Multi-State Analysis Of Unconventional Oil And Gas Development And Rates Of Sexually Transmitted Infections In The United States

Nicholaus Peter Johnson

This thesis is restricted to Yale network users only. It will be made publicly available on 08/28/2021



Advances in unconventional oil and gas development (UOGD), including hydraulic fracturing and horizontal drilling, have helped position the United States as a world leader in the production of both crude oil and natural gas hydrocarbons. This significant increase in production would not have been possible, however, without the assemblage of a workforce that is predominantly young, male, and mobile. This is significant given that a number of studies have associated similar types of mobile workforces – especially those employed by resource extractive industries – with impacts on the sexual health of the host communities in which these workers operate (Westwood & Orenstein, 2016; Udoh et al., 2008). For instance, two ecologic studies have explored the association between shale activity and county level rates of reportable STI in counties overlying the Marcellus Shale, with positive associations being reported in each. While their findings are novel, these studies are restricted in their geographic scope. Some of the largest and most productive oil and gas formations are located in other regions of the United States including Colorado, North Dakota, and Texas. A multi-state, multi-region analysis of the association between shale activity and county level rates of reportable STI – chlamydia, gonorrhea, and primary/secondary syphilis – is needed to confirm whether this previously observed association is a unique place-based phenomenon or a generalizable result of UOGD.


To evaluate the association between shale drilling activity and rates of STI in Colorado, North Dakota, and Texas, we conducted an ecologic study that utilized annual census, STI, and drilling activity data at the county level from 2000-2016. We requested the annual rates (cases per 100,000 population) of chlamydia, gonorrhea, and primary/secondary syphilis for the 64 counties in Colorado, 53 counties in North Dakota, and 254 counties in Texas using the Centers for Disease Control and Prevention’s (CDC) National Center for HIV/AIDS, Viral Hepatitis, STD, and TB Prevention (NCHHSTP) Atlas Plus tool. Data pertaining to drilling activity was collected annually for each county through DrillingInfo’s DI Application. Each county year was classified according to its shale activity: none (0 spuds targeting shale), low (1-49 spuds targeting shale), and high (≥ 50 spuds targeting shale). Data for sociodemographic covariates was obtained from the Decennial Census and American Community Survey (ACS) provided by the U.S. Census Bureau’s American Fact Finder tool. We used multivariable mixed effects Poisson regression models to estimate rate ratios (RR) with 95% CI for the association between county year shale activity and cases of chlamydia, gonorrhea, and primary/secondary syphilis for each state, while adjusting for potential confounders and secular trends.


Compared to county years with no shale activity, Texas county years with high shale activity (≥ 50 shale spuds) had 10% (RR = 1.10; 95% CI = 1.04, 1.17) increased rates of chlamydia and 15% (RR = 1.15; 95% CI = 1.04, 1.28) increased rates of gonorrhea. No association was observed for syphilis in Texas. An association between shale activity and annually reported rates of chlamydia, gonorrhea, and syphilis were not observed for Colorado and North Dakota.


The first multi-state, multi-region analysis of shale activity and annual STI rates confirmed previously observed associations in Texas counties, for the association between shale drilling activity and annual rates of chlamydia and gonorrhea. While elevated rates in the unadjusted models in Colorado and North Dakota suggested a positive association, these effect estimates did not remain after rigorous adjustments for sociodemographic covariates, secular trends, county-level random effects, and observation-level random effects.