Guest post by David Middleton
This article and the underlying “study” can only be explained by a childhood diet rich in lead paint chips…
Study finds pregnant women who live near fracking areas more likely to have underweight babies
[…]
While scientists are unsure whether the pollution is coming from air, water, onsite chemicals or increased traffic, they say the results prove “hydraulic fracturing does have an impact on our health.”
They are now looking forward to investigating the source of the pollution and challenging lawmakers consider the dangers on health.
[…]
They can’t identify a fracking-related pollutant or its source… Neither could Obama’s EPA. Yet, they conclusively prove that fracking kills babies (hyperbole intentional).
The “paper“* is mind-bogglingly idiot.
Table 1 explores differences in maternal characteristics, infant characteristics, and health outcomes between mothers who were potentially exposed to fracturing and those who were not. The first two columns show variable means for mothers whose residences were less than 1 km from a location (or multiple locations) that fractured. Columns (3) and (4) report the means for births to mothers who live within 3 to 15 km of a well location. These samples are further divided into those whose infants were born before the spud date (that is, the commencement of drilling)—thus, never exposed to fracking—and those whose babies were born after the spud date. When the mother is within 1 km of multiple locations, we use the earliest spud date to align with the approach used in the regression analysis.
The define fracking as “after the spud date” of a well that was frac’ed during the completion process. This is Gasland-stupid and/or Gasland-dishonest.
This is their “evidence”:
There is absolutely no meaningful trend. Just a spike on the sample closest to “fracking,” which is also, by far, the smallest sample.
How does crap like this pass peer review and get published? This was actually published by the American Association for the Advancement of Science. The only “journal” worthy of crap like this is:
[*]Currie, J. et al., Hydraulic fracturing and infant health: New evidence from Pennsylvania. Science Advances 13 Dec 2017: Vol. 3, no. 12, e1603021
DOI: 10.1126/sciadv.1603021
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
That assertion is utterly ludicrous. Oh, by the way, I saw my grandmother leap over the moon last night…
Does that make you 1/4 bovine?
I believe they would see more correlation if they checked birth weight against proximity to a school.
But, but, but, isn’t fracking is a major cause of erectile dysfunction and women faking orgasms…?
Love those confidence intervals, particularly for the first km.
Regarding: “Just a spike on the sample closest to “fracking,” which is also, by far, the smallest sample.” It’s more ridiculous than that. The spike is not on the closest sample, but the second-closest one.
The spike, the only anomalous deflection, is the far left sample…
Take a look at the graph again. The leftmost sample is the one under the low end to the left of the spike.
The left-most sample (0-1 km) is the spike. It is the only point that the infant health index drops below -0.03…
All of the other points oscillate around 0.00.
Oops, my bad. The low end is a downward anomaly and the main anomaly. I was looking at the peak over the second sample when I was looking for a spike. The peak over the second sample corresponds to babies being probably slightly healthier than average.
That’s OK. The paper is extremely convoluted in its torture of data.
Well it does make sense that there are almost no samples right on top of a well bore. And the authors concede that the effects are highly local (0-1km) from the well bore.
“We find the largest effects for mothers living within 1 km of a site—a 25% increase in the probability of a low–birth weight birth (<2500 g) and significant declines in average birth weight, as well as in an index of infant health. There are also reductions in infant health for mothers living within 1 to 3 km of a fracking site, but the estimates are about one-third to one-half of the size of those for mothers within the 0- to 1-km band. There is little evidence of health effects at further distances, suggesting that health impacts are highly local."
http://advances.sciencemag.org/content/3/12/e1603021.full
Links to the full paper were twice provided in the post.
They demonstrated no statistically meaningful relationship between fracking and infant health. They didn’t even correlate infant health with fracking operations.
They defined fracking as anything after the spud date of a well.
This is moronically meaningless…
For example, Fig. 3 shows that the coefficient on the indicator for maternal residence within 1 km of a site is approximately 0.01, indicating a 0.01 percentage point increase in the probability of low birth weight relative to people who live 15 km or more away from a site. The effect of living 1 to 2 km from a site is near zero, but the effect of living 2 to 3 km from a site again appears to be positive.
Figure 3…
http://advances.sciencemag.org/content/advances/3/12/e1603021/F3.medium.gif
A 25% increase in a vanishingly small number is still a vanishingly small number.
Regardless,
Correlation does not equal causality.
I grew up in rural PA during the summers. Tailings from open pit and other coal mining could just as easily be the reason — if they tend to drill fracking wells near old coal mine sites.
They don’t “tend to drill “fracking wells” near old coal mine sites.”
Black triangles = coal mines
Blue-gray circles (so dense, looks like a blov) = gas wells
https://www.eia.gov/state/?sid=PA
There’s no such thing as a “fracking well.” Fracking is a well completion procedure used on most oil & gas wells.
This should really be displayed with the observed points scattered around and the error bars centered on the null hypothesis. I expect that the “results” are totally unreproduceable.
