiovascular mortality in China under future global warming scenarios. Therefore, there is an urgent need to obtain projections of YLL from cardiovascular diseases. Here we applied nineteen global-scale climate models (GCMs) and three Representative Concentration Pathway emission scenarios (RCPs) in the 2050s and 2070s for temperature-related YLL projection in Tianjin, China. We found the relationships between daily maximum temperatures with YLL from cardiovascular mortality were basically U-shaped. We observed increasing net annual YLL across a range of multiple models under different climate scenarios, suggesting that increasing heat-related YLL from cardiovascular mortality could offset decreasing cold-related YLL from cardiovascular mortality. The largest temperature-related YLL from cardiovascular mortality were observed under the RCP8.5 scenario and increased more rapidly in the 2070s versus the 2050s. Monthly analyses of percent changes in YLL from cardiovascular mortality showed that the largest percent increases occurred from May to September. If warm adaptation occurs, only the adverse effects under RCP2.6 could be fully offset in both 2050 and 2070. Our exploration provided further evidence for the potential health impacts of global warming and highlighted that government should develop environmental policies for future health risks.

Hot Weather Impacts on New York City Restaurant Food Safety Violations and Operations

Previous studies have shown that higher ambient air temperature is associated with increased incidence of gastrointestinal illnesses, possibly as a result of leaving potentially hazardous food in the temperature danger zone for too long. However, little is known about the effect of hot weather on restaurant practices to maintain safe food temperatures. We examined hot weather impacts on restaurant food safety violations and operations in New York City using quantitative and qualitative methods. We used data from 64,661 inspections conducted among 29,614 restaurants during May to September, 2011 to 2015. We used Poisson time-series regression to estimate the cumulative relative risk (CRR) of temperature-related food safety violations across a range of daily maximum temperature (13 to 40°C [56 to 104°F]) over a lag of 0 to 3 days. We present CRRs for an increase in daily maximum temperature from the median (28°C [82°F]) to the 95th percentile (34°C [93°F]) values. Maximum temperature increased the risk of violations for cold food holding above 5°C (41°F) (CRR, 1.19; 95% CI, 1.14, 1.25) and insufficient refrigerated or hot holding equipment (CRR, 2.37; 95% CI, 2.02, 2.79). We also conducted focus groups among restaurant owners and managers to aid interpretation of findings and identify challenges or knowledge gaps that prevent hot weather preparedness. Focus group participants cited refrigeration issues as a common problem during hot weather. Participants expressed the need for more guidance on hot weather and power outages to be delivered concisely. Our findings suggest that hotter temperatures may compromise cold and hot food holding, possibly by straining refrigeration or other equipment. The findings have public health implications because holding potentially hazardous foods in the temperature danger zone allows foodborne pathogens to proliferate and increases risk for foodborne illness. Distribution of simple guidelines that can be easily accessed during emergencies could help restaurants respond better.

Temporal and seasonal variations of mortality burden associated with hourly temperature variability: A nationwide investigation in England and Wales

BACKGROUND

Sudden temperature change may elevate short-term mortality and remains an important global health threat in the context of climate change. To date, however, little available temperature-mortality evidence has taken into account both intra- and inter-day temperature variability (TV), thus largely limiting the comprehensive understanding of mortality burden due to unstable weather. Moreover, seasonal and temporal patterns in TV-mortality associations were sparsely discussed, nationally and regionally.

OBJECTIVES

We aimed to assess the nationwide association of all-cause mortality with hourly temperature variability (HTV), quantify HTV-attributable mortality, and further explore the temporal and seasonal variations of mortality burden due to HTV in United Kingdom.

METHODS

Fourteen-year time-series data on temperature and mortality were collected from 10 regions in England and Wales during 1993-2006, totally including 7,573,716 all-cause deaths. HTV was calculated from the standard deviation of hourly temperature records within two neighboring days. A three-stage analytic approach was adopted to assess HTV-associated mortality burden. We first applied a time-series quasi-Poisson regression to estimate region-specific HTV-mortality associations, then pooled these associations at the national level using a multivariate meta-analysis, and finally estimated the HTV-attributable mortality fraction and illustrated its seasonal and temporal variations by conducting season- and period-specific analyses based on time-varying distributed lag models.

