Table 2.

Pooled associations between PM_2.5 exposure (per 10 μg/m³ increment) during pregnancy and preterm birth risks in different subgroups

Subgroups	No. of studies	Heterogeneity test		Summary OR (95 % CI)	Hypothesis test		I2 (%)	Egger’s test
		Q	P		Z	P		t	P
Exposure during the entire pregnancy	13	80.51	<0.001	1.13* (1.03–1.24)	2.59	0.010	91.4	2.20	0.051
Specific trimester
First trimester exposure	10	89.14	<0.001	1.08 (0.92–1.26)	0.96	0.334	91.3	0.68	0.517
Second trimester exposure	5	138.69	<0.001	1.09 (0.82–1.44)	0.60	0.548	98.7	0.311	0.776
Third trimester exposure	9	44.83	<0.001	1.08** (0.99–1.17)	1.70	0.089	92.1	1.58	0.157
First month of gestation	3	22.03	<0.001	1.10 (0.92–1.30)	1.03	0.301	91.0	0.58	0.666
Within one month before birth	6	51.49	<0.001	1.01 (0.86–1.19)	0.09	0.926	96.8	0.03	0.980
Exposure assessment methoda
Individual exposure	3	4.94	0.085	1.11 (0.89–1.37)	0.93	0.352	61.3	1.74	0.332
Semi-individual exposure	9	55.86	<0.001	1.14 (0.97–1.35)	1.56	0.119	93.0	0.35	0.737
Regional level	4	46.19	<0.001	1.07 (0.94–1.23)	1.00	0.319	93.8	0.11	0.921
Study designa
Retrospective studies	9	70.98	<0.001	1.10* (1.01–1.21)	2.12	0.034	93.3	2.31	0.055
Prospective studies	4	4.64	0.201	1.42* (1.08–1.85)	2.52	0.012	39.5	0.10	0.927
Study settinga
USA	8	50.49	<0.001	1.16* (1.05–1.29)	2.73	0.006	90.6	1.80	0.121
Others	5	7.90	0.095	0.98 (0.95–1.01)	1.11	0.268	0.1	1.62	0.205

^a: All of these subgroup analyses were conducted for the studies that assessed the association between PM_2.5 exposure during the entire pregnancy and preterm birth. All of these estimates were ORs for each 10 μg/m³ increment of PM_2.5 exposure during the entire pregnancy

*: p < 0.05

**: 0.05 < p < 0.10