Searching for the best modeling specification for assessing the effects of temperature and humidity on health: a time series analysis in three European cities

Heat, humidity and health impacts: how causal diagrams can help tell the complex story

The global health burden associated with exposure to heat is a grave concern and is projected to further increase under climate change. While physiological studies have demonstrated the role of humidity alongside temperature in exacerbating heat stress for humans, epidemiological findings remain conflicted. Understanding the intricate relationships between heat, humidity, and health outcomes is crucial to inform adaptation and drive increased global climate change mitigation efforts. This article introduces 'directed acyclic graphs' (DAGs) as causal models to elucidate the analytical complexity in observational epidemiological studies that focus on humid-heat-related health impacts. DAGs are employed to delineate implicit assumptions often overlooked in such studies, depicting humidity as a confounder, mediator, or an effect modifier. We also discuss complexities arising from using composite indices, such as wet-bulb temperature. DAGs representing the health impacts associated with wet-bulb temperature help to understand the limitations in separating the individual effect of humidity from the perceived effect of wet-bulb temperature on health. General examples for regression models corresponding to each of the causal assumptions are also discussed. Our goal is not to prioritize one causal model but to discuss the causal models suitable for representing humid-heat health impacts and highlight the implications of selecting one model over another. We anticipate that the article will pave the way for future quantitative studies on the topic and motivate researchers to explicitly characterize the assumptions underlying their models with DAGs, facilitating accurate interpretations of the findings. This methodology is applicable to similarly complex compound events.

Increase in summertime ambient temperature is associated with decreased sick leave risk in Helsinki, Finland

OBJECTIVES

Climate change has increased attention to the health effects of high ambient temperatures and heatwaves worldwide. Both cause-specific mortality and hospital admissions are studied widely, mainly concentrating on warmer climates, but studies focusing on more subtle health effects and cold climates lack. This study investigated the effect of summertime daily ambient temperatures and heatwaves on sick leaves in the employed population in Helsinki, Finland, a Nordic country with a relatively cold climate.

METHODS

We obtained from the City of Helsinki personnel register data on sick leaves for the summer months (June-August) of 2002-2017. We estimated the overall cumulative association of all and short (maximum 3-day) sick leaves with daily mean temperature over a 21-day lag period using a negative binomial regression model coupled with a penalized distributed lag non-linear model (penalized DLNM). The association of sick leaves with heatwaves (cut-off temperature 20.8 °C), and prolonged heatwaves, was estimated using a negative binomial regression model coupled with DLNM. We adjusted the time series model for potential confounders, such as air pollution, relative humidity, time trends, and holidays.

RESULTS

Increasing daily temperature tended to be associated with decreased overall cumulative risk of sick leaves and short sick leaves over a 21-day lag period. In addition, heatwaves and prolonged heatwaves were associated with decreased overall cumulative risk of sick leaves compared to all other summer days: RR 0.87 (95 % CI 0.78 to 0.97) and RR 0.83 (95 % CI 0.70 to 0.98), respectively.

CONCLUSIONS

This research suggests that summertime daily temperatures that are high for this northern location have protective effects on the health of the working population.

Associations and burdens of relative humidity with cause-specific mortality in three Chinese cities

This study aimed to investigate the association between relative humidity (RH) and various cause of mortality, and then quantify the RH-related mortality fraction of low and high RH under the assumption that causal effects exist. Daily cause-specific mortality counts from 2008 to 2011, and contemporaneous meteorological data in three Chinese cities were collected. Distributed lag nonlinear models were adopted to quantify the nonlinear and delayed effects of RH on mortality risk. Low and high RH were defined as RH lower or higher than the minimum mortality risk RH (MMRH), respectively. Corresponding RH-related mortality fractions were calculated in the explanatory analysis. From the three cities, 736,301 deaths were collected. RH (mean ± standard deviation) were 50.9 ± 20.0 for Beijing, 75.5 ± 8.6 for Chengdu, and 70.8 ± 14.6 for Nanjing. We found that low RH in Beijing and high RH (about 80-90%) in Chengdu was associated with increased all-cause mortality risk. Both low and high RH may increase the CVD mortality risk in Beijing. Both low and high (about 80-85%) RH may increase the COPD mortality risk in Chengdu. Low RH (about < 45%) was associated with increased diabetes mortality risk in Nanjing. Effects of extreme low and extreme high RH were delayed in these cities, except that extreme low effects on COPD mortality appeared immediately in Chengdu. The effects of extreme low RH are higher than that of the extreme high RH in Beijing and Nanjing, while contrary in Chengdu. Finally, under the causal effect assumption, 6.80% (95% eCI: 2.90, 10.73) all-cause mortality and 12.48% (95% eCI: 7.17, 16.80) CVD deaths in Beijing, 9.59% (95% eCI: 1.38, 16.88) COPD deaths in Chengdu, and 23.79% (95% eCI: 0.92, 387.93) diabetes mortality in Nanjing were attributable to RH. Our study provided insights into RH-mortality risk, helped draw relative intervention policies, and is also significant for future predictions of climate change effects under different scenarios.