RESULTS

We found strong evidence that large HTV exposure elevated short-term mortality risk in England and Wales, with a pooled estimate of 1.13% (95% confidence interval (CI): 0.88, 1.39) associated with a 1-°C increase in HTV. During the whole study period, HTV accounted for a national average attributable fraction of 2.52% (95% empirical confidence interval (eCI): 2.27, 2.76) of the total deaths. This HTV-attributable mortality estimate showed a significant temporal decrease (p < 0.001) from 2.72% (95% eCI: 2.58, 2.87) in 1993-99 to 2.28% (95% eCI: 2.13, 2.43) in 2000-06. Additionally, clear seasonal variations were observed for HTV-attributable mortality burden, with the largest estimate of 3.08% (95% eCI: 2.80, 3.38) in summer, followed by 2.71% (95% eCI: 2.44, 2.98) in spring, 2.40% (95% eCI: 2.16, 2.63) in autumn, and 2.00% (95% eCI: 1.81, 2.20) in winter.

CONCLUSIONS

Despite clear evidence observed for the reduction, mortality burden caused by temperature variability remained a great public health threat, especially in warm seasons. It highlighted the importance of specific interventions targeted to unstable weather as well as temperature extremes, so as to reduce climate-related mortality burden.

Association between Precipitation and Diarrheal Disease in Mozambique

Diarrheal diseases are a leading cause of morbidity and mortality in Africa. Although research documents the magnitude and pattern of diarrheal diseases are associated with weather in particular locations, there is limited quantification of this association in sub-Saharan Africa and no studies conducted in Mozambique. Our study aimed to determine whether variation in diarrheal disease was associated with precipitation in Mozambique. In secondary analyses we investigated the associations between temperature and diarrheal disease. We obtained weekly time series data for weather and diarrheal disease aggregated at the administrative district level for 1997-2014. Weather data include modeled estimates of precipitation and temperature. Diarrheal disease counts are confirmed clinical episodes reported to the Mozambique Ministry of Health (n = 7,315,738). We estimated the association between disease counts and precipitation, defined as the number of wet days (precipitation > 1 mm) per week, for the entire country and for Mozambique's four regions. We conducted time series regression analyses using an unconstrained distributed lag Poisson model adjusted for time, maximum temperature, and district. Temperature was similarly estimated with adjusted covariates. Using a four-week lag, chosen a priori, precipitation was associated with diarrheal disease. One additional wet day per week was associated with a 1.86% (95% CI: 1.05-2.67%), 1.37% (95% CI: 0.70-2.04%), 2.09% (95% CI: 1.01-3.18%), and 0.63% (95% CI: 0.11-1.14%) increase in diarrheal disease in Mozambique's northern, central, southern, and coastal regions, respectively. Our study indicates a strong association between diarrheal disease and precipitation. Diarrheal disease prevention efforts should target areas forecast to experience increased rainfall. The burden of diarrheal disease may increase with increased precipitation associated with climate change, unless additional health system interventions are undertaken.

Association of diurnal temperature range with daily mortality in England and Wales: A nationwide time-series study

BACKGROUND

Diurnal temperature range (DTR) reflects within-day temperature variability and is closely associated with climate change. In comparison to temperature extremes, up-to-date DTR-health evidence at the regional and national scales has been still very limited worldwide, especially in Europe.

OBJECTIVES

This study aimed to provide nationwide estimates for DTR-associated effects on mortality, and explore whether season and regional-level characteristics modify DTR-mortality relation in United Kingdom.

METHODS

Fourteen-year time-series data on weather and mortality were collected from 10 regions in England and Wales during 1993-2006, including 7,573,716 total deaths. A quasi-Poisson regression incorporated with distributed lag non-linear model was first applied to estimate region-specific DTR-mortality relationships. Then, a multivariate meta-analysis was employed to derive the pooled DTR effects at the national level. Also, the modifying effects of some regional characteristics (e.g., geographical and climatological) were examined by conducting multivariate meta-regression.

RESULTS

A non-linear DTR-mortality relationship was identified in UK. At the national level, increasing DTR raised the mortality risk observably when DTR exposure was below 25th percentile or above 90th percentile of DTR distribution, with an intermediate risk plateau indicating no associations. Extremely high DTR exhibited greater adverse effect estimates in hot season compared with in cold and transitional season, whereas entirely different association patterns were observed for the season-specific effects of extremely low DTR. In addition to season, regional latitudes, average temperature and humidity were also found to significantly modify DTR-mortality relationship.

CONCLUSIONS

Our study added strong evidence that extremely high DTR increased short-term mortality, whereas the effects of extremely low DTR exhibited entirely different seasonal patterns. Also, mortality vulnerability to DTR extremes varied greatly by regional latitudes and climate conditions.

