Skip to main content

Table 1 Summary of effect estimates (excess risk per 10 μg/m 3 ) from cohort studies on particulate matter (PM 10 or PM 2.5 ) and mortality from all causes and cardiovascular diseases

From: Long-term air pollution exposure and cardio- respiratory mortality: a review

Study Study population Follow-up period Pollutant Conca(μg/m3) Spatial scaleb % change in risk (95% CI) in mortality associated with a 10 μg/m3increase PM References
       All cause Cardiovascular c  
Harvard six cities 8111 adults in six US cities 1976 - 1989 PM2.5 18 (11–30) City 13(4, 23) 18 (6, 32) [15]
Harvard six cities 8096 adults in six US cities 1979 -1998 PM2.5 15 (10–22) City 16 (7, 26) 28 (13,44) [14]
Harvard six cities 8096 adults in six US cities 1974 - 2009 PM2.5 16 (11–24) City 14 (7, 22) 26 (14, 40) [16]
American Cancer Society (ACS) study 552, 800 adults from 51 US cities 1982 - 1989 PM2.5 18 (9–34) City 26 (8, 47) NA [17]
ACS study 500,000 adults from 51 US cities 1982 -1998 PM2.5 18 (4) City 6 (2, 11) 9 (3, 16)c [18]
ACS sub-cohort study 22,905 subjects in Los Angeles area 1982 - 2000 PM2.5 (~9 – 27) Zip code (Int) 17 (5, 30) 26 (1, 60)c [19]
German cohort 4752 women in Ruhr area 1985 – 2003 PM10 44 (35–53) Address (near) 12 (−9, 37) 52 (8, 114) [20]
German cohort 4752 women in Ruhr and surrounding area 1985 - 2008 PM10 44 (35–53) Address (near) 22 (6, 41) 61 (26, 104) [21]
Women’s Health Initiative Observational Study 65,893 postmenopausal women from 36 US metropolitan areas 1994-1998 PM2.5 14 (3–28) Zip code (near) NA 76 (25,147) [22]
Netherlands Cohort Study 120, 852 subjects from Netherlands 1987 -1996 PM2.5 28 (23–37) Address (LUR) 6 (−3, 16) 4 (−10, 21) [23]
Nurses’ Health Study 66,250 women from the US north eastern metropolitan areas 1992-2002 PM10 22 (4) Address (LUR) 11 (1,23) 35 (3, 77) [24]
Nurses’ Health Study 66,250 women from the US north eastern metropolitan areas 1992-2002 PM2.5 14 (6–28) Address (LUR) 26 (2, 54) NA [25]
Medicare national cohort 13.2 million elderly Medicare recipients across the USA 2000 - 2005 PM2.5 13 (4) Zip code (Mean) 4 (3, 6)d   [26]
California teachers study 45,000 female teachers 2002 -2007 PM2.5 18 (7–39) Address (near) 6 (−4, 16) 19 (5, 36)c [27]
Swiss national cohort National census data linked with mortality 2000 - 2005 PM10 19 (>40)e Address (Disp) NA −1 (−3, 0) [28]
Health professionals follow-up study 17,545 highly educated men in the midwestern and northeastern US 1989 – 2003 PM2.5 18 (3) Address (LUR) −14 (−28,2) 3 (−17, 26) [29]
Vancouver cohort 452,735 Vancouver residents 45–85 yr 1999 – 2002 PM2.5 4 (0 – 10) Address (LUR) NA 7 (-14, 32) [30]
China nat. hypertension survey 70,497 men and women 1991 - 2000 TSP 289 (113–499) City 0.3 (0, 1) 1 (0, 2) [31]
US trucking industry cohort 53,814 men in the US trucking industry 1985 -2000 PM2.5 14 (4) Address (near) 10 (3, 18) 5 (−7, 19) [32]
Chinese retrospective cohort study 9,941 adults from five districts of Shenyang city 1998 -2009 PM10 154 (78–274)f District (mean) 53 (50, 56) 55 (51, 60) [33]
Canadian national cohort 2.1 million nonimmigrant Canadians . > 25 yr 1991 - 2001 PM2.5 9 (2 – 19) Enumeration area, N = 45710 (satellite) 10 (5, 15) 15 (7, 24) [34]
New Zealand Census mortality study 1.06 million adults in urban areas from 1996 census 1996 -1999 PM10 8 (0 – 19) Census tract (Disp) 7 (3, 10) 6 (1, 11) [35]
California teachers study 101,784 female teachers 1997- 2005 PM2.5 16 (3–28) Address (Inter) 1 (−5, 9) 7 (−5, 19) [36]
Nippon data cohort 7,250 adults > 30 yr throughout Japan 1980 - 2004 PM10 <27 - > 43 District (near) −2 (−8, 4) −10 (−19, 0) [37]
Rome longitudinal study 1,265,058 adults from Rome 2001 - 2010 PM2.5 23 (7 – 32) Address (DISP, 1 km grid) 4 (3, 5) 6 (4, 8) [38]
  1. a Mean with minimum – maximum in parentheses (μg/m3). One number in parentheses is standard deviation.
  2. b Spatial scale of exposure assignment, in parentheses exposure assignment method. City = average of monitors within the city; Near = nearest monitor concentration; LUR = land use regression; Disp = dispersion modeling; Inter = interpolation.
  3. c Cardio-pulmonary mortality reported if cardiovascular mortality not available.
  4. d Combining the estimates from the three regions of the USA.
  5. e Median and 90th percentile reported.
  6. f Very high pollution levels that changed significantly during follow-up changing the ranking of the five districts.
  7. Studies adjusted for individual smoking except references [26, 28, 30, 34, 38, 56].