Yes, the high confidence interval on the first point indicates that there is no significant variation even for the first point, so there is no spike. To say there is a spike is a variation on the data miner’s salvation- those ever present “non-significant results”.
Thanks for all your insights David. Should be mind-bogglingly “idiotic” I’d guess.
Could this incidence of underweight babies have been caused by all those young strapping frackers coming to town?
This turned out to be the case in Cumbria’s leukemia cluster … and the story is also a good example of the pub(l)ic science process.
http://www.telegraph.co.uk/technology/3346407/Did-Sellafield-workers-seed-leukaemia.html
[PS, has anybody done a study on the correlation between Alien Abductions and the incidence of extra-marital affairs?]
“PS, has anybody done a study on the correlation between Alien Abductions and the incidence of extra-marital affairs?”
No I havent, but I heard that you can lead a horse to water, but it takes lead for every pencil.
Tony how about extra-martian affairs?
Well that was interesting…..Washington and Bradford counties in Penn have the highest number of fracking…and also high poverty and drug use
Indeed, and malnutrition and drug use do result in low birth-weight babies….
PMK
pameladragon – December 18, 2017 at 4:00 pm
To be more precise, ……. drug addiction by the parent(s) result in their birthed children being: (1) low birth-weight; ……. (2) learning disabled; ……. (3) physically deformed/handicapped; …… and/or (4) stunted growth and low body mass due to neglect and malnutrition during their “formative” pre-teen years.
The above stated facts can be verified by checking US Public School Records for the numbers of “learning disabled” and “birth-handicapped” children being enrolled as students each and very year. The numbers of the aforesaid students started increasing exponentially after the 1970’s drug usage escalated and the “users” started birthing children.
The learning disabled/birth-handicapped problem has become so great that every Public School System in the US of A now employees at least one (1) Special Education Teacher, something that was never heard of or needed prior to the “Wacky Tobacca Revolution”.
So drug use causes fracking.
I always wondered about that.
These people seem to imply if you can find a well where the babies health is better within 1k then that would prove fracking is good for babies.
Assuming that the effect is real, you have a second problem. People living that close to a wellhead aren’t representative. They are much more likely to be poorer and rural, This is the same issue as people living right next to industrial districts, railroads, and high power lines. It’s a proxy for wealth. The only issue is that it’s a much poorer proxy than other options, leading to a much weaker effect.
This is yet another study that, after you cut out the preconceived answer, boils down to that it stinks to be poor.
Given the things the “study” did not control for, one could as easily claim the birthweight anomaly was caused by noise pollution from the rig. Or the mother’s exposure to meth. Or opiates. Or just poverty in general.
I don’t see why fracturing would have an impact. The truck traffic and rig noise could be stressful, especially if they are used to a fairly low noise environment. The data doesn’t show much of a trend, but I do feel companies need to do a better job studying these issues ahead of time.
Many years ago I had a team leader in my group go out and spot a well pad about 100 meters from a ranch house whose owner was a very nice old man, who owned thousands of acres. I heard about it from the field manager who felt the site was going to cause bad relations with the man, so I got in my car with a surveyor, a drilling engineer, and the surface construction team leader, visited the site, and found a perfectly good location about 400 meters away, which required a bit more road but made the old man really happy. After that I had to give a pep talk to the whole office about improving the way they related to landowners, tenants, and others in the area, watch out for better ways to route truck traffic etc.
I’m explaining this because even though the study looks bogus, the oil industry does have yahoos who need straightening out so they have better manners.
+1000
fernando
” ,. . . . the oil industry does have yahoos who need straightening out so they have better manners.”
The oil industry is far from unique.
Very far.
Auto
fernandoleanme – December 18, 2017 at 7:51 am
Fernand, given the fact you are a self-defined expert on things associated with activities involving the “oil/gas” exploration/drilling etc., ……please tell everyone just “how many days” a homeowner would have to tolerate truck traffic and drilling rig noise before the well was “shut-in” and all noisy equipment moved off-site.
And ps, Fernando, …… when my mother was pregnant with me, an extremely loud noise scared the bejesus out of her, …….causing her to drop several 78 RPM phonograph records on the hardwood floor, ….. but that incident never affected me, ….. affected me, ….. affected me, ….. affected me.
Cheers, Eritas
Notice that there is no reason to expect any correlation here on births. Claiming causality (correlations) strictly on the basis of sample correlations is a no-no. One can ALWAYS
find a correlation that is “statistically significnat’ (which often is actually insignifcant in the effects)
is one looks at enough dependent variables. For example, these folks could have selected all manner of variates and calculated a correlation for each – number of headaches, cancer incidence,
lifespan, shoe size, height, weight, chloresterol, you name it. Do this with enough variates are you are pretty much guaranteed that you will find a fracking-distance correlation that is “statistically significant” at, say the .05 level. The proper next step is to cross validate that correlation using a new, different set of data points. The correlation will then disappear. This is a perfect example of
violation of the science rule that the researcher must present a theory of a suspected correlation
stating the reason for the correlation, BEFORE doing any sampling, and calculating correlation coefficients. In this particular case, the data doesn’t even support a linear correlation and should
never have been published by any science journal just on that basis alone.