Projecting temperature-related years of life lost under different climate change scenarios in one temperate megacity, China

Temperature has been associated with population health, but few studies have projected the future temperature-related years of life lost attributable to climate change. To project future temperature-related disease burden in Tianjin, we selected years of life lost (YLL) as the dependent variable to explore YLL attributable to climate change. A generalized linear model (GLM) and distributed lag non-linear model were combined to assess the non-linear and delayed effects of temperature on the YLL of non-accidental mortality. Then, we calculated the YLL changes attributable to future climate scenarios in 2055 and 2090. The relationships of daily mean temperature with the YLL of non-accident mortality were basically U-shaped. Both the daily mean temperature increase on high-temperature days and its drop on low-temperature days caused an increase of YLL and non-accidental deaths. The temperature-related YLL will worsen if future climate change exceeds 2 °C. In addition, the adverse effects of extreme temperature on YLL occurred more quickly than that of the overall temperature. The impact of low temperature was greater than that of high temperature. Men were vulnerable to high temperature compared with women. This analysis highlights that the government should formulate environmental policies to reach the Paris Agreement goal.

Association between temperature and maternal stress during pregnancy

BACKGROUND

Maternal psychological stress during pregnancy has essentially been conceptualized as a teratogen. However, little is known about the effect of temperature on maternal stress during pregnancy. The aim of this study is to investigate the relationship between temperature and maternal stress during pregnancy.

METHODS

In 2010, a total of 1931 eligible pregnant women were enrolled across Shanghai from four prenatal-care clinics during their mid-to-late pregnancy. Maternal life-event stress and emotional stress levels during pregnancy were assessed by the "Life Event Scale for Pregnant Women" (LESPW) and "Symptom Checklist-90-Revised Scale" (SCL-90-R), respectively. Exposure to ambient temperature was evaluated based on daily regional average in different moving average and lag days. The generalized estimating equations were used to evaluate the relationship between daily average temperature/temperature difference and maternal stress.

RESULTS

After adjusting for relevant confounders, an U-shaped relationship was observed between daily average temperature and maternal Global-Severity-Index (GSI) of the SCL-90-R. Cumulative exposures to extremely low temperatures (< P5, 1.4-10.5℃, lag 0-1 days, 0-2 days and 0-5 days) and extremely high temperatures (≥ P95, 31.2-34.1℃, lag 0-1 days and 0-2 days), and acute exposures to extremely low (lag day 0, 1, 2 and 3) and high (lag day 0, 1) temperatures, all induced higher risks of high GSI (the highest tertile), compared to the risk induced by exposed to an optimal temperature range (20-25℃) (P< 0.05). Increased temperature difference was associated with high maternal GSI (P< 0.05). However, non-significant associations were observed between daily average temperatures/temperature differences and maternal log-transferred LESPW scores.

CONCLUSIONS

Cumulative and acute exposures to extremely low/high temperatures may both induce emotional stress during pregnancy.

Temperature exposure during pregnancy and birth outcomes: An updated systematic review of epidemiological evidence

Seasonal patterns of birth outcomes have been observed worldwide, and there was increasing evidence that ambient temperature played as a trigger of adverse birth outcomes, such as preterm birth (PTB), low birth weight (LBW), and stillbirth. To systematically review updated epidemiological evidence about the relationship between temperature exposure during pregnancy and PTB, LBW, and stillbirth, we searched for related studies published in English from electronic databases and references of identified papers. We only included original articles that directly reported the effects of prenatal temperature exposure on birth outcomes. The characteristics and main findings of included studies were examined. A total of 36 epidemiological studies were finally included in this review. Most of these studies focused on PTB and LBW, while less attention has been paid to stillbirth that was relatively rare in the occurrence. Several designs including ecological (e.g., descriptive and time-series) and retrospective cohort studies (e.g., case-crossover and time-to-event) were applied to assess temperature effects on birth outcomes. Temperature metrics and exposure windows varied greatly in these investigations. Exposure to high temperature was generally found to be associated with PTB, LBW, and stillbirth, while several studies also reported the adverse impact of low temperature on birth outcomes of PTB and LBW. Despite no conclusive causality demonstrated, the current evidence for adverse effect on birth outcomes was stronger for heat than for cold. In summary, the evidence linking birth outcomes with ambient temperature was still very limited. Consequently, more related studies are needed worldwide and should be conducted in diversified climate zones, so as to further ascertain the association between temperature and birth outcomes. Future studies should focus on more sophisticated study designs, more accurate estimation of temperature exposure during pregnancy, and more efficient methods to find out the exposure windows, as well as cold-related effects on birth outcomes.