Interdepartmental Centre of Bioclimatology, University of Florence, Firenze, Italy.
Int J Biometeorol. 2012 Sep;56(5):801-10. doi: 10.1007/s00484-011-0481-y. Epub 2011 Aug 7.
High ambient temperatures have been associated with increased mortality across the world. Several studies suggest that timely preventive measures may reduce heat-related excess mortality. The main aim of this study was to detect the temporal modification of heat-related mortality, in older adults (aged 65-74) and in elderly ≥75 years old, in the Florentine area by comparing previous (1999-2002) and subsequent (2004-2007) periods to the summer of 2003, when a regional Heat-Health Warning System (HHWS) was set up. Mortality data from 1999 to 2007 (May-September) were provided by the Mortality Registry of the Tuscany Region (n = 21,092). Weather data were used to assess daily apparent temperatures (AT). Case-crossover time-stratified designs and constrained segmented distributed lag models were applied. No significant heat-related mortality odds ratio (OR) variations were observed among the sub-periods. Nevertheless, a general OR decrease dating from 1999-2002 (OR 1.23; lack of HHWS) to 2004-2005 (OR 1.21; experimental HHWS running only for Florence) and to 2006-2007 (OR 1.12; official HHWS extended to the whole Florentine area) was observed when the maximum AT was considered. This modification was only evident in subjects ≥75 years old. The heat effect was higher and sustained for more days (until lag 9) during the period 1999-2002 than 2004-2007. The decrease of the excessive heat effect on mortality between periods with the absence and existence of a HHWS is also probably due to the mitigation of preventive measures and the implementation of a HHWS with specific interventions for safeguarding the health of the "frail elderly".
高温环境与全球范围内的死亡率增加有关。有几项研究表明，及时采取预防措施可能会降低与热有关的超额死亡率。本研究的主要目的是通过比较之前（1999-2002 年）和之后（2004-2007 年）两个时期与 2003 年夏季的数据，检测佛罗伦萨地区老年人（65-74 岁）和老年人（≥75 岁）与热有关的死亡率的时间变化，当时建立了区域热健康预警系统（HHWS）。托斯卡纳地区死亡率登记处提供了 1999 年至 2007 年（5 月至 9 月）的死亡率数据（n=21092）。使用天气数据评估每日明显温度（AT）。应用病例交叉时间分层设计和约束分段分布滞后模型。在各子时段内，未观察到与热有关的死亡率比值比（OR）的显著变化。然而，从 1999-2002 年（无 HHWS，OR1.23）到 2004-2005 年（仅对佛罗伦萨运行的实验 HHWS，OR1.21）再到 2006-2007 年（正式 HHWS 扩展到整个佛罗伦萨地区，OR1.12），当考虑最高 AT 时，观察到一般 OR 下降。这种变化仅在≥75 岁的人群中明显。与 2004-2007 年相比，1999-2002 年期间热效应更高且持续时间更长（直至滞后 9 天）。在没有和存在 HHWS 的两个时期之间，对死亡率的过高热效应的降低可能也是由于预防措施的缓解和实施 HHWS 以及为保护“脆弱老年人”的健康而采取的具体干预措施所致。
