Skip to main content
PMC home page
JAMA Network logoLink to JAMA Network
. 2022 Apr 4;176(6):585–592. doi: 10.1001/jamapediatrics.2022.0306

Association Between Residential Proximity to Hydraulic Fracturing Sites and Adverse Birth Outcomes

Zoe F Cairncross 1, Isabelle Couloigner 2,3, M Cathryn Ryan 4, Carly McMorris 5, Lucija Muehlenbachs 6, Nickie Nikolaou 7, Ron Chik-Kwong Wong 8, Selwynne M Hawkins 7, Stefania Bertazzon 2, Jason Cabaj 9, Amy Metcalfe 1,9,10,✉
PMCID: PMC8981068  PMID: 35377398

This cohort study examines the association between residential proximity to hydraulic fracturing sites and adverse birth outcomes.

Key Points

Question

Is residential proximity to hydraulic fracturing sites associated with adverse birth outcomes?

Findings

In this population-based cohort study including all reproductive-aged individuals who had a pregnancy in rural Alberta, Canada, from 2013 to 2018, those individuals living within 10 km of 100 or more hydraulically fractured wells during 1 year preconception or pregnancy had a significantly increased risk of spontaneous preterm birth and small for gestational age birth.

Meaning

In this study, living in proximity to a high density of hydraulic fracturing sites was associated with adverse birth outcomes.

Abstract

Importance

The association between hydraulic fracturing and human development is not well understood. Several studies have identified significant associations between unconventional natural gas development and adverse birth outcomes; however, geology and legislation vary between regions.

Objective

To examine the overall association between residential proximity to hydraulic fracturing sites and adverse birth outcomes, and investigate whether well density influenced this association.

Design, Setting, and Participants

This population-based retrospective cohort study of pregnant individuals in rural Alberta, Canada, took place from 2013 to 2018. Participants included reproductive-aged individuals (18-50 years) who had a pregnancy from January 1, 2013, to December 31, 2018, and lived in rural areas. Individuals were excluded if they lived in an urban setting, were outside of the age range, or were missing data on infant sex, postal code, or area-level socioeconomic status.

Exposures

Oil and gas wells that underwent hydraulic fracturing between 2013 to 2018 were identified through the Alberta Energy Regulator (n = 4871). Individuals were considered exposed if their postal delivery point was located within 10 km of 1 or more wells that was hydraulically fractured during 1 year preconception or during pregnancy.

Main Outcomes and Measures

Outcomes investigated were spontaneous and indicated preterm birth, small for gestational age, major congenital anomalies, and severe neonatal morbidity or mortality.

Results

After exclusions, the sample included 26 193 individuals with 34 873 unique pregnancies, and a mean (SD) parental age of 28.2 (5.2) years. Small for gestational age and major congenital anomalies were significantly higher for individuals who lived within 10 km of at least 1 hydraulically fractured well after adjusting for parental age at delivery, multiple births, fetal sex, obstetric comorbidities, and area-level socioeconomic status. Risk of spontaneous preterm birth and small for gestational age were significantly increased in those with 100 or more wells within 10 km.

Conclusions and Relevance

Results suggest that individuals who were exposed to hydraulic fracturing within pregnancy may be at higher risk of several adverse birth outcomes. These results may be relevant to health policy regarding legislation of unconventional oil and gas development in Canada and internationally.

Introduction

Hydraulic fracturing, commonly referred to as fracking, is a component of a cost-efficient technology used to drill horizontal wells that extract gas from unconventional (eg, shale) formations.1,2 Although the relationship between hydraulic fracturing and human development is not well understood, members of the public and environmental groups have expressed concern over potential adverse health outcomes.3 Presently, the World Health Organization classifies the association between fracking and human health as “mostly unknown.”4 Legislation and regulations attempt to mitigate the health and environmental hazards of hydraulic fracturing; nonetheless, exposure to potential exposures cannot be entirely eliminated through the regulatory regime alone.5 Hydraulic fracturing may be associated with individual health through several processes: drilling, extraction, and the use of toxic chemicals may lead to groundwater contamination, and heavy truck traffic may create additional air pollution.6,7

Few studies have examined the association between residential exposure to hydraulic fracturing and human reproductive outcomes. In a 2019 systematic review, significant associations were identified between unconventional natural gas development and preterm birth,2,8,9 infant mortality,10 low birth weight,8,11,12 and small for gestational age (SGA).2,8,9 Most of these studies were conducted in the US; hence their findings might not be generalizable to other countries.

Our objectives were 3-fold: (1) to examine the association between residential proximity to hydraulic fracturing sites and adverse birth outcomes in Alberta, Canada, (2) to investigate the influence of well distance and well density on this association, and (3) to conduct sensitivity analyses measuring the influence of timing of exposure on this association.

