We thank Dr. Hernandez1 for his interest in our article,2 which reported associations between extreme temperatures and mortality for 15 cities in Northeast Asia. Dr. Hernandez first commented that the categorization for the cause of mortality (i.e., cardiovascular, respiratory, and others) is quite broad and future research should be directed to investigate the causes related to weather change with more refined categories. Although sub-categorized mortality data could have been used in our study, the death counts for each sub-cause are small, which would produce larger uncertainty in results.2(Figure 4) This may be a limitation not only in our study but also in many other time-series studies for environmental exposure (e.g., weather and air pollution) and mortality.3,4 Perhaps focusing on vulnerable populations in larger spatial units would allow collection of enough mortality data to obtain reliable results, although exposure calculation should also be refined to allow for the larger spatial unit. We agree that future study of more cause-specific outcomes is valuable. In addition, in pointing out one of our results—that annual average daily mean barometric pressure (hereafter, pressure) is negatively associated with cold effects,2(Table) Dr. Hernandez noted that it may imply a protective effect of pressure on cold-related mortality. Although increased pressure was associated with decreased cold effect (2.36% decrease in cold-related mortality per interquartile range increase in pressure), this result should be interpreted with caution. The largest five values for pressure correspond to five cities, all in Korea, where the smallest five cold effects were estimated.2(Figure S4) Therefore, the estimated regression line with a negative slope may have been driven by a country effect. To determine the meaning of this negative association, a larger number of cities should be considered in Korea as well as in other countries ensuring more variations in pressure and cold effects both between and within countries. However, the observed association merits further investigation for a possible modification of the relation between cold temperature and mortality by pressure. We examined this hypothesis by adding an interaction between temperature and pressure in our city-specific models.2(Equation 1) Effects of cold on all-cause and cause-specific mortality were estimated to change across pressure though the directions vary across cities. Our findings may relate to Dr. Hernandez’s comment that pressure may cause physiologic changes through oxygen saturation5 but causal pathways from weather changes to mortality are not well understood. In previous time series, studies for environmental exposure and mortality, pressure was mostly considered as a potential confounder that covaries with exposure and mortality.3,4 Our findings, along with Dr. Hernandez’s comments, however, suggest that pressure may play more complex roles in the pathway, providing new information toward disentangling specific causes of mortality associated with environmental stressors. Therefore, future studies should expand to incorporate the complexity of environmental conditions with more attention to the uncertain role of pressure. Yeonseung Chung Department of Mathematical Sciences Korea Advanced Institute of Science and Technology Daejeon, South Korea Michelle L. Bell School of Forestry and Environmental Studies Yale University New Haven, CT Ho Kim Department of Biostatistics and Epidemiology Graduate School of Public Health & Institute of Health and Environment Seoul National University Seoul, South Korea [email protected]
Read full abstract
Sep 14, 2023
S Nayak + 3
Open Access