Forecasting of non-accidental, cardiovascular, and respiratory mortality with environmental exposures adopting machine learning approaches

Environmental exposure constantly changes with time and various interactions that can affect health outcomes. Machine learning (ML) or deep learning (DL) algorithms have been used to solve complex problems, such as multiple exposures and their interactions. This study developed predictive models for cause-specific mortality using ML and DL algorithms with the daily or hourly measured meteorological and air pollution data. The ML algorithm improved the performance compared to the conventional methods, even though the optimal algorithm depended on the adverse health outcomes. The best algorithms were extreme gradient boosting, ridge, and elastic net, respectively, for non-accidental, cardiovascular, and respiratory mortality with daily measurement; they were superior to the generalized additive model reducing a mean absolute error by 4.7%, 4.9%, and 16.8%, respectively. With hourly measurements, the ML model tended to outperform the conventional models, even though hourly data, instead of daily data, did not enhance the performance in some models. The proposed model allows a better understanding and development of robust predictive models for health outcomes using multiple environmental exposures.

The need for location-specific biometeorological indexes in Taiwan

Objective

As most available biometeorological indexes were developed decades ago in western countries, the benefit of using these indexes to study the effect of weather on human health in modern eastern countries is questionable. This study aimed to reconfirm the effectiveness of applying these biometeorological indexes when analyzing demand for daily emergency ambulance services (EAS) in Taipei.

Methods

More than 370,000 EAS usage records were analyzed in this study. The records were first allotted into different time-series data by age, gender, triage level, and case nature (trauma/non-trauma) in order to represent different kinds of daily EAS demand. They were then regressed on biometeorological indexes [Apparent Temperature (AT) and Net Effective Temperature (NET)]; the indexes' additional descriptive power to describe the daily EAS demand over traditional weather factors was then assessed.

Results

No significant difference was observed in the descriptive powers in terms of effect on daily EAS demand of the biometeorological indexes and traditional weather factors. The largest improvement on the regression models' adjusted-R2 using NET and AT was only 0.008.

Conclusion

It may not be a good idea to make direct use of the biometeorological indexes developed in western countries decades ago. Taiwan should have a tailor-made biometeorological index for a better representation of its unique situation.

Short-Term Effects of Apparent Temperature on Cause-Specific Mortality in the Urban Area of Thessaloniki, Greece

Although there is a growing interest in the association between ambient temperatures and mortality, little evidence is available for Thessaloniki, the second largest city of Greece. In this study, we present an assessment of the effects of temperature on daily mortality from 2006 to 2016 in the urban area of Thessaloniki, by describing the exposure-lag-response association between temperature and cause-specific mortality with the use of a distributed lag non-linear model (DLNM). A J-shaped relationship was found between temperature and mortality. The highest values of risk were evident for respiratory (RR > 10) and cardiovascular causes (RR > 3), probably due to the fact that health status of individuals with chronic respiratory and cardiovascular diseases rapidly deteriorates during hot periods. Cold effects had longer lags of up to 15 days, whereas heat effects were short-lived, up to 4 days. Percentage change in all- and cause-specific mortality per 1 °C change above and below Minimum Mortality Temperature showed a larger increase for all-cause mortality in heat (1.95%, 95% CI: 1.07–2.84), in contrast to a smaller increase in cold (0.54%, 95% CI: 0, 1.09). Overall, 3.51% of all-cause deaths were attributable to temperature, whereas deaths attributed to heat (2.34%) were more than deaths attributed to cold (1.34%). The findings of this study present important evidence for planning public-health interventions, to reduce the health impact of extreme temperatures.