Methods

Study Setting and Sample

Alberta is Canada’s fourth largest province with a population of 4.4 million and approximately 50 000 births annually.13 More than 450 000 energy wells have been drilled in Alberta, of which more than 150 000 are presently active.14,15 Wells are considered active if they have had production of oil or gas in the last 12 months. Current legislation requires that sites be at least 100 m away from residential locations.16

We conducted a population-based cohort study including all reproductive-aged individuals (18-50 years) who were living in a rural area and had a pregnancy between January 1, 2013, and December 31, 2018, in Alberta, Canada. Individuals were excluded if they did not live in a rural area, and if they were missing data on infant sex, postal code, or area-level socioeconomic status. No pregnant individuals living in urban areas resided within 10 km of a hydraulically fractured well. Rural residence was determined by 6-digit postal code, wherein the occurrence of a numeral 0 in the second digit of a postal code indicates a rural residence.17

This study received ethical approval from the Health Research Ethics Board of Alberta, Cancer Committee (REB19-0631). Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines were used. Informed consent was waived given secondary analysis of deidentified data.

Data Sources

Information about wells that were hydraulically fractured in Alberta from 2013 to 2018 was downloaded from the Alberta Energy Regulator (AER) website.18 Unique well identifier (UWI), production type, fracture start/end date, chemicals that were used, volume of water that was used, and bottom hole geographic (North American Datum of 1983) locations were available. In 2013, the AER mandated that all companies must report details of every hydraulic fracturing job in Alberta. The data are regularly inspected and audited to ensure compliance.19 The wellhead geographic locations and their status were obtained from GeoSCOUT using the UWIs for wells that were hydraulically fractured. As the health data were made available at the postal-code level, delivery postal points shapefile (Developing Mathematical Thinking Institute [DMTI] Spatial Inc) was downloaded from the University of Calgary digital library, which provided the geographic locations of the delivery points. Delivery points correspond to the local postal delivery units where the mail delivery person sorts the mail for their route.

Scrambled personal health numbers, sex, and dates of birth were used to deterministically link mothers and children across administrative health data holdings. Data on health outcomes and residential location were obtained from Alberta Health and Alberta Health Services. The Physician Claims database includes all fee-for-service billing data for outpatient physician visits, including the date of the visit and diagnosis/procedure codes.20 The Discharge Abstract Database and the National Ambulatory Care Reporting System contain data on acute hospitalizations and emergency department visits, respectively.

Exposure

Using ESRI ArcGIS, version 10.5, all data were projected into North American Datum of 1983/10TM AEP (resources), a projected coordinated reference system for province-wide spatial data management and mapping as defined by information from the government of Alberta.21 Unique postal delivery points (n = 504) were found for Alberta by cropping the DMTI postal delivery points shapefile to the province of Alberta (CanMap Postal Code Suite, DMTI Spatial Inc). The distance (in meters) between the surface location of hydraulically fractured wells within 10 km of postal delivery points, and the angle of the vector between the surface location of those wells and the delivery points were then calculated. The tabular data from postal points, GeoSCOUT, and the AER were linked using the UWI for further analyses with the health data. Maps of the surface locations of hydraulically fractured wells and of the postal delivery points were created for visualization.

Individuals were considered exposed if their postal delivery point was located within 10 km from at least 1 active well that was hydraulically fractured during preconception (1 year prior to conception date) and/or pregnancy (conception date to delivery date). A distance of 10 km or less was chosen to ensure adequate sample size and maintain comparability with other studies.11,22 Several characteristics were determined for each pregnancy: (1) distance to the nearest well active during preconception or pregnancy from their postal delivery point, (2) the number of wells within a 10-km radius of their postal delivery point (1-24, 25-49, 50-74, ≥100), (3) whether exposure occurred in preconception, pregnancy, or both, and (4) if exposed in pregnancy, whether exposure occurred in the first, second, third, or multiple trimesters.

Outcomes

Outcomes investigated were spontaneous and indicated preterm birth (< 37 weeks’ gestation), SGA, major congenital anomalies, and severe neonatal morbidity/mortality. Spontaneous preterm birth is defined as preterm labor with cervical dilation or preterm rupture of membranes, while indicated preterm birth is defined as births induced or delivered by cesarean birth owing to parental or fetal illness. Indicated preterm births were identified based on a Canadian Classification of Health Interventions code for labor induction in conjunction with a validated algorithm for identifying prelabor cesarean delivery.23,24,25,26 SGA, defined as newborn weight below the 10th percentile for gestational age, was determined based on Canadian reference standards.27 Major congenital anomalies easily visible or diagnosed at birth were identified using a validated algorithm.28 Severe neonatal morbidity/mortality was measured using the neonatal adverse outcome indicator, a validated algorithm that combines diagnosis and procedure codes related to the occurrence or treatment of severe events in the first 28 days after delivery.29

Covariates

Covariates included in our analyses were parental age at delivery, multiple births (ie, twins, triplets), infant sex, obstetric comorbidities, and area-level socioeconomic status. Obstetric comorbidities were identified using a validated composite measure that includes preexisting physical comorbidities, pregnancy-related conditions, and substance-related conditions.30,31 Area-level socioeconomic status at the postal-code level was measured using the Pampalon Index, a composite score comprising 6 indices of material and social deprivation.32,33 The score is divided into quintiles within Alberta, with quintile 1 representing the least deprived areas (highest socioeconomic status), and quintile 5 representing the most deprived areas (lowest socioeconomic status).

