www.ehjournal.net

Table 3
Recent Studies Identifying Vulnerable Subgroups of Mortality from High Ambient Temperature
Reference	Study location	Study design	Exposure	Causes of death	Result

Baccini 2008 [12]	15 European cities, April-September 1990-2000 (5-11 years depending on data availability for city)	Time-series	Maximum apparent temperature (threshold 29.4°C Mediterranean cities and 23.3°C north-continental cities)	Daily all-cause mortality	Respiratory diseases among 75+ years

Basu and Ostro 2008 [14]	9 California counties, May to September 1999-2003	Case-crossover	Mean daily apparent temperature	Cause-specific mortality; all-cause mortality by age, race/ethnicity, gender, education level	Cardiovascular, higher specifically for ischemic heart disease, myocardial infarction, and congestive heart failure, ≤ 1 year, ≤ 5 years, elderly, Black race, out of hospital death; no elevated risks for cerebrovascular, diabetes, respiratory; no difference by gender or high school graduation

Bell 2008 [15]	Sao Paulo, Brazil, Santiago, Chile and Mexico City, Mexico, 1998-2002	Case-crossover	Same day apparent temperature	Daily all-cause mortality	65+ years, women in Mexico City, but men in Santiago and Sao Paulo, less educated in Sao Paulo

Ishigami 2008 [24]	Budapest, London and Milan, 2003	Time-series	Mean daily temperature (lag0 and lag1), PM₁₀(TSP in Budapest), ozone	Daily all-cause mortality	Increased age, females 65+ years greater risk in London and Milan and non-elderly adults in Milan; mortality from external causes, respiratory and cardiovascular diseases

Stafoggia 2008 [30]	4 Italian cities, 1997-2004	Case-crossover	Apparent temperature 30°C compared to 20°C	Deaths in hospitals for those with 2+ days in hospital	Increased age, single general medicine compared to high and intensive care units, history of psychiatric disorders, cerebrovascular diseases, heart failure, stroke, chronic pulmonary diseases

Vaneckova 2008a [46]	Sydney, Australia, October to March 1993-2001	Time-series	Temporal Synoptic Index (TSI); ratio of highest 10% mortality days within air mass and % frequency of air mass occurrence	Daily all-cause mortality	65+ years, women

Yip 2008 [52]	Maricopa County, Arizona, June to September 2000-2005	Time-series	Heat index	Heat-related deaths	Young and old outdoors, but greater risk for elderly indoors

Hajat 2007 [25]	England and Wales, 1993-2003	Time-series	Heat (> 95^th%) and cold (< 5^th%) thresholds	All-cause mortality	Elderly, those in nursing care homes respiratory and external causes, women; not modified by deprivation in London

Medina-Ramon 2007 [21]	50 US cities in cold (November to March) and warm (May to September) seasons	Case-crossover	Binary variable as extreme temperature and continuous; ozone	All-cause and CVD mortality	Cities with milder summers, less air conditioning and higher population density

Diaz 2006 [35]	Madrid, January 1986-December 1997	Time-series	T(hwave) = Tmax-36.5C if Tmax>36.5C; 5^th% to 95^th% temperature, NO₂	AR = (RR-1)/RR for daily mortality	Circulatory causes, males 45-64 years

Stafoggia 2006 [16]	Bologna, Milan, Rome, Turin, 1997-2003	Case-crossover	30°C mean apparent temperature (lag01) relative to 20°C; odds ratio	All-cause mortality and previous hospitalization	Increased age and greater for women, widows and widowers, psychiatric disorders, depression, heart and circulatory disorders

Hajat 2005 [48]	Delhi, Sao Paulo, London, January 1991-December 1994	Time-series	Daily temperature (lag 0,1) greater than 20°C	Daily all-cause mortality	Respiratory deaths in Sao Paulo and London; children in Delhi

O'Neill, Zanobetti and Schwartz 2005 [37]	Chicago, Detroit, Minneapolis, Pittsburgh, 1988-1993 for Chicago and 1986-1993 for other cities	Time-series	Percent change daily mean temperature 29°C relative to 15°C (lag0), barometric pressure, day of the week, PM₁₀	Mortality, prevalence of air conditioner (AC)	Black race, lack of air conditioner

Gouveia 2003 [33]	Sao Paulo, Brazil, 1991-1994	Time-series	Daily mean temperature (lag01), SO₂, PM₁₀, CO, O₃, NO₂, day of the week, season, humidity	Daily all-cause mortality, excluding violent deaths, cardiovascular and respiratory mortality	Greatest for 65+years and < 15 years, also increased for15-64 years; elderly cardiovascular, respiratory for adults and elderly; no modification by socioeconomic status

O'Neill 2003 [38]	7 US cities, 1986-1993	Time-series	Mean daily apparent temperature (% change 29°C and -5°C), PM₁₀	Daily all-cause mortality, looking at effect modification by demographics & other variables	Black race, less educated, and outside hospital

Rainham and Smoyer-Tomic 2003 [42]	Toronto, May 1 to September 30, 1980-1996	Time-series	Humidex, CO, O₃, NO₂, SO₂	Daily all-cause mortality	Females

Curriero 2002 [39]	11 Eastern US cities, 1973-1994	Time-series	Daily mean temperature, dew point temperature; minimum mortality temperature (MMT) range: 65.2-90.3	Daily all-cause mortality, excluding accidents	Higher latitude, more poverty, less air conditioning or heating

Basu Environmental Health 2009 8:40 doi:10.1186/1476-069X-8-40

Table 3