INTERVIEW: OSU’s Mary Willis Talks About Birth Weight Near Drilling Sites

It’s been a known fact that drilling for oil brings with it the potential for good and bad. The good largely lies in better fortunes for both landowners and workers. But do we really know all of the bad?

Dr. Mary Willis of Oregon State University has been part of a study concerning birth weights in areas closest to oil drilling and fracking.

TCA: Hi, I’m Sally Lehman, and I’m with The Advocate. Today, I’m speaking with Dr. Mary Willis, who has recently written up a post-doctoral study linking lower birth weight to proximity to oil drilling rigs. Thank you for being here today, Dr. Willis.

Willis: Thank you so much for inviting me.

TCA: What is your doctorate in?

Willis: My doctorate is in environmental health with a minor in biological data science. So functionally, when you combine those together, it’s environmental epidemiology.

TCA: And why did you choose this area to research?

Willis: When I was a Masters student in undergrad, I was living in Rochester, New York during the height of all of the fracking discussions there – with respect to should fracking be allowed in New York state. I was fortunate enough to meet Dr Elaine Hill, who was working on some oil and gas work in Pennsylvania and very heavily involved in the policy discussions in New York. 

As I worked with her, I realized that this was a heavily under researched area, yet it’s a rapidly expanding hazard for local populations. So I continued that through my doctorate and some of my postdoctoral work.

TCA: According to your study, babies gestated near oil drilling will tend to have a lower birth rate. How much lower were the birth rates?

Willis: It’s oil and gas drilling – what we’re looking at here. So we’re not specifically looking at just oil or just gas. 

In this study, we find about a seven to ten gram decrease in birth weight among mothers who are residing near drilling during their pregnancy.

TCA: Is that a significant difference?

Willis: It’s a very small difference at the individual level, but it has potential implications for population health.

TCA: Your study found that these changes occur for infants born three kilometers or just under two miles from the drilling rigs. Were there higher incidence or other noteworthy issues closer to the drilling?

Willis: In our study, surprisingly, not particularly. We didn’t see any sort of dose response gradient that you’d expect.

TCA: You also say that there were more early pregnancy losses and difficulty in conceiving among the families in that area. Were those numbers significantly higher than the norm?

Willis: Oh, that wasn’t in our study. That’s just on future things we’re going to be working on. Our study was exclusively on term births. So births between 37 and 42 weeks gestation.

TCA: You say that the health impact doesn’t dissipate at three kilometers. At what distance does the health impact lessen?

Willis: So we designed this study to be very specifically related to what could be coming out of the drilling site itself. And based on existing literature by groups of amazing scientists around the world, about three kilometers away is when we wouldn’t expect it to be directly an impact from the drilling site itself anymore. So for our study, we didn’t go out any further. We just looked at zero to three kilometers and different buffer distances. 

TCA: Could this be more of a story of correlation rather than causation? Meaning is it possible that the lower birth rate could be attributed to the fact that housing closer to the [drilling] rigs tends to be cheaper and perhaps the women living there couldn’t afford healthy foods or were more likely to smoke?

Willis: While we controlled for smoking in our study, we can’t address any of that really cause it’s observational research. All observational research has the potential for unmeasured confounding infiltrating our results. So something that we couldn’t measure in this massive birth cohort. 

We looked at 2.5 million births, roughly over a 15 year period. That’s the trade off of using this massive big data set. We can’t get at all of those little potential individual differences in the study.

TCA: Another issue raised was flaring. Can you explain that for us?

Willis: So when you have a natural gas or oil development site, you occasionally have times at which there is too much of the resource coming out of the ground – either too much of the resource for your local infrastructure to handle what’s on the site, like it could physically blow out, or it’s perhaps not economically viable. If that happens, in either of those scenarios, a site has the option to vent or flare it, which would be just either letting the gas into the atmosphere or burning it off and having all of those combustion products going into the air. You probably see a lot of them in different news sources of just the stacks with the flame coming off of it. 

TCA: Dramatic to see. Yes. So what about flaring off would affect the children born?

Willis: Our study doesn’t address that, but Dr Lara Cushing at University of California, San Francisco just published an amazing study in the same journal where our study was published on showing that infants within five kilometers of a flare had a lower birth weights and higher risk of preterm birth, if I recall correctly. So our study can’t address that, but other people have.