Statistical Analyses

Descriptive characteristics were computed using frequencies and percentages. Modified Poisson regression was used to derive adjusted risk ratios (aRRs) with 95% CIs for all analyses. The modified Poisson model uses a sandwich variance estimator to compute risk ratios of independent binary outcomes.34 Models were further fitted using generalized estimating equations to account for correlation within those individuals who had multiple unique pregnancies in the time frame of the study. We first measured the overall association between living within 10 km or less of a well hydraulically fractured during preconception or pregnancy and adverse birth outcomes compared with individuals who lived 10 km or more from the closest well (the referent). To evaluate whether there was a stronger association in those who lived closer to wells, we investigated the association between each outcome and the continuous distance in kilometers to the nearest well, among the exposed cohort. We then investigated well density by measuring the association between the number of wells within 10 km (1-24, 25-49, 50-74, 75-99, ≥100 wells) and each outcome.

Additionally, 2 sensitivity analyses were completed to evaluate whether associations differed depending on timing of exposure. The first investigated associations with outcomes when exposed in preconception alone, pregnancy alone, or both preconception and pregnancy, compared with those unexposed. The second investigated associations with SGA, neonatal morbidity/mortality, and any outcome when exposed in first, second, third, or multiple trimesters compared with those unexposed. Spontaneous and indicated preterm birth were not evaluated owing to risk of immortal time bias.35,36,37 Complete case analysis was used. Statistical analyses were conducted using SAS version 9.4 (SAS Institute Inc).

Results

A visualization of the 504 unique postal delivery point locations and the unique surface wells that underwent hydraulic fracturing between 2013 to 2018 in Alberta is presented in Figure 1. Of the 4871 unique surface wells that underwent hydraulic fracturing between 2013 and 2018, 226 of the 504 postal delivery points were less than 10 km from a well, with an average of 32 wells per postal point (range, 1-310). Descriptive characteristics of the study sample are presented in Table 1. After exclusions for missing postal codes (n = 1845), urban residence (n = 260 319), age outside 18 to 50 years (n = 691), missing infant sex (n = 65), missing area-level socioeconomic status (n = 4605), and living within 10 km of a hydraulic fracturing site without exposure during preconception or pregnancy (n = 6505), 26 193 individuals had 34 873 unique pregnancies during the study period. Comparisons of descriptive data for those with and without missing postal code data are presented in eTable 1 in the Supplement. Twenty-seven percent (n = 9158) were exposed to hydraulic fracturing within 10 km during preconception or pregnancy.

Figure 1. Visualization of Unique Surface Wells That Underwent Hydraulic Fracturing (HF) Between 2013 and 2018.

Figure 1.

Open in a new tab

Labeled are the second most populous and the capital city of Alberta, Canada (Edmonton) and the most populous city in Alberta, Canada (Calgary).

Table 1. Descriptive Characteristics of the Study Sample (N = 34 873).

Characteristic No. (%)
Parental age at delivery, mean (SD), y 28.2 (5.2)
Birth weight, mean (SD), g 3390 (565)
Obstetric comorbidities, any 10 557 (30.3)
Infant sex
Male 17 908 (51.4)
Female 16 965 (48.7)
Multiple births 478 (1.4)
Gestational age at delivery, wk
<32 401 (1.2)
32-33 312 (0.9)
34-36 1904 (5.5)
37-38 9357 (26.8)
39-40 18 693 (53.6)
≥41 4196 (12.0)
Area-level SES, quintile
1st (Highest SES) 926 (2.7)
2nd 3527 (10.1)
3rd 5387 (15.5)
4th 9072 (26.0)
5th (Lowest SES) 15 961 (45.8)
Adverse birth outcomes
Spontaneous preterm birth 1655 (4.8)
Indicated preterm birth 972 (2.8)
Small for gestational age 2355 (6.8)
Major congenital anomalies 257 (0.7)
Severe neonatal morbidity or mortality 2543 (7.3)
Open in a new tab

Abbreviation: SES, socioeconomic status.

Table 2 presents frequencies of each outcome, as well as crude and aRRs for exposure to a hydraulic fracturing site within 10 km in preconception or pregnancy compared with those unexposed. Among those exposed, there was a significant association with an increase in SGA (aRR, 1.12; 95% CI, 1.03-1.23) and major congenital anomalies (aRR, 1.31; 95% CI, 1.01-1.69). eTable 2 in the Supplement presents the crude and aRRs for the association between distance to the nearest well as a continuous measure and adverse birth outcomes among the exposed cohort. No association was found between any of the adverse birth outcomes and distance to the nearest well.

Table 2. Risk Ratios and 95% CIs for Adverse Birth Outcomes in Those Exposed Compared With Those Unexposed.

Outcome No./Total No. (%) RR (95% CI)
No. exposed with outcome/total exposed (%) No. unexposed with outcome/total unexposed (%) Crude Adjusteda
Spontaneous preterm birth 422/9158 (4.6) 1233/25 715 (4.8) 0.96 (0.86-1.07) 0.97 (0.87-1.08)
Indicated preterm birth 260/9158 (2.8) 712/25 715 (2.8) 1.03 (0.89-1.18) 1.01 (0.88-1.16)
Small for gestational age 671/9158 (7.3) 1684/25 715 (6.5) 1.13 (1.03-1.23) 1.12 (1.03-1.23)
Major congenital anomalies 82/9158 (0.9) 175/25 715 (0.7) 1.32 (1.01-1.71) 1.31 (1.01-1.69)
Severe neonatal morbidity or mortality 694/9158 (7.6) 1849/25 715 (7.2) 1.05 (0.97-1.15) 1.07 (0.97-1.17)
Open in a new tab

Abbreviation: RR, risk ratio.

a

All modified Poisson regression models adjusted for parental age at delivery, area-level socioeconomic status, multiple births, infant sex, and obstetric comorbidities.