What is the value of delivering on time?

Purpose

Delivery punctuality is essential in supply chain management, yet the cost of untimely delivery is usually assumed to be given or based on intuition and not quantified by facts.

Design/methodology/approach

The authors used a data set containing detailed transaction data for a nine-year period on orders and deliveries of sport goods. The methodology is based on applying a polynomial distributed lag model to longitudinal data on supply chain transactions.

Findings

The results indicate that small delivery delays up to two weeks decrease the sales by maximum 10% during a period of 3–4 weeks. Longer delays, up to 45 days, have a larger negative effect on sales, which can also last longer. For this case company, the estimated lost sales due to late deliveries (=5 days) were 5.1% of the delivery value. The longer delays got, the large the cost was: delays at least 45 days long were the most costly causing almost 40% of the estimated lost sales.

Practical implications

This study offers a methodology for quantifying lost sales due to delivery delays and estimating how long the poor delivery performance affects retailers' order behaviour.

Originality/value

The results give a quantitative decision-making tool for supply chain managers to estimate the profitability of investments in the supply chain performance, especially on improving punctuality.

The Role of Humidity in Associations of High Temperature with Mortality: A Multicountry, Multicity Study

BACKGROUND

There is strong experimental evidence that physiologic stress from high temperatures is greater if humidity is higher. However, heat indices developed to allow for this have not consistently predicted mortality better than dry-bulb temperature.

OBJECTIVES

We aimed to clarify the potential contribution of humidity an addition to temperature in predicting daily mortality in summer by using a large multicountry dataset.

METHODS

In 445 cities in 24 countries, we fit a time-series regression model for summer mortality with a distributed lag nonlinear model (DLNM) for temperature (up to lag 3) and supplemented this with a range of terms for relative humidity (RH) and its interaction with temperature. City-specific associations were summarized using meta-analytic techniques.

RESULTS

Adding a linear term for RH to the temperature term improved fit slightly, with an increase of 23% in RH (the 99th percentile anomaly) associated with a 1.1% [95% confidence interval (CI): 0.8, 1.3] decrease in mortality. Allowing curvature in the RH term or adding terms for interaction of RH with temperature did not improve the model fit. The humidity-related decreased risk was made up of a positive coefficient at lag 0 outweighed by negative coefficients at lags of 1-3 d. Key results were broadly robust to small model changes and replacing RH with absolute measures of humidity. Replacing temperature with apparent temperature, a metric combining humidity and temperature, reduced goodness of fit slightly.

DISCUSSION

The absence of a positive association of humidity with mortality in summer in this large multinational study is counter to expectations from physiologic studies, though consistent with previous epidemiologic studies finding little evidence for improved prediction by heat indices. The result that there was a small negative average association of humidity with mortality should be interpreted cautiously; the lag structure has unclear interpretation and suggests the need for future work to clarify. https://doi.org/10.1289/EHP5430.

Assessment of the combination of temperature and relative humidity on kidney stone presentations

Temperature and relative humidity have opposing effects on evaporative water loss, the likely mediator of the temperature-dependence of nephrolithiasis. However, prior studies considered only dry-bulb temperatures when estimating the temperature-dependence of nephrolithiasis. We used distributed lag non-linear models and repeated 10-fold cross-validation to determine the daily temperature metric and corresponding adjustment for relative humidity that most accurately predicted kidney stone presentations during hot and cold periods in South Carolina from 1997 to 2015. We examined three metrics for wet-bulb temperatures and heat index, both of which measure the combination of temperature and humidity, and for dry-bulb temperatures: (1) daytime mean temperature; (2) 24-h mean temperature; and (3) most extreme 24-h temperature. For models using dry-bulb temperatures, we considered four treatments of relative humidity. Among 188,531 patients who presented with kidney stones, 24-h wet bulb temperature best predicted kidney stone presentation during summer. Mean cross-validated residuals were generally lower in summer for wet-bulb temperatures and heat index than the corresponding dry-bulb temperature metric, regardless of type of adjustment for relative humidity. Those dry-bulb models that additionally adjusted for relative humidity had higher mean residuals than other temperature metrics. The relative risk of kidney stone presentations at the 99th percentile of each temperature metric compared to the respective median temperature in summer months differed by temperature metric and relative humidity adjustment, and ranged from an excess risk of 8-14%. All metrics performed similarly in winter. The combination of temperature and relative humidity determine the risk of kidney stone presentations, particularly during periods of high heat and humidity. These results suggest that metrics that measure moist heat stress should be used to estimate the temperature-dependence of kidney stone presentations, but that the particular metric is relatively unimportant.

