. 2016 Oct 7;125(3):349–354. doi: 10.1289/EHP98

Table 5.

Cumulative percent change (95% confidence interval) in number of nonaccidental, circulatory, and respiratory deaths by 10 μg/m³ increase in particulate air pollution among studies.

Study	Age group	Air pollutant	Period	Percent change^a (95% confidence interval)
Study	Age group	Air pollutant	Period	Cumulative	Single lag
Non-accidental deaths
São Paulo^b (this study)	Elderly	PM₁₀	Lag 0 to 40	0.00 (–0.92, 0.92)	0.37 (0.20, 0.55)^c
Milan (Zanobetti et al. 2000)^b	All ages	TSP	Lag 0 to 45	6.70 (3.80, 9.60)	2.20 (1.40, 3.10)
Boston (Schwartz 2000a)^d	All ages	PM_2.5	60-day window	3.75 (3.20, 4.30)	2.10 (1.50, 4.30)
10 European Cities (Zanobetti et al. 2002)^b	All ages	PM₁₀	Lag 0 to 40	1.57 (0.26, 2.88)	0.70 (0.43, 0.97)
Dublin (Goodman et al. 2004)^b	All ages	BS	Lag 0 to 40	1.10 (0.80, 1.30)	0.40 (0.30, 0.60)
Circulatory deaths
São Paulo^b (this study)	Elderly	PM₁₀	Lag 0 to 30	–1.11 (–2.21, 0.01)	0.01 (–0.27, 0.29)^e
10 European Cities (Zanobetti et al. 2003)^e	All ages	PM₁₀	Lag 0 to 40	1.72 (1.20, 2.25)	0.69 (0.31, 1.08)
Dublin (Goodman et al. 2004)^b	All ages	BS	Lag 0 to 40	1.10 (0.70, 1.50)	0.40 (0.20, 0.70)
Respiratory deaths
São Paulo^b (this study)	Elderly	PM₁₀	Lag 0 to 30	2.81 (0.99, 4.66)	1.33 (0.83, 1.83)^e
10 European Cities (Zanobetti et al. 2003)^e	All ages	PM₁₀	Lag 0 to 40	2.62 (0.19, 5.11)	0.74 (–0.17, 1.66)
Dublin (Goodman et al. 2004)^b	All ages	BS	Lag 0 to 40	3.60 (3.00, 4.30)	0.20 (0.00, 0.50)
Notes: BS, black smoke; PM_2.5, particulate matter smaller than 2.5 μm; PM₁₀, particulate matter smaller than 10 μm; TSP, total suspended particulate. ^aInterquartile range increase in Zanobetti et al. (2000) and 10-μg/m³ increase in other studies. ^bDistributed lag model constrained with polynomial structure. ^cLag 0. ^dSTL (standard template library) algorithms with LOESS (locally weighted smoothing). ^eUnconstrained distributed lag model.