Figure 2 and eTable 3 in the Supplement present aRRs of each outcome comparing the association of the number of hydraulic fracturing wells within 10 km (1-24, 25-49, 50-74, 75-99, ≥100) using 1 to 24 wells as the reference. In individuals exposed to and living with 100 or more wells within 10 km (n = 414) there was an association with an increased risk of spontaneous preterm birth (aRR, 1.64; 95% CI, 1.04-2.60) and SGA (aRR, 1.65; 95% CI, 1.10-2.48). No association was found at any other level of exposure (25-49, 50-74, 75-99 wells) for any of the adverse birth outcomes.

Figure 2. Forest Plot of Adjusted Risk Ratios With 95% CIs for Adverse Birth Outcomes in Pregnant Individuals .

Figure 2.

Open in a new tab

Results of the first sensitivity analysis measuring the association regarding exposure during 1 year preconception, pregnancy, both, or neither are presented in Table 3. No differences in risk were found for spontaneous preterm birth, indicated preterm birth, or major congenital anomalies. Exposure in both preconception and pregnancy was associated with increased risk of SGA (aRR, 1.18; 95% CI, 1.05-1.33) and severe neonatal morbidity or mortality (aRR, 1.19; 95% CI, 1.07-1.33).

Table 3. Risk Ratios and 95% CIs for Adverse Birth Outcomes in Individuals Living Within or Near a Hydraulic Fracturing Site During Preconception Compared With Those Unexposed .

Outcome Timing of exposure RR (95% CI)
Crude Adjusteda
Spontaneous preterm birth Unexposed 1 [Reference] 1 [Reference]
Preconception 1.17 (0.98-1.40) 1.18 (0.99-1.40)
Pregnancy 0.93 (0.77-1.12) 0.92 (0.77-1.12)
Both 0.87 (0.74-1.01) 0.88 (0.75-1.03)
Indicated preterm birth Unexposed 1 [Reference] 1 [Reference]
Preconception 1.16 (0.92-1.47) 1.12 (0.89-1.41)
Pregnancy 0.79 (0.60-1.04) 0.78 (0.59-1.02)
Both 1.09 (0.90-1.31) 1.08 (0.90-1.30)
Small for gestational age Unexposed 1 [Reference] 1 [Reference]
Preconception 0.99 (0.84-1.17) 0.99 (0.84-1.17)
Pregnancy 1.15 (1.00-1.32) 1.14 (0.99-1.32)
Both 1.19 (1.06-1.33) 1.18 (1.05-1.33)
Major congenital anomalies Unexposed 1 [Reference] 1 [Reference]
Preconception 1.32 (0.84-2.08) 1.34 (0.86-2.11)
Pregnancy 1.42 (0.93-2.18) 1.43 (0.94-2.19)
Both 1.22 (0.85-1.74) 1.19 (0.83-1.70)
Severe neonatal morbidity or mortality Unexposed 1 [Reference] 1 [Reference]
Preconception 1.04 (0.90-1.21) 1.05 (0.91-1.22)
Pregnancy 0.88 (0.75-1.03) 0.89 (0.76-1.05)
Both 1.16 (1.04-1.30) 1.19 (1.07-1.33)
Open in a new tab

Abbreviation: RR, risk ratios.

a

All modified Poisson regression models adjusted for parental age at delivery, area-level socioeconomic status, multiple births, infant sex, and obstetric comorbidities.

Results of the second sensitivity analysis measuring the association for exposure by trimester of pregnancy are presented in eTable 4 in the Supplement. Risk of SGA was increased in those exposed in the third trimester alone (aRR, 1.22; 95% CI, 1.02-1.47) compared with those unexposed. Risk of severe neonatal morbidity or mortality was increased in those exposed in the second trimester alone (aRR, 1.25; 95% CI, 1.03-1.52).

Discussion

In this study, we measured the association between residential proximity to hydraulic fracturing sites and birth outcomes in pregnant individuals in rural Alberta, Canada. We found that the risk of SGA and major congenital anomalies was increased for pregnant individuals living less than 10 km from a hydraulically fractured well during 1 year preconception or pregnancy. Risk of any of the measured outcomes did not increase when exposure was limited to distance to the nearest well, but risk of spontaneous preterm birth and SGA were increased in those individuals living within 10 km of more than 100 wells. Timing of exposure may also be relevant to the association: exposure in both the preconception period and during pregnancy showed increased risk of several adverse outcomes, but exposure in only 1 of the 2 time periods had no association. Together, these results suggest that the association of hydraulic fracturing is not in proximity but in the density of operations in a given area and the cumulative amount of exposure an individual may have.

