Health Effects of Aerosols - Aerosols: Measurement, Dosimetry, and Health Effects - page 614

TABLE 23.17
Selected Time Series Studies That Evaluate the Relationship between Changes in Speciic
Mass Components of PM Aerosol and Daily Mortality

Study
Population

Pollutants and
Components

Years Methods Results References

Studies that report predominant effects for ine component of mass fraction
Spokane, WA

1989-1996 Days with dust

storms as surrogate
for increased
coarse PM, PM 10

PR a with
indicator for
days with
dust storms

RR for days after dust storms
1.01 (0.87-1.17)

[126]

Wasatch
Front, UT

1985-1995 PM 10 (PM 2.5

∼70%-90% of
PM 10 , PM 1.0 ),
windblown dust
episodes surrogate
for PM 10-2.5

PR, GAM Stagnant air episodes

characterized by high
concentrations of primary and
secondary combustion source

[316]

PM more associated with
mortality than windblown
dust episodes characterized by
PM with larger contribution
from crustal elements

Harvard Six
Cities

1976-1987 PM 10 , b coarse PM,

PM 2.5 , H + , SO 4 2−

PR, GAM Combined % increase mortality

over all cities/10 μg/m 3
increase on same day PM
metric

[128]

PM 2.5 1.3% (0.9-1.7)
Coarse PM 0.4% (20.2, 0.9)
(Results for two cities for
course PM show comparable
effects with PM 2.5 )

Harvard Six
Cities

1979-1988 Sources based on 15

elements,
converted to ive
“factors”

PR, GAM Percent increase in

mortality/10 μg/m 3 increase in
mass concentration from
speciic source

[121]

Mobile sources (ine PM)
3.4% (1.7-5.2)
Crustal sources 22.3%
(25.8, 1.2)

Meta-
analysis
based on 19
studies from
United
States

Studies
span
1973-
1990

PM 10 , TSP, SO 2 ,
CO, NO 2 , O 3 ,
PM 2.5 /PM 10 ratio)
for 14 U.S. cities

Random
effects

Percent change in
mortality/10 μg/m 3 increase in
PM 10 (ratio used directly)

[229]

PM 10 0.67 (0.46-0.88)

Bayes
summaries

PM 2.5 /PM 10 ratio as grouped
variable: 0.31% (ratio
<0.57), 0.68 (ratio
0.57-0.64), 0.81 (ratio
>0.65) (95% CI excludes 0
for the highest two groups)