Humidity does not Confound Temperature-Mortality Relationship in Japan

DOI Web Site 12 References
  • HONDA Yasushi
    Institute of Health and Sport Sciences, University of Tsukuba
  • ONO Masaji
    Division of Health Sciences, National Institute for Environmental Studies
  • UCHIYAMA Iwao
    Department of Occupational Health, National Institute of Public Health

Bibliographic Information

Other Title
  • わが国において,湿度は気温—死亡の関係に影響を与えない

Search this article

Description

Several studies have reported a “V”-shaped relationship between short-term temperature and mortality rates, characterized by mortality rates that are higher when temperature is extremely low or high than when moderate. To quantify the confounding effect of humidity on the V-shaped short-term temperature-mortality relationship, we studied Japanese prefecture-specific data from 1972 to 1990. We classified the prefectures into 5 areas according to the 19-year mean temperature. For each area, we calculated all-cause mortality rates of 65+ years old subjects by gender, year, daily maximum temperature and daily relative humidity. We then calculated two types of Mantel-Haenszel rate ratios: one controlled for year only (MHRRy), and the other controlled for year and daily relative humidity (MHRRyh). The ratio of MHRRy to MHRRyh would be unity, if humidity were not a confounder. This ratio (MHRRy/MHRRyh) by daily maximum temperature, area and gender ranged from 0.989 to 1.037, and 89% of the ratios was between 0.990 and 1.010, i.e., the discrepancy was less than 1%. This suggests that we need not control for humidity in evaluating the short-term temperature effect on all-cause mortality, at least for 65+ years old age group in Japan.

Journal

References(12)*help

See more

Keywords

Details 詳細情報について

Report a problem

Back to top