Three key advantages arise from using data from Alberta. First is the ability to examine health outcomes associated with hydraulic fracturing in the absence of an income shock. In the US, landowners are often also the owners of the subsurface mineral rights and therefore can see large windfalls in the form of royalties and bonus payments when fracking takes place on their properties, which is a concern given income is a known confounder of birth outcomes.38,39 In Alberta, most minerals are owned by the Crown and therefore beyond a small surface lease payment, we do not expect household income to change with nearby fracking.40 Second, by focusing on Alberta, bias arising from self-selection into treatment was less of a concern. In the US context, many households own the subsurface minerals, which means they have the choice of whether to allow drilling on their properties, a choice that could be correlated with factors also influencing health outcomes.38 Crown ownership of minerals in Alberta means that households do not have the choice as to whether a well will be drilled on their property.40,41 Third, the high density of drilling in Alberta provided unique variation in the extent to which mothers were exposed. We observed large variation in the number of wells fracked nearby, allowing us to examine differential associations by development intensity.

Our study adds to a growing body of literature in which several studies have found associations between unconventional oil and gas development and adverse birth outcomes. To our knowledge, only 1 other study of this type has been conducted in Canada; Caron-Beaudoin et al22 investigated hydraulic fracturing in northeastern British Columbia and identified nonlinear associations between proximity and density of hydraulically fractured wells, low birth weight, and preterm birth. All other studies of hydraulic fracturing and birth outcomes have been conducted in the US, in a different legislative, geographical, and geological context. Casey et al2 found that residential proximity to hydraulic fracturing increased the odds of preterm birth (odds ratio, 1.4; 95% CI, 1.0-1.9) and high-risk pregnancies (odds ratio, 1.3; 95% CI, 1.1-1.7). Currie et al11 reported that parental residential exposure to hydraulic fracturing during pregnancy was negatively associated with birth weight (P < .001) and a composite outcome composing low birth weight, preterm birth, congenital anomalies, and other adverse neonatal events. Hill et al8 observed that residential proximity to hydraulic fracturing during pregnancy was associated with a 3% increase in preterm birth and a 7% increase in low birth weight.

There are multiple mechanisms whereby residential proximity may lead to human exposure to chemicals used in the fracking process and pollutants that are released into the environment.6 Although groundwater contamination from hydraulic fracturing via subsurface pathways has not been clearly demonstrated, groundwater changes have been attributed to surface spills.42 This is particularly important in the Alberta context, as it is estimated that for approximately 90% of rural Albertans, the primary source of drinking water is groundwater.43 The drilling and extraction process can also lead to changes in air quality.6 Hydraulic fracturing involves the transport of large quantities of water (majority surface-sourced in Alberta), often via heavy trucks, which could lead to increased air pollution.6,44 Overall, thousands of chemicals are used in fracking fluids; however, only 240 of these have toxicity information available.1,7 Of those 240 chemicals, 103 (43%) are linked to reproductive toxicity.1,7 Thus, pregnant individuals may be especially susceptible to hydraulic fracturing exposure that occurs during critical stages of gestation.11,45 Two biomonitoring studies have shown increased levels of contaminants associated with hydraulic fracturing in the urine and hair of pregnant individuals living in areas with high oil and gas activity.46,47

Limitations

Our study has several limitations. As we only know the postal code of individuals in the data set and not their exact address, their locations are approximated to the delivery point of their postal code. In less populated areas where postal codes are larger, using the delivery point may underestimate or overestimate the distance from residence to nearest well. We also have not delineated between the different phases of oil and gas production (eg, drilling, production) which have been shown to be associated with adverse birth outcomes and should be considered in future studies.2,48 We have also used categorical rather than continuous variables to measure well density. This was decided a priori to ensure adequate sample size in each group but subsequently limits our ability to capture the true association. We encourage future groups to find optimal cut points in their analyses. Administrative health data do not capture all variables relevant to reproductive health. While we have attempted to approximate some of these variables, there are multiple confounders that we cannot control for and we acknowledge that our results do not reflect the full complexity of pregnancy-related health. For example, pregnancy morbidity is associated with lack of access to healthy food choices, a variable that cannot be properly approximated by the simplicity of rural living vs urban living.49 An E-value analysis was completed to quantify unmeasured confounding and is included in eTable 5 in the Supplement.50,51

Conclusions

In this retrospective cohort study, close residential proximity to high numbers of hydraulic fracturing sites was correlated with adverse birth outcomes. Though we cannot ascertain in this analysis the mechanisms (eg, air, water) through which hydraulic fracturing and oil and gas development may be increasing the risk for adverse health outcomes, we encourage additional research to further elucidate these potential pathways.

Supplement.

eTable 1. Descriptive statistics of those with (n=34,873) and without (n=1,845) postal code.

eTable 2. Risk ratios (RR) and 95% confidence intervals (CI) for adverse birth outcomes in women living within 10km of ≥1 hydraulic fracturing site during preconception or pregnancy (n=9,158), using distance in kilometres from the closest hydraulic fracturing site as a continuous independent variable.

eTable 3. Risk ratios (RR) and 95% confidence intervals (CI) for adverse birth outcomes in women living within 10km of ≥1 hydraulic fracturing site during preconception or pregnancy (n=9,158), comparing the effect of number of wells in a 10km radius.

eTable 4. Risk ratios (RR) and 95% confidence intervals (CI) for adverse birth outcomes in women living within 10km of ≥1 hydraulic fracturing site during their first trimester (n=1,154), second trimester (n=1,108), third trimester (n=1,330), or across multiple trimesters (n=3,236), compared to those unexposed (n=25,715).

eTable 5. E-values for each analysis with statistically significant results. E-values indicate the association a confounder would need to have with both the exposure and the outcome to entirely explain the association found.

