ABSTRACTS Abstracts published in EPIDEMIOLOGY have been reviewed by the societies at whose meetings the abstracts have been accepted for presentation. These abstracts have not undergone review by the Editorial Board of EPIDEMIOLOGY. ISEE 22nd Annual Conference, Seoul, Korea, 28 August–1 September 2010 CLIMATE CHANGE AND ENVIRONMENTAL HEALTH O-29A1-1 Heat-related Mortality and Heat Watch-warning Systems in the United States: Recent Developments Scott Sheridan, 1 Laurence Kalkstein, 2 Adam Kalkstein, 3 and Scott Greene 4 1 Kent State University, Kent, OH; 2 University of Miami, Coral Gables, FL; 3 United States Military Academy, West Point, NY; and 4 University of Oklahoma, Norman, OK. Background/Aims: Heat watch-warning systems have increased in coverage, and in some US locales, have now been operational for over 15 years. Heat-related mortality has decreased over the last few decades, partially as a result of increased awareness, but it is still statistically significant in many locations. We present recent research results on several projects that assess these changes as well as projected future changes, and the spatiotemporal variability in heat-related mortality across the United States. Methods: Trends are assessed using both the synoptic climatological methodology, utilized in the authors’ heat watch-warning systems, and the Heat Stress Index, a numerical quantifier of the severity of heat relative to location. Estimates of heat-related mortality are developed for 40 large US cities, using climate models for 3 decades over the next hundred years and various emissions scenarios. Results: Over the past 30 years, the heat-mortality relationship has tended to converge across the United States, with areas in the warmer climates experiencing generally slight increases in mortality response, and areas in colder locations, which were historically more sensitive, experiencing more significant decreases. On a seasonal level, there is an asymmetry in heat-related mortality, with most cities exhibiting the largest increases before the summer solstice; in some cases in the southern United States, the hottest days in late summer exhibit a mortality decrease. Heat- mortality relationships are much more consistent across larger cities than smaller cities. Conclusion: The impacts of an anticipated climate change will vary among urban areas. The results show an increase in excessive heat event days and increased heat-attributable mortality across the cities with the most pronounced increase in the Southeast and Northeast. An evaluation to determine how various intervention activities would ameliorate this rise shows that public health responses and a nationalization of heat warning systems can dramatically lessen the increase posed by a climate change. O-29A1-2 Mortality Related to Temperature and Persistent Extreme Temperatures—A Study of Cause-Specific and Age Stratified Mortality Joacim Rocklo ¨v, 1,2 Kristie Ebi, 3 and Bertil Forsberg 1,2 Departments of 1 Public Health and Clinical Medicine and 2 Occupational and Environmental Medicine, Umeå University, Umeå, Sweden; and 3 Department of Global Ecology, Carnegie Institution for Science, Stanford, CA. Background/Aims: High and low ambient temperatures cause large numbers of deaths annually. Many studies show higher mortality rates during heat waves, with mortality increasing with heat wave duration. However, such effects have not been explicitly incorporated in models of temperature and mortality; while from a physiological basis, dehydration followed by cardiovascular stress is more likely to occur in such conditions. Methods: We established time-series Poisson regression models based on cause-specific mortality and age stratified mortality in Stockholm County, 1990 –2002, adjusting for time trends and potential confounders. We studied the effects estimates as functions of time within summers and winters. Results: Warmer temperatures increase all natural causes of death, while decreasing colder temperatures increases the risk of cardiovascular deaths. Moreover, persistent extreme heat exposure is associated substantially with additional deaths, and the risk of death increase significantly with longer heat exposure. Extreme exposure to heat is associated with higher death rates in people with lower age, compared to a rise in temperature. Furthermore, the impact of warm and cold temperatures decreases within the season, while the impact of persistent extreme heat exposure remains similar throughout the summer. Conclusion: This study is the first to show that additional impacts of persistence of extreme heat exposure are important to account for in models of mortality related to ambient temperatures to avoid negatively biased estimates of the associated effects. Moreover, it appears that the impacts of temperature depend on the size of the pool of fragile individuals, while persistent extreme heat exposure increases the influx of new individuals to the pool and, thus, remain large throughout the season. O-29A1-3 Diurnal Temperature Range and Cause-specific Cardiovascular Hospital Admission in Seoul, Korea—Time-series Analysis and Temperature- Matched Case-crossover Design Youn-Hee Lim and Ho Kim School of Public Health, Seoul, Republic of Korea. Background/Aims: Rapid temperature change within a day may increase blood pressure, clot, or cardiovascular work load. Hence, diurnal temperature range (DTR) may cause higher risk of cardiovascular disease (CVD) for population vulnerable to thermal stress. In this study, we hypothesized an increase of DTR which was associated with cardiovascular-related hospital admission in Seoul, South Korea. Methods: Data of daily hospital admission counts of 6 cause-specific CVD diseases in Seoul were collected from Korea National Health Insurance Corporation during 2003–2006. CVD diseases were arrhythmia, cardiac disease, cardiac failure, ischemic heart disease, myocardial infarction, and stroke. Results of time-series analysis and temperature- matched case-crossover design were compared. Associations in time-series analysis were evaluated using generalized linear model (glm) adjusted for confounding factors. In temperature-matched case-crossover design, Epidemiology • Volume 22, Number 1, January Supplement 2011 www.epidem.com | S13