Vortragssitzung
Temperature and Health
Vorträge
Effects of Extreme Weather Events on Morbidity, Mortality and Economic Burden in German Inpatient Healthcare
Jona Frasch, Universitätsklinikum Hamburg-Eppendorf
Einleitung / Introduction
Climate change results in extreme weather events such as heat waves, cold waves, storms and torrential rain with higher frequency, intensity and longer duration in Germany. These hazards in turn lead to an increased morbidity and mortality, putting further health and health-economic pressure on an already strained healthcare system. In order to mitigate these negative effects, improve public health and increase the resilience of the healthcare system, a detailed understanding of the effects and their preconditions are paramount. We aimed to quantify the effects of heat and cold waves as frequent extreme weather events on mortality and the economic burden in German inpatient care.
Methode / Method
In this observational study, we combined the Diagnostic Related Group (DRG) Statistic of the German Federal Statistical Office containing anonymized information on all cases in German hospitals between 2010 and 2019 with daily measurements of weather stations of the German Meteorological Service. Using inverse distance weighting, we estimated weather conditions at the patients’ county of residence on the day of hospital admission. We calculated inpatient costs from payer’s perspective based on the DRG cost weights. Days were classified as heat days or cold days based on the maximum (> 30°) and minimum temperature (< -10°) respectively. Using Generalized Linear Models (GLM) with fixed effects, we estimated the effect of a heat or cold day on mortality and cost rates.
Ergebnisse / Results
In total, 173.917.600 hospital cases were observed within 10 years. Across counties, this mounted to 71.7 hospital cases per 100.000 residents (res.) per day. Patients had a mean age of 55.2 years and 52.8% were female. The average length of stay was 6.2 days. On average, 11.3 heat days and 5.1 cold days were counted per county per year. Initial analyses showed that mean inpatient mortality and costs were higher on heat days (mortality: 0.0119 per 100.000 res.; costs: 1856.58€ per 100.000 res.) than on non-heat days (mortality: 0.0118 per 100.000 res.; costs: 1808.63€ per 100.000 res.). Similarly, on cold days inpatient mortality and costs exceeded those on non-cold days, with 0.015 vs. 0.012 inpatient deaths per 100.000 res. and hospital costs of 2043.67€ vs. 1807.34€ per 100.000 res. respectively. Analyses with GLM affirmed the direction and statistical significance of these effects. Further analyses, e.g. with different thresholds for heat and cold days, are planned.
Zusammenfassung / Conclusion
This study corroborates findings from existing literature on the effect of heat and cold waves on hospital mortality and broadens our understanding of the health-economic impact of climate change. Its effects on public health are substantial and pervasive, putting further emphasis on the necessity of climate change prevention and adaption measures.
AutorInnen
Jona J. Frasch, Universitätsklinikum Hamburg-Eppendorf
Hans-Helmut König, Universitätsklinikum Hamburg-Eppendorf
Claudia Konnopka, Universitätsklinikum Hamburg-Eppendorf
Effects of Droughts on Recall and Health
Soschia Karimi, Leibniz Universität Hannover
Einleitung / Introduction
A growing body of literature indicates that heat stress and precipitation deficencies can pose a critical threat to human health, particularly in less developed countries with low coping capacities and high exposure. A frequently used source to identify drought periods is self-reported data and as such it can be subject to misreporting. The aims of this study for rural Thailand are twofold. First, we shed light on the recall of drought events by linking survey data with objective meteorological data. Here, an anomaly in the survey design serves as a natural experiment. Due to random exclusion, the interview interval for one province was shorter than in the other province. We find that a shorter time interval between surveys has a large positive effect on households correctly reporting a drought event. Second, we find a strong negative association between droughts and health applying high-dimensional fixed-effects regressions over seven panel waves. We also find that households with internet or smartphone access exhibit milder negative health impacts of a drought. Additional to better access to information, a larger network potentially helps affected households to get support from network members who were not affected by the drought.
AutorInnen
Annika Herr, Leibniz Universität Hannover
Julian Wichert, Leibniz Universität Hannover
Cold temperature at conception and metabolic outcomes in adulthood: Evidence from the UK Biobank
Timo Münz, Johannes Gutenberg-Universität Mainz
Einleitung / Introduction
Animal studies and correlational evidence among humans have linked paternal cold exposure around conception to an increased amount of active brown fat in adulthood, which is associated with improved metabolic outcomes such as lower risks for obesity and diabetes. However, causal evidence on humans is lacking. At the same time, this topic has a high public health relevance by providing evidence on the long-term externalities of rising global temperatures and climate change.
Methode / Method
We combine detailed historical UK temperature data from MIDAS with UK Biobank individual-level health data on ca. 450,000 individuals born between 1934 and 1971. Metabolic health is measured in terms of anthropometric measures (BMI and waist circumference) and biomarkers for cardiovascular disease (total cholesterol and triglycerides). We estimate the month of conception by calculating backwards from participants’ dates of birth. Average temperature in this month and location is calculated, and exposure is defined as the deviation from the long-run historical mean for the given location and month. Temperature deviations around conception are then assigned to each participant by the nearest neighbor approach. In an additional analysis, temperature deviations are assigned to individuals by taking an inverse-distance weighted average of recordings from all available stations which are nearby. Our regressions control for sex, age and year of assessment. Furthermore, including region-by-month of birth fixed effects allows us to exploit quasi-random variation in temperatures per geographic area and thus provide causal estimates.
Ergebnisse / Results
Our findings suggest that individuals who were prenatally exposed to colder temperatures have a lower BMI (-0.011, 95% CI: -0.022, -0.001, p-value: 0.034) and waist circumference (-0.024, 95% CI: -0.050, 0.002, p-value: 0.068). Further analyses on key markers in blood assays showed that individuals exposed to colder temperatures around conception had lower levels of triglycerides (-0.002, 95% CI: -0.005, -0.000, p-value: 0.043) and cholesterol (-0.008, 95% CI: -0.012, -0.005, p-value: < 0.001).
Zusammenfassung / Conclusion
Our results show that cold exposure around conception can result in positive effects on metabolic health in later life. In the future, fewer people will be periconceptually exposed to cold spells as winters are getting warmer. Thus, this protective effect may diminish. Climate change and improved house insulation may therefore contribute to the pandemic in obesity and metabolic disturbances. Moreover, our findings demonstrate that adulthood metabolic health can be influenced by comparably subtle prenatal shocks such as temperature at conception.