Internet searches and heat-related emergency department visits in the United States

Vulnerability assessment of heat waves within a risk framework using artificial intelligence

Current efforts to adapt to climate change are not sufficient to reduce projected impacts. Vulnerability assessments are essential to allocate resources where they are needed most. However, current assessments that use principal component analysis suffer from multiple shortcomings and are hard to translate into concrete actions. To address these issues, this article proposes a novel data-driven vulnerability assessment within a risk framework. The framework is based on the definitions from the Sixth Assessment Report of the Intergovernmental Panel on Climate Change, but some definitions, such as sensitivity and adaptive capacity, are clarified. Heat waves that occurred between 2001 and 2018 in Quebec (Canada) are used to validate the framework. The studied impact is the daily mortality rates per cooling degree-days (CDD) region. A vulnerability map is produced to identify the distributions of summer mortality rates in aggregate dissemination areas within each CDD region. Socioeconomic and environmental variables are used to calculate impact and vulnerability. We compared abilities of AutoGluon (an AutoML framework), Gaussian process, and deep Gaussian process to model the impact and vulnerability. We offer advice on how to avoid common pitfalls with artificial intelligence and machine-learning algorithms. Gaussian process is a promising approach for supporting the proposed framework. SHAP values provide an explanation for the model results and are consistent with current knowledge of vulnerability. Recommendations are made to implement the proposed framework quantitatively or qualitatively.

Heat wave exposure and increased heat-related hospitalizations in young children in South Korea: A time-series study

BACKGROUND

Numerous studies have investigated the association between heat wave exposure increased heat-related hospitalizations in the general population. However, little is known about heat-related morbidity in young children who are more vulnerable than the general population. Therefore, we aimed to evaluate the association between hospitalization for heat-related illness in children and heat wave exposure in South Korea.

METHODS

We used the National Health Insurance Service (NHIS) database, which provides medical records from 2015 to 2019 in South Korea. We defined daily hospitalizations for heat-related illness of children younger than five years during the summer period (June to August). We considered the definition of heat waves considering the absolute temperature and percentile. A total of 12 different heat waves were used. A time-series analysis was used to investigate the association between heat wave exposure and heat-related hospitalization among children younger than five years. We used a two-stage design involving a meta-analysis after modeling by each region.

RESULTS

We included 16,879 daily heat-related hospitalizations among children younger than five years. Overall, heat wave exposure within two days was most related for heat-related hospitalizations in young children. The relative risk (RR) due to heat wave exposure within two days (lag2) (12 definitions: 70th to 90th percentile of maximum temperature) ranged from 1.038 (95% confidence interval (CI): 0.971, 1.110) to 1.083 (95% CI: 1.036, 1.133). We found that boys were more vulnerable to heat exposure than girls. In addition, we found that urban areas were more vulnerable to heat exposure than rural areas.

CONCLUSIONS

In our study, heat wave exposure during summer was found to be associated with an increased risk of hospitalization for heat-related illness among children younger than five years. Our findings suggest the need for summer heat wave management and prevention for children.

Rescue service deployment data as an indicator of heat morbidity in Frankfurt / Main, Germany (2014-2022) - Trend association with various heat exposure indicators and considerations for outreach

Many publications dealt with the monitoring of heat-related mortality. Fewer analyses referred to indicators of heat-related morbidity. The aim of this work was to describe the heat-related morbidity using rescue service data from the city of Frankfurt/Main, Germany for the time period 2014-2022, with regard to the questions: 1) How do rescue service deployments develop over the years? Is there a trend identifiable towards a decrease in deployments over the years, e.g. as an effect of either (physiological) adaptation of the population or of the measures for prevention of heat-related morbidity? 2) Which heat parameters (days with a heat warning, heat days, heat weeks, heat waves) are most strongly associated with heat-related morbidity in terms of rescue service deployments and might therefore be additionally used as an easily communicable and understandable heat-warning indicator? Rescue service data were provided by the interdisciplinary medical supply compass system "IVENA" and adjusted for population development including age development. The effect of various indicators for heat exposure, such as days with a heat warning from the German meteorological service based on the scientific concept of "perceived heat", heat days, heat wave days and heat week days on different endpoints for heat morbidity (deployments in total as well as for heat associated diagnoses) was calculated using both difference-based (difference ± 95% CI) and ratio-based (ratio ± 95% CI) effect estimators. Rescue services deployments in summer months increased overall from 2014 to 2022 in all age groups over the years (2698 to 3517/100.000 population). However, there was a significant decrease in 2020, which could be explained by the special situation of the COVID-19 pandemic, probably caused by the absence of tourists and commuters from the city. In addition, no data are available on the actual implementation of the measures by the population. Therefore, an effect of the measures taken to prevent heat-associated morbidity in Frankfurt am Main could not be directly demonstrated, and our first question cannot be answered on the basis of these data. Almost all heat definitions used for exposure (day with a heat warning, heat day, heat wave day, heat week day) showed significant effects on heat-associated diagnoses in every year. When analysing the effect on all deployments, the effect was in part strongly dependent on individual years: Heat wave days and heat week days even showed negative effects in some years. The definition heat day led to a significant increase in rescue service deployments in all single years between 2014 and 2022 (ratio 2014-2022 1.09 (95CI 1.07-1.11); with a range of 1.05 (95CI 1.01-1.09) in 2020 and 1.14 (95CI 1.08-1.21) in 2014), this was not the case for days with a heat warning (ratio 2014-2022 1.04 (95CI1.02-1.05); with a range of 1.01 (95CI 0.97-1.05) in 2017 and 1.16 (95CI 1.10-1.23). Thus being not inferior to the heat warning day, the "heat day" defined as ≥32 °C maximum temperature, easily obtainable from the weather forecast, can be recommended for the activities of the public health authorities (warning, surveillance etc.) regarding heat health action planning.

Use of smartphone mobility data to analyze city park visits during the COVID-19 pandemic

INTRODUCTION

The COVID-19 pandemic focused attention on city parks as important public resources. However, monitoring park use over time poses practical challenges. Thus, pandemic-related trends are unknown.

METHODS

We analyzed monthly mobility data from a large panel of smartphone devices, to assess park visits from January 2018 to November 2020 in the 50 largest U.S. cities.

RESULTS

In our sample of 11,890 city parks, visits declined by 36.0 % (95 % CI [27.3, 43.6], p < 0.001) from March through November 2020, compared to prior levels and trends. When we segmented the COVID-19 period into widespread closures (March-April) and reopenings (May-November), we estimated a small rebound in visits during reopenings. In park service areas where a greater proportion of residents were White and highincome, this rebound effect was larger.

CONCLUSIONS

Smartphone data can address an important gap for monitoring park visits. Park visits declined substantially in 2020 and disparities appeared to increase.