老年人口自殺與熱島效應之空間關聯分析：以2000年至2008年台北都會區生態調查研究為例

Spatial correlation analysis of elderly suicides and Urban Heat Island Effects: an ecological study in Taipei, 2000-2008

目標：本研究旨在應用地理資訊系統與統計模型，以探討台北都會區2000至2008年間熱島效應強度與65歲以上老年人口自殺死亡率之關聯。方法：本研究利用氣象站監測資料以及克利金空間推估以量化熱島效應；其次，針對熱島效應以及老年人口之自殺死亡率進行空間以及時間趨勢分析，並進一步探討自殺死亡率在空間上的群聚情形；最後，於校正大氣溫度、雨量、平均每戶綜合所得、教育程度、65歲以上性別比以及離婚率等因子，並且考慮時間、空間自相關之情況下，利用廣義加乘混合模型，以釐清熱島強度與老年人口自殺死亡率之關聯性。結果：在考慮空間、時間自相關、隨機效應、性別比、人口結構以及各項重要之社會經濟因子等風險因子後，熱島強度對於老年人口自殺死亡率之相對風險達統計上之顯著水準（相對風險：1.0549；p < 0.01）。結論：本研究發現，熱島效應之強度與老年人口自殺死亡具有統計上之顯著關聯，熱島效應強度增加1℃，可能會使得老年人口之自殺死亡率相對上升5.49%。

Objectives: The aim of this study was to investigate the association between Urban Heat Island Effects (UHIE) and elderly suicides in metropolitan Taipei from 2000 to 2008 using Geographic Information System (GIS) and statistical analysis. Methods: The UHIE intensity in each township was estimated based on onsite air temperature observations using Kriging interpolation. Spatial/temporal trends in UHIE and elderly suicides were examined, and the spatial clustering effect of elderly suicides was assessed. A Generalized Additive Mixed Model (GAMM) was then utilized to assess the association between UHIE and elderly suicides after adjusting for confounders including air temperature, precipitation, income tax, education level, sex ratio of the elderly population, and marital status. Results: Based on the GAMM analysis, UHIE reached the level of statistical significance (Relative Risk: 1.0549; p < 0.01) while accounting for the effects of the other risk factors in the model. Conclusions: The intensity of UHIE was statistically associated with elderly suicides in Taipei.