Changes in relative fit of human heat stress indices to cardiovascular, respiratory, and renal hospitalizations across five Australian urban populations

Various human heat stress indices have been developed to relate atmospheric measures of extreme heat to human health impacts, but the usefulness of different indices across various health impacts and in different populations is poorly understood. This paper determines which heat stress indices best fit hospital admissions for sets of cardiovascular, respiratory, and renal diseases across five Australian cities. We hypothesized that the best indices would be largely dependent on location. We fit parent models to these counts in the summers (November-March) between 2001 and 2013 using negative binomial regression. We then added 15 heat stress indices to these models, ranking their goodness of fit using the Akaike information criterion. Admissions for each health outcome were nearly always higher in hot or humid conditions. Contrary to our hypothesis that location would determine the best-fitting heat stress index, we found that the best indices were related largely by health outcome of interest, rather than location as hypothesized. In particular, heatwave and temperature indices had the best fit to cardiovascular admissions, humidity indices had the best fit to respiratory admissions, and combined heat-humidity indices had the best fit to renal admissions. With a few exceptions, the results were similar across all five cities. The best-fitting heat stress indices appear to be useful across several Australian cities with differing climates, but they may have varying usefulness depending on the outcome of interest. These findings suggest that future research on heat and health impacts, and in particular hospital demand modeling, could better reflect reality if it avoided "all-cause" health outcomes and used heat stress indices appropriate to specific diseases and disease groups.

Estimated Effect of Temperature on Years of Life Lost: A Retrospective Time-Series Study of Low-, Middle-, and High-Income Regions

BACKGROUND

Numerous studies have reported a strong association between temperature and mortality. Additional insights can be gained from investigating the effects of temperature on years of life lost (YLL), considering the life expectancy at the time of death.

OBJECTIVES

The goal of this work was to assess the association between temperature and YLL at seven low-, middle-, and high-income sites.

METHODS

We obtained meteorological and population data for at least nine years from four Health and Demographic Surveillance Sites in Kenya (western Kenya, Nairobi), Burkina Faso (Nouna), and India (Vadu), as well as data from cities in the United States (Philadelphia, Phoenix) and Sweden (Stockholm). A distributed lag nonlinear model was used to estimate the association of daily maximum temperature and daily YLL, lagged 0-14 d. The reference value was set for each site at the temperature with the lowest YLL.

RESULTS

Generally, YLL increased with higher temperature, starting day 0. In Nouna, the hottest location, with a minimum YLL temperature at the first percentile, YLL increased consistently with higher temperatures. In Vadu, YLL increased in association with heat, whereas in Nairobi, YLL increased in association with both low and high temperatures. Associations with cold and heat were evident for Phoenix (stronger for heat), Stockholm, and Philadelphia (both stronger for cold). Patterns of associations with mortality were generally similar to those with YLL.

CONCLUSIONS

Both high and low temperatures are associated with YLL in high-, middle-, and low-income countries. Policy guidance and health adaptation measures might be improved with more comprehensive indicators of the health burden of high and low temperatures such as YLL. https://doi.org/10.1289/EHP1745.

Humidity May Modify the Relationship between Temperature and Cardiovascular Mortality in Zhejiang Province, China