Click here for additional data file. (184.8KB, pdf)

References

  • 1.Balise VD, Meng C-X, Cornelius-Green JN, Kassotis CD, Kennedy R, Nagel SC. Systematic review of the association between oil and natural gas extraction processes and human reproduction. Fertil Steril. 2016;106(4):795-819. doi: 10.1016/j.fertnstert.2016.07.1099 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Casey JA, Savitz DA, Rasmussen SG, et al. Unconventional natural gas development and birth outcomes in Pennsylvania, USA. Epidemiology. 2016;27(2):163-172.doi: 10.1097/EDE.0000000000000387 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Canadian Association of Physicians for the Environment (CAPE) . Fractures in the bridge: unconventional (fracked) natural gas, climate change and human health. 2020. https://cape.ca/wp-content/uploads/2020/01/CAPE-Fracking-Report-EN.pdf
  • 4.Watterson A. Speakers corner. to frack or not to frack? why is that not a global public health question and how should public health practitioners address it? J Epidemiol Community Health. 2016;70(3):219-220. doi: 10.1136/jech-2014-205359 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Reap E. The risk of hydraulic fracturing on public health in the UK and the UK’s fracking legislation. Environ Sci Eur. 2015;27(1):27. doi: 10.1186/s12302-015-0059-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Field RA, Soltis J, Murphy S. Air quality concerns of unconventional oil and natural gas production. Environ Sci Process Impacts. 2014;16(5):954-969. doi: 10.1039/C4EM00081A [DOI] [PubMed] [Google Scholar]
  • 7.Elliott EG, Ettinger AS, Leaderer BP, Bracken MB, Deziel NC. A systematic evaluation of chemicals in hydraulic-fracturing fluids and wastewater for reproductive and developmental toxicity. J Expo Sci Environ Epidemiol. 2017;27(1):90-99. doi: 10.1038/jes.2015.81 [DOI] [PubMed] [Google Scholar]
  • 8.Hill EL. Shale gas development and infant health: evidence from Pennsylvania. J Health Econ. 2018;61:134-150. doi: 10.1016/j.jhealeco.2018.07.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Whitworth KW, Marshall AK, Symanski E. Maternal residential proximity to unconventional gas development and perinatal outcomes among a diverse urban population in Texas. PLoS One. 2017;12(7):e0180966. doi: 10.1371/journal.pone.0180966 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Busby C, Mangano JJ. There’s a world going on underground—infant mortality and fracking in Pennsylvania. J Environ Prot. 2017;8(4):381. doi: 10.4236/jep.2017.84028 [DOI] [Google Scholar]
  • 11.Currie J, Greenstone M, Meckel K. Hydraulic fracturing and infant health: new evidence from Pennsylvania. Sci Adv. 2017;3(12):e1603021. doi: 10.1126/sciadv.1603021 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Stacy SL, Brink LL, Larkin JC, et al. Perinatal outcomes and unconventional natural gas operations in southwest Pennsylvania. PLoS One. 2015;10(6):e0126425. doi: 10.1371/journal.pone.0126425 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Government of Alberta . Canada. Vital statistics (births and deaths)—Alberta, census divisions and economic regions. Updated September 29, 2020. Accessed date March 2, 2022. https://open.alberta.ca/opendata/vital-statistics-births-and-deaths-alberta-census-divisions-economic-regions
  • 14.Government of Alberta, Canada . Oil and gas liabilities management. Accessed March 2, 2022. https://www.alberta.ca/oil-and-gas-liabilities-management.aspx
  • 15.Abboud J, Watson T, Ryan M. Fugitive methane gas migration around Alberta’s petroleum wells. Greenh Gases. 2021;11(1):37-51. doi: 10.1002/ghg.2029 [DOI] [Google Scholar]
  • 16.Government of Alberta . Distance requirements. December 1, 2013. Accessed January 21, 2019. https://open.alberta.ca/dataset/e8b7dda0-31a0-4350-be47-bcb52fc815b8/resource/5b4aa7ae-6cd6-457c-9b61-adb532cb6360/download/2013-ed2006-15-distance-requirements-updated-dececember-1-2013.pdf
  • 17.Canada Post . Addressing guidelines. Updated January 18, 2021. Accessed March 2, 2022. https://www.canadapost.ca/tools/pg/manual/PGaddress-e.asp?ecid=murl10006450
  • 18.Alberta Energy Regulator . Hydraulic fracturing water use data. Updated December 17, 2020. Accessed March 2, 2022. https://www1.aer.ca/ProductCatalogue/530.html
  • 19.Alberta Energy Regulator . Hydraulic fracturing. Accessed March 2, 2022. https://www.aer.ca/providing-information/by-topic/hydraulic-fracturing
  • 20.Cunningham CT, Cai P, Topps D, Svenson LW, Jetté N, Quan H. Mining rich health data from Canadian physician claims: features and face validity. BMC Res Notes. 2014;7(1):682. doi: 10.1186/1756-0500-7-682 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.GeoRepository . NAD83(CSRS) / Alberta 10-TM (resource); projected CRS used in Canada—Alberta. Updated October 19, 2020. March 2, 2022. https://epsg.org/crs_3403/NAD83-CSRS-Alberta-10-TM-Resource.html