TCA: What would be the long term effects to a child born with a lower birth weight?

Willis: In this particular, we don’t necessarily know exactly. We can’t tell you what is the long term effect of fracking or oil and gas drilling or any sorts of those components. 

There is potential that having a lower birth weight in utero could be associated with long term chronic health conditions later in life. A lot of other studies have done work on that, just not specifically with respect to oil and gas drilling. So no one’s done a cohort, but it’s been followed up for a really long time.

TCA: Your study takes into account the economic benefits to being near a drilling site. What are those and how would they improve a child’s life in the long term?

Willis: Our study is designed to take into account that when a drilling boom comes to an area, there are massive changes to the community. 

So you’ve got, as we talked about already, the environmental pollution from the oil and gas drilling itself. You’ve simultaneously got a change in your local demographics from new people coming into your community, new job opportunities, that sort of thing. And then you also have a new tax base coming into your area potentially, that could yield different money being allocated into your community or communities that are further away, maybe benefiting as well. The literature is mixed on if it’s a true benefit to the people living closest to it, or if it’s mostly a benefit to people living further away. 

However, our study was concerned that that fundamental shift in dynamics in the community needs to be taken into account in order to fully understand the association between oil and gas drilling in adverse birth outcomes. So our study design is called a difference in differences design. We take into account what the community was like beforehand – before drilling occurred, and then compare in the same area afterwards, as well as simultaneously comparing to a control group that’s farther away before and after. 

So we’re uniquely able to look at how this shift that’s happening, not necessarily measuring every little component of it, but we know there is a shift and then seeing do we still see an effect on birth weight, and we do in this study. And that aligns with the other two studies out there that used similar studies in Pennsylvania.

TCA: So Oregon does have a ban on offshore drilling. Are there any bans on fracking in the state of Oregon?

Willis: Well, I believe there is technically a ban in Oregon. However, Oregon’s geology isn’t necessarily somewhere where fracking would be viable in the long run. If we [were] necessarily on a pocket of shale, that would be somewhere where current technologies would be viable to go drill.

TCA: And where will your post-doctoral research go from here?

Willis: Right now, I am jointly working at Oregon State University as a postdoctoral scholar, and I’m also physically at Boston University working in the Department of Epidemiology. 

My work at Boston University is with the doctors Lauren Wise and [Elizabeth] Hatch, where we’re looking at pregnancy study online – a preconception cohort study that we’ll be able to potentially look at the effects of oil and gas drilling on fertility –so are couples actually able to get pregnant living near drilling, and looking at other things like pregnancy outcomes. So bringing the observational work that I’ve done here, that we’re discussing, into an actual cohort framework so we can interact with or ask questions of the people and control for the things you brought up before – like what are they eating, are they smokers, that sort of thing.

TCA: Is there a reason so much of your research has been aimed toward pregnant women and the weights of newborn babies?

Willis: To a degree, yes. 

One of the things that’s uniquely understudied is just women’s health in general. And so when you start bringing it towards infants, at least that is something where you have a lot of data available, like vital statistics data – birth certificates are available for everyone. So you’re able to get access to enough data to look at these really small changes in health effects that you wouldn’t be able to necessarily look at otherwise. 

And then on top of that, we don’t necessarily always look at pregnancy as a women’s health condition. It’s often all about the baby. So bringing this back towards my future work is bringing this more towards what is going on with these women in the long run after they have their kid.

TCA: Where do you see your research going in the next five to ten years?

Willis: I’m planning on continuing with this oil and gas drilling framework and to see what else is going on here. I mean, there’s roughly 30 studies on the health effects of the oil and gas industry and an epidemiologic context. Yet, 5.4% of the US population lives within one mile of an oil and gas drilling site, that’s 17.6 million people if I recall correctly. It’s vastly understudied. So for now, as long as we’re continuing to have oil and gas development in the US, I plan to continue working on figuring out what the health effects may be to better help policymakers have the data they need to make the right decisions, whatever that may be.

TCA: Thank you so much for your time, Dr. Willis, and I appreciate you being here today.

Willis: Thank you so much for having me.

By Sally K Lehman