Background: The evidence of increased mortality attributable to extreme temperatures is widely characterized in climate-health studies. However, few of these studies have examined the role of humidity on temperature-mortality association. We investigated the joint effect between temperature and humidity on cardiovascular disease (CVD) mortality in Zhejiang Province, China. Methods: We collected data on daily meteorological and CVD mortality from 11 cities in Zhejiang Province during 2010–2013. We first applied time-series Poisson regression analysis within the framework of distributed lag non-linear models to estimate the city-specific effect of temperature and humidity on CVD mortality, after controlling for temporal trends and potential confounding variables. We then applied a multivariate meta-analytical model to pool the effect estimates in the 11 cities to generate an overall provincial estimate. The joint effects between them were calculated by the attributable fraction (AF). The analyses were further stratified by gender, age group, education level, and location of cities. Results: In total, 120,544 CVD deaths were recorded in this study. The mean values of temperature and humidity were 17.6 °C and 72.3%. The joint effect between low temperature and high humidity had the greatest impact on the CVD death burden over a lag of 0–21 days with a significant AF of 31.36% (95% eCI: 14.79–38.41%), while in a condition of low temperature and low humidity with a significant AF of 16.74% (95% eCI: 0.89, 24.44). The AFs were higher at low temperature and high humidity in different subgroups. When considering the levels of humidity, the AFs were significant at low temperature and high humidity for males, youth, those with a low level of education, and coastal area people. Conclusions: The combination of low temperature and high humidity had the greatest impact on the CVD death burden in Zhejiang Province. This evidence has important implications for developing CVD interventions.

Adaptation to Climate Change: A Comparative Analysis of Modeling Methods for Heat-Related Mortality

BACKGROUND

Multiple methods are employed for modeling adaptation when projecting the impact of climate change on heat-related mortality. The sensitivity of impacts to each is unknown because they have never been systematically compared. In addition, little is known about the relative sensitivity of impacts to "adaptation uncertainty" (i.e., the inclusion/exclusion of adaptation modeling) relative to using multiple climate models and emissions scenarios.

OBJECTIVES

This study had three aims: a) Compare the range in projected impacts that arises from using different adaptation modeling methods; b) compare the range in impacts that arises from adaptation uncertainty with ranges from using multiple climate models and emissions scenarios; c) recommend modeling method(s) to use in future impact assessments.

METHODS

We estimated impacts for 2070-2099 for 14 European cities, applying six different methods for modeling adaptation; we also estimated impacts with five climate models run under two emissions scenarios to explore the relative effects of climate modeling and emissions uncertainty.

RESULTS

The range of the difference (percent) in impacts between including and excluding adaptation, irrespective of climate modeling and emissions uncertainty, can be as low as 28% with one method and up to 103% with another (mean across 14 cities). In 13 of 14 cities, the ranges in projected impacts due to adaptation uncertainty are larger than those associated with climate modeling and emissions uncertainty.

CONCLUSIONS

Researchers should carefully consider how to model adaptation because it is a source of uncertainty that can be greater than the uncertainty in emissions and climate modeling. We recommend absolute threshold shifts and reductions in slope. https://doi.org/10.1289/EHP634.

Associations between environmental factors and hospital admissions for sickle cell disease

Sickle cell disease is an increasing global health burden. This inherited disease is characterized by a remarkable phenotypic heterogeneity, which can only partly be explained by genetic factors. Environmental factors are likely to play an important role but studies of their impact on disease severity are limited and their results are often inconsistent. This study investigated associations between a range of environmental factors and hospital admissions of young patients with sickle cell disease in London and in Paris between 2008 and 2012. Specific analyses were conducted for subgroups of patients with different genotypes and for the main reasons for admissions. Generalized additive models and distributed lag non-linear models were used to assess the magnitude of the associations and to calculate relative risks. Some environmental factors significantly influence the numbers of hospital admissions of children with sickle cell disease, although the associations identified are complicated. Our study suggests that meteorological factors are more likely to be associated with hospital admissions for sickle cell disease than air pollutants. It confirms previous reports of risks associated with wind speed (risk ratio: 1.06/standard deviation; 95% confidence interval: 1.00–1.12) and also with rainfall (1.06/standard deviation; 95% confidence interval: 1.01–1.12). Maximum atmospheric pressure was found to be a protective factor (0.93/standard deviation; 95% confidence interval: 0.88–0.99). Weak or no associations were found with temperature. Divergent associations were identified for different genotypes or reasons for admissions, which could partly explain the lack of consistency in earlier studies. Advice to patients with sickle cell disease usually includes avoiding a range of environmental conditions that are believed to trigger acute complications, including extreme temperatures and high altitudes. Scientific evidence to support such advice is limited and sometimes confusing. This study shows that environmental factors do explain some of the variations in rates of admission to hospital with acute symptoms in sickle cell disease, but the associations are complex, and likely to be specific to different environments and the individual’s exposure to them. Furthermore, this study highlights the need for prospective studies with large numbers of patients and standardized protocols across Europe.