  • 22.Caron-Beaudoin É, Whitworth KW, Bosson-Rieutort D, Wendling G, Liu S, Verner MA. Density and proximity to hydraulic fracturing wells and birth outcomes in Northeastern British Columbia, Canada. J Expo Sci Environ Epidemiol. 2021;31(1):53-61. doi: 10.1038/s41370-020-0245-z [DOI] [PubMed] [Google Scholar]
  • 23.Gregory KD, Korst LM, Gornbein JA, Platt LD. Using administrative data to identify indications for elective primary cesarean delivery. Health Serv Res. 2002;37(5):1387-1401. doi: 10.1111/1475-6773.10762 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Henry OA, Gregory KD, Hobel CJ, Platt LD. Using ICD-9 codes to identify indications for primary and repeat cesarean sections: agreement with clinical records. Am J Public Health. 1995;85(8 pt 1):1143-1146. doi: 10.2105/AJPH.85.8_Pt_1.1143 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Liu S, Joseph KS, Hutcheon JA, et al. ; Maternal Health Study Group of Canadian Perinatal Surveillance System . Gestational age-specific severe maternal morbidity associated with labor induction. Am J Obstet Gynecol. 2013;209(3):209.e1-209.e8. doi: 10.1016/j.ajog.2013.05.033 [DOI] [PubMed] [Google Scholar]
  • 26.Liu S, Liston RM, Joseph KS, Heaman M, Sauve R, Kramer MS; Maternal Health Study Group of the Canadian Perinatal Surveillance System . Maternal mortality and severe morbidity associated with low-risk planned cesarean delivery versus planned vaginal delivery at term. CMAJ. 2007;176(4):455-460. doi: 10.1503/cmaj.060870 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Kramer MS, Platt RW, Wen SW, et al. ; Fetal/Infant Health Study Group of the Canadian Perinatal Surveillance System . A new and improved population-based Canadian reference for birth weight for gestational age. Pediatrics. 2001;108(2):E35. doi: 10.1542/peds.108.2.e35 [DOI] [PubMed] [Google Scholar]
  • 28.Metcalfe A, Sibbald B, Lowry RB, Tough S, Bernier FP. Validation of congenital anomaly coding in Canada’s administrative databases compared with a congenital anomaly registry. Birth Defects Res A Clin Mol Teratol. 2014;100(2):59-66. doi: 10.1002/bdra.23206 [DOI] [PubMed] [Google Scholar]
  • 29.Lain SJ, Algert CS, Nassar N, Bowen JR, Roberts CL. Incidence of severe adverse neonatal outcomes: use of a composite indicator in a population cohort. Matern Child Health J. 2012;16(3):600-608. doi: 10.1007/s10995-011-0797-6 [DOI] [PubMed] [Google Scholar]
  • 30.Bateman BT, Mhyre JM, Hernandez-Diaz S, et al. Development of a comorbidity index for use in obstetric patients. Obstet Gynecol. 2013;122(5):957-965. doi: 10.1097/AOG.0b013e3182a603bb [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Metcalfe A, Lix LM, Johnson JA, Currie G, et al. Validation of an obstetric comorbidity index in an external population. BJOG. 2015;122(13):1748-1755. doi: 10.1111/1471-0528.13254 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Adhikari Dahal K, Premji S, Patel AB, Williamson T, Peng M, Metcalfe A. Variation in maternal co-morbidities and obstetric interventions across area-level socio-economic status: a cross-sectional study. Paediatr Perinat Epidemiol. 2017;31(4):274-283. doi: 10.1111/ppe.12370 [DOI] [PubMed] [Google Scholar]
  • 33.Pampalon R, Raymond G. A deprivation index for health and welfare planning in Quebec. Chronic Dis Can. 2000;21(3):104-113. [PubMed] [Google Scholar]
  • 34.Zou GY, Donner A. Extension of the modified Poisson regression model to prospective studies with correlated binary data. Stat Methods Med Res. 2013;22(6):661-670. doi: 10.1177/0962280211427759 [DOI] [PubMed] [Google Scholar]
  • 35.Corrao G, Rea F, Franchi M, Beccalli B, Locatelli A, Cantarutti A. Warning of immortal time bias when studying drug safety in pregnancy: application to late use of antibiotics and preterm delivery. Int J Environ Res Public Health. 2020;17(18):6465. doi: 10.3390/ijerph17186465 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Matok I, Azoulay L, Yin H, Suissa S. Immortal time bias in observational studies of drug effects in pregnancy. Birth Defects Res A Clin Mol Teratol. 2014;100(9):658-662. doi: 10.1002/bdra.23271 [DOI] [PubMed] [Google Scholar]
  • 37.Neophytou AM, Kioumourtzoglou M-A, Goin DE, Darwin KC, Casey JA. Educational note: addressing special cases of bias that frequently occur in perinatal epidemiology. Int J Epidemiol. 2020;50(1):337-345. doi: 10.1093/ije/dyaa252 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Fitzgerald T. Importance of mineral rights and royalty interests for rural residents and landowners. Choices. 2014;29(4):1-7. doi: 10.22004/ag.econ.195351 [DOI] [Google Scholar]
