Oil and gas development and spontaneous preterm birth risk in the San Joaquin Valley, CA, USA

Supplementary Materials

 

David J.X. Gonzalez, Allison R. Sherris, Wei Yang, David K. Stevenson, Amy Padula, Michael Baiocchi, Marshall Burke, Mark R. Cullen, and Gary M. Shaw

 

 

Figure S1. The locations of well sites in preproduction by month, 1997 to 2011.

 

 

Screen%20Shot%202018-05-22%20at%205

Figure S2. Flow diagram of inclusion and exclusion criteria for births.

 

 

Figure S3. Count of wells spudded and completed by month, California, USA.

 

 

Table S1. Characteristics of the study population stratified by exposure tertile, San Joaquin Valley, California. For this table, exposure was assessed cumulatively throughout the pregnancy.

 

Exposure tertile

0 (none)

1

2

3

n

181,354

14,674

14,673

14,673

Gestational age (%)

 

 

 

 

   20-27 weeks

1.0

1.0

1.0

1.0

   28-31 weeks

1.4

1.3

1.4

1.5

   32-36 weeks

10.0

9.5

10.0

10.3

   37-41 weeks (term)

87.6

88.2

87.6

87.2

Age, years (mean ± SD)

26.1 ± 5.97

26.2 ± 6.02

25.7 ± 6.09

25.5 ± 5.89

Race/ethnicity (%)

   Asian

7.4

8.8

6.5

3.4

   Black

4.3

6.6

6.9

6.2

   Hispanic

56.4

55.2

64.5

55.5

   White, non-Hispanic

29.8

27.5

19.8

32.4

   Other

1.5

1.4

1.5

1.8

   Missing

0.5

0.5

0.6

0.8

Education (%)

   < 12 years

32.6

31.0

37.6

32.3

   12 years

31.3

33.2

32.5

35.5

   > 12 years

34.9

33.4

26.7

27.8

   Missing

1.3

2.4

3.3

4.4

Parity (%)

   1

34.9

35.4

34.9

35.4

   2 or more

65.0

64.5

65.1

64.5

   Missing

0.1

0.1

0.1

0.1

Infant sex (% female)

48.6

49.3

48.9

49.2


 

Table S2. Summary statistics for inverse distance-squared weighted (IDW) indices, organized by trimester of exposure. The values are for each exposure tertile. Exposed births were divided into tertiles; tertile “0” comprises unexposed births. An index of 1 is equivalent to having one well located 1 km away from the maternal residence, or ten wells located 10 km away.

 

Trimester

Tertile

Mean ± SD

Range

1

0 (unexposed)

0

0

1

0.02 ± 0.01

0.01, 0.03

2

0.07 ± 0.03

0.03, 0.14

3

 

0.86 ± 1.54

0.14, 58.56

2

0

0

0

1

0.02 ± 0.01

0.01, 0.03

2

0.07 ± 0.03

0.03, 0.15

3

 

0.93 ± 1.72

0.15, 84.3

3

0

0

0

1

0.02 ± 0.01

0.01, 0.03

2

0.07 ± 0.03

0.03, 0.14

3

0.91 ± 1.66

0.15, 76.68

 

 

Table S3. Unadjusted odds ratios (95% confidence intervals) for the association between high exposure to drilling and preterm birth. This analysis compares the most exposed births (tertile 3) with unexposed births.

 

 

OR (95% CI)

Trimester

20-27 weeks

28-31 weeks

32-36 weeks

1

1.19 (0.99, 1.41)

1.37 (1.19, 1.58)

1.10 (1.03, 1.17)

2

1.12 (0.93, 1.34)

1.27 (1.09, 1.47)

1.11 (1.05, 1.18)

3

1.11 (1.05, 1.18)

 

 

Table S4. Odds ratios (95% confidence intervals) for the association between high exposure to drilling and preterm birth, adjusted for observed maternal covariates as well as birth year. This analysis compares the most exposed births (tertile 3) with unexposed births.

 

 

OR (95% CI)

Trimester

20-27 weeks

28-31 weeks

32-36 weeks

1

1.12 (0.93, 1.34)

1.31 (1.13, 1.52)

0.99 (0.93, 1.06)

2

1.06 (0.88, 1.27)

1.20 (1.03, 1.40)

1.00 (0.94, 1.07)

3

1.00 (0.94, 1.07)

 

 

Table S5. Results adjusted for maternal covariates for births with all tertiles of exposure compared with unexposed births, stratified by gestational age category. Results are presented as odds ratio (95% confidence interval). Exposure in the third trimester was assessed for the last 30 days prior for delivery.

 

 

 

OR (95% CI)

Gestational age

Exposure Tertile

Trimester 1

Trimester 2

Trimester 3

20-27 weeks

0 (unexposed)

1.0

1.0

1

1.15 (0.95, 1.36)

0.95 (0.78, 1.14)

2

1.02 (0.84, 1.23)

1.00 (0.82, 1.20)

3

1.21 (1.00, 1.44)

1.14 (0.95, 1.37)

28-31 weeks

0

1.0

1.0

1

1.07 (0.91, 1.25)

1.00 (0.85, 1.18)

2

1.10 (0.94, 1.29)

1.13 (0.97., 1.32)

3

1.35 (1.16, 1.56)

1.24 (1.06, 1.44)

32-36 weeks

0

1.0

1.0

1.0

1

1.03 (0.97, 1.10)

0.98 (0.92, 1.05)

1.02 (0.96, 1.08)

2

1.04 (0.97, 1.10)

1.04 (0.97, 1.10)

1.09 (1.03, 1.16)

3

1.10 (1.03, 1.17)

1.12 (1.05, 1.19)

1.12 (1.05, 1.18)

 

 

Table S6. Results for primary analysis with exposure assessed by gestational month, stratified by gestational age category. Results are presented for as odds ratio (95% confidence interval) for each month, from B1 (the month before conception) to P1 (the first month of gestation) to P2 (second month of gestation) up to P7.

 

 

OR (95% CI)

Month

20-27 weeks

28-31 weeks

32-36 weeks

B1

1.21 (0.99, 1.45)

1.30 (1.10, 1.52)

1.10 (1.03, 1.18)

P1

1.24 (1.02, 1.50)

1.30 (1.11, 1.53)

1.10 (1.03, 1.18)

P2

1.26 (1.04, 1.52)

1.32 (1.12, 1.55)

1.11 (1.04, 1.18)

P3

1.26 (1.03, 1.51)

1.35 (1.15, 1.57)

1.12 (1.05, 1.20)

P4

1.23 (1.01, 1.48)

1.30 (1.10, 1.53)

1.14 (1.07, 1.22)

P5

1.18 (0.96, 1.42)

1.24 (1.05, 1.46)

1.13 (1.05, 1.20)

P6

 

1.25 (1.06, 1.46)

1.11 (1.04, 1.19)

P7

 

 

1.11 (1.04, 1.19)

 

 

Table S7. Results from sensitivity analysis with alternative exposure assessment assumptions. For these analyses, we assessed exposure using different parameters. We assessed exposure to wells within (a) 3, (b) 5, and (c) 15 km of the maternal residence using inverse distance-squared weighting, and to wells within 10 km with (d) inverse distance weighting and (e) inverse distance-square root weighting. Results are stratified by gestational age category and reported as odds ratios (95% confidence intervals). For assessment (a), exposure is assigned as a binary indicator (exposed/unexposed) rather than divided into tertiles, because the range of IDW indices were comparable to that of high tertile births in other assessments. All analyses are adjusted for observed maternal covariates.

 

 

 

OR (95% CI)

Exposure Parameters

Trimester

20-27 weeks

28-31 weeks

32-36 weeks

 

1

1.46 (1.10, 1.88)

1.05 (0.79, 1.36)

1.05 (0.95, 1.17)

(a) 3 km radius, (1/d2)

2

1.29 (0.96, 1.69)

1.15 (0.88, 1.48)

1.05 (0.94, 1.16)

 

3

 

1.20 (1.01, 1.41)

 

1

1.44 (1.09, 1.86)

1.34 (1.05, 1.68)

1.12 (1.02, 1.24)

(b) 5 km radius, (1/d2)

2

1.13 (0.83, 1.50)

1.32 (1.03, 1.66)

1.15 (1.04, 1.26)

 

3

 

1.15 (1.05, 1.26)

 

1

1.18 (1.01, 1.36)

1.27 (1.12, 1.43)

1.09 (1.04, 1.15)

(c) 15 km radius, (1/d2)

2

1.06 (0.91, 1.23)

1.21 (1.07, 1.37)

1.08 (1.03, 1.14)

 

3

 

1.12 (1.07, 1.18)

 

1

1.26 (1.05, 1.50)

1.40 (1.20, 1.61)

1.13 (1.06, 1.20)

(d) 10 km radius, (1/d)

2

1.15 (0.96, 1.38)

1.30 (1.11, 1.50)