  • 39.Kinnaman TC. The economic impact of shale gas extraction: A review of existing studies. Ecol Econ. 2011;70(7):1243-1249. doi: 10.1016/j.ecolecon.2011.02.005 [DOI] [Google Scholar]
  • 40.Stewart FL. Non-compliance in a dangerous time: the pitfalls of section 27 of the surface rights act. The Negotiator. March 2, 2022. https://www.academia.edu/36047189/Non_Compliance_in_a_Dangerous_Time_The_Pitfalls_of_Section_27_of_The_Surface_Rights_Act
  • 41.Fluker S. The right to public participation in resources and environmental decision-making in Alberta. Alta Law Rev. 2014;52(3):567. doi: 10.29173/alr24 [DOI] [Google Scholar]
  • 42.Kassotis CD, Harkness JS, Vo PH, et al. Endocrine disrupting activities and geochemistry of water resources associated with unconventional oil and gas activity. Sci Total Environ. 2020;748:142236. doi: 10.1016/j.scitotenv.2020.142236 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Government of Alberta, Canada . Facts about water in Alberta. Updated 2010. Accessed August 1, 2021. https://open.alberta.ca/dataset/1832cd36-bbeb-4997-ae81-67d3eedfcfe5/resource/18a9d64b-bad8-413a-8c63-77a548ec9d88/download/4888138-2010-facts-about-water-in-alberta-2010-12.pdf
  • 44.Muehlenbachs L, Staubli S, Chu Z. The accident externality from trucking: evidence from shale gas development. Reg Sci Urban Econ. 2021;88:103630. doi: 10.1016/j.regsciurbeco.2020.103630 [DOI] [Google Scholar]
  • 45.Peng L, Meyerhoefer C, Chou SY. The health implications of unconventional natural gas development in Pennsylvania. Health Econ. 2018;27(6):956-983. doi: 10.1002/hec.3649 [DOI] [PubMed] [Google Scholar]
  • 46.Caron-Beaudoin É, Bouchard M, Wendling G, et al. Urinary and hair concentrations of trace metals in pregnant women from Northeastern British Columbia, Canada: a pilot study. J Expo Sci Environ Epidemiol. 2019;29(5):613-623. doi: 10.1038/s41370-019-0144-3 [DOI] [PubMed] [Google Scholar]
  • 47.Caron-Beaudoin É, Valter N, Chevrier J, Ayotte P, Frohlich K, Verner M-A. Gestational exposure to volatile organic compounds (VOCs) in northeastern British Columbia, Canada: a pilot study. Environ Int. 2018;110:131-138. doi: 10.1016/j.envint.2017.10.022 [DOI] [PubMed] [Google Scholar]
  • 48.Walker Whitworth K, Kaye Marshall A, Symanski E. Drilling and production activity related to unconventional gas development and severity of preterm birth. Environ Health Perspect. 2018;126(3):037006. doi: 10.1289/EHP2622 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Tipton MJ, Wagner SA, Dixon A, Westbay L, Darji H, Graziano S. Association of living in a food desert with pregnancy morbidity. Obstet Gynecol. 2020;136(1):140-145. doi: 10.1097/AOG.0000000000003868 [DOI] [PubMed] [Google Scholar]
  • 50.Mathur MB, Ding P, Riddell CA, VanderWeele TJ. The web site and R package for computing E-values. Epidemiology. 2018;29(5):e45-e47. doi: 10.1097/EDE.0000000000000864 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.VanderWeele TJ, Ding P. Sensitivity analysis in observational research: introducing the E-value. Ann Intern Med. 2017;167(4):268-274. doi: 10.7326/M16-2607 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplement.

eTable 1. Descriptive statistics of those with (n=34,873) and without (n=1,845) postal code.

eTable 2. Risk ratios (RR) and 95% confidence intervals (CI) for adverse birth outcomes in women living within 10km of ≥1 hydraulic fracturing site during preconception or pregnancy (n=9,158), using distance in kilometres from the closest hydraulic fracturing site as a continuous independent variable.

eTable 3. Risk ratios (RR) and 95% confidence intervals (CI) for adverse birth outcomes in women living within 10km of ≥1 hydraulic fracturing site during preconception or pregnancy (n=9,158), comparing the effect of number of wells in a 10km radius.

eTable 4. Risk ratios (RR) and 95% confidence intervals (CI) for adverse birth outcomes in women living within 10km of ≥1 hydraulic fracturing site during their first trimester (n=1,154), second trimester (n=1,108), third trimester (n=1,330), or across multiple trimesters (n=3,236), compared to those unexposed (n=25,715).

eTable 5. E-values for each analysis with statistically significant results. E-values indicate the association a confounder would need to have with both the exposure and the outcome to entirely explain the association found.

Click here for additional data file. (184.8KB, pdf)

Articles from JAMA Pediatrics are provided here courtesy of American Medical Association

RESOURCES