1.10 (1.03, 1.17)

 

3

 

1.14 (1.07, 1.21)

 

1

1.24 (1.04, 1.48)

1.39 (1.20, 1.61)

1.13 (1.06, 1.20)

(e) 10 km radius, (1/d½)

2

1.17 (0.97, 1.40)

1.26 (1.08, 1.46)

1.12 (1.05, 1.19)

 

3

1.14 (1.07, 1.21)

 

 

Table S8. Results from analyses stratified by (a) maternal race/ethnicity, (b) maternal education, and (c) birth year. We fit unadjusted logistic regression models for each stratum, and each trimester and gestational age category within the stratum. For the stratification on maternal race/ethnicity, we omitted the strata for Asian and other due to insufficient sample size.

 

 

 

 

OR (95% CI)

 

Stratum

Trimester

20-27 weeks

28-31 weeks

32-36 weeks

 

 

1

1.22 (0.96, 1.53)

1.41 (1.16, 1.71)

1.13 (1.04, 1.23)

(a) Maternal

race/ethnicity

Hispanic

2

1.18 (0.93, 1.49)

1.27 (1.03, 1.54)

1.15 (1.06, 1.25)

 

3

 

1.13 (1.04, 1.23)

Non-Hispanic

Black

 

1

1.40 (0.82, 2.26)

1.43 (0.91, 2.16)

1.20 (0.96, 1.50)

2

1.20 (0.67, 1.98)

1.21 (0.75, 1.87)

1.26 (1.01, 1.57)

3

 

1.17 (0.94, 1.44)

Non-Hispanic

White

 

1

1.21 (0.84, 1.68)

1.39 (1.05, 1.80)

1.03 (0.93, 1.15)

2

1.12 (0.78, 1.57)

1.36 (1.03, 1.77)

1.04 (0.93, 1.15)

 

3

 

1.08 (0.98, 1.20)

 

 

1

1.31 (0.96, 1.73)

1.59 (1.26, 1.98)

1.25 (1.13, 1.39)

(b) Maternal

education

< 12 years

2

1.26 (0.92, 1.67)

1.41 (1.11, 1.77)

1.26 (1.14, 1.40)

 

3

 

1.25 (1.13, 1.37)

 

1

1.11 (0.82, 1.46)

1.19 (0.92, 1.50)

1.07 (0.96, 1.19)

12 years

2

1.04 (0.76, 1.39)

1.17 (0.90, 1.48)

1.13 (1.02, 1.25)

 

3

 

1.08 (0.98, 1.19)

> 12 years

 

1

1.15 (0.78, 1.62)

1.22 (0.88, 1.65)

0.95 (0.84, 1.07)

2

1.07 (0.73, 1.52)

1.08 (0.76, 1.49)

0.92 (0.82, 1.04)

 

3

 

0.99 (0.88, 1.11)

 

 

1

1.18 (0.90, 1.50)

1.23 (0.99, 1.50)

1.12 (1.02, 1.23)

(c) Birth year

1998-2006

2

1.21 (0.93, 1.55)

1.20 (0.96, 1.48)

1.14 (1.04, 1.25)

 

3

 

1.09 (0.99, 1.19)

 

1

1.07 (0.83, 1.35)

1.42 (1.16, 1.72)

0.92 (0.95, 1.00)

2007-2011

2

0.93 (0.71, 1.19)

1.25 (1.01, 1.52)

0.92 (0.95, 1.00)

 

 

3

0.96 (0.88, 1.03)

 

 

Table S9. Results from sensitivity analysis, lagged exposure

 

[not available]

 

 

Table S10. Association between drilling exposure and preterm birth (delivery at < 37 weeks) comparing births with high exposure to unexposed births, stratified by exposure to traffic-related copollutants. Births with low traffic exposure were in the lowest three quartiles for exposures to all four traffic-related copollutants. Births with high traffic exposure were in the highest quartile for all four copollutants.

 

 

OR (95% CI)

Trimester

Low Traffic

(n = 40,427)

High Traffic

(n = 7,039)

All
(n = 47,466)

1

1.15 (0.91, 1.43)

1.07 (0.87, 1.31)

1.20 (1.04, 1.39)

2

1.25 (1.00, 1.55)

1.03 (0.84, 1.26)

1.22 (1.07, 1.41)

3

0.98 (0.73, 1.27)

1.07 (0.86, 1.32)

1.13 (0.97, 1.32)

 

 

Table S11. Results from secondary analysis, effect of drilling on air quality

 

[not available]