Fracking is a lightning rod that attracts junk science research like this.
arthur
“Fracking is a lightning rod that attracts junk science research like this.”
“Fracking is a lightning rod that attracts junk research like this.”
Fixed?
Auto
The fact this gets published tells us all we need to know about the status of scientific endeavor today. There are many skilled, reputable scientists who know how to do research and who publish honestly but they apear to be vastly outnumbered by incompetent, dishonest opportunists.
This paper was thoroughly destroyed the moment it became public, but that won’t stop the headlines. Or maybe it’s true pregnant women should stay between 1-2 km away from fracking operations at all times if they want the healthiest baby possible.
Yes it ignores the look-elsewhere-effect but that doesn’t seem to stop anything in Pseudosciences.
Or could the effect possibly be caused by stress induced by fracking protesters occupying the the vicinity of the fracking site.
How does this cr@p get published?
Because it fits the current ecoloon paradigm that fracking is evil.
No. propagandists, long ago, figured out they can swing undecideds to their message with THE “it’s for the Children” ploy.. If you favor fracking you hate children. It has become a Hail Mary strategy in the CAGW propaganda movement.
If you believe Currie et al’s Figure 4, the healthiest babies are born between 1 and 2 km of a fracking well.
So let’s draw a ring between 1 at 2 km of every fracking well in the USA, and build maternity wards in every ring. Just don’t get too close…
The cite says J. Currie. Is that defamatory against Judith Curry?
Pretty simple really, most locations where fracking occurs are low-income rural areas, and the incidence of problem / premature births is extremely high. I don’t have the Census data handy, it’s in the micro-data, and I don’t have time to roll it up, cluster by cluster, but just take my word for it.
In regard to the authors, Currie in particular has some really odd-ball ideas about public health, or health in general, and seems to completely divorce it from any personal action. To her it’s all about economics, and she’s and endless shill for a nanny-state mentality
Elevation does have an effect, elevation of wells in Colorado would cause proximity to correlate with smaller birth weight, but not be the cause.
Maybe to much pop and chips eaten by pregnant mothers? See it all the time in the grocery stores after the baby is born, don’t have enough money for all the groceries, so put the pablum and diapers back and keep the pop and chips, not to mention keeping the cigarettes too. Yes the father is there too, someone has to drive the new pickup truck.
http://advances.sciencemag.org/content/3/12/e1603021.full
Now we need a graph for the prevalence of neck injuries caused by head shaking due to retarded studies.
Fracking babies? … Seriously?! … This is supposed to be evidence? … [Ahhh!, I just pulled a neck muscle!]
Robert,
From this [Eastern] side of the Ditch – I have literally no idea of what the map represents.
Apparently it is an are with – taking logarithmic mid-points of birth estimates – with something like 100 000 – 150 000 births per year. The UK has about 700 000 births per year, so this would seem to cover a total population of – roughly 10-20 million [with some approximations made in my head, making those bounds definitely not hard bounds].
I have, by inspection, discounted all the lower 48 – as states.
But – is it a county, or some other bundled area.
Is it some part of a state – Oklahoma? Texas?
There has been much fracking, so I guess not California.
[Mods – /Sarc for the last comment, just in case you missed it!
Is California still anti hydropower, still, too?].
Auto
You are looking at the State of Pennsylvania.
The area in the southeast corner with the highest birth rate is the region around Philadelphia. The area in the southwest corner is the region around Pittsburg.
SMC
Thanks.
I didn’t pick that up at all.
Much appreciated.
Auto
Auto,
The fractured wells represented in the map are in the Marcellus Formation. The Marcellus is the largest shale gas reservoir in the USA. Here is a primer from Wikipedia…
https://en.wikipedia.org/wiki/Marcellus_Formation
In general, the wells are in sparsely populated areas.
American Association for the Advancement of Science needs to change it’s name to the American Association for the Advancement of Propaganda.
The AASS jumped the shark some decades ago when they started accepting parapsychology as science.
American Association for the Advancement of Stupidity … [no letter changes required]
The 80 20 rule does not work on campuses. The 90 10 rule must be applied in their case because the checks and balances of society do not apply there.
This explains their applications as trade protectionist researchers also…
https://www.eurekalert.org/pub_releases/2017-12/cul-ro121417.php
Maybe undersized babies will make up for all the growth hormones in our food that snowflakes are wetting their pants over.
You can find a lot more correlations like this one at:
http://www.tylervigen.com/spurious-correlations
A sample:
US spending on science, space, and technology correlates with Suicides by hanging, strangulation and suffocation
Correlation: 99.79% (r=0.99789126)
Number of people who drowned by falling into a pool correlates with Films Nicolas Cage appeared in
Correlation: 66.6% (r=0.666004)
Per capita cheese consumption correlates with Number of people who died by becoming tangled in their bedsheets
Correlation: 94.71% (r=0.947091)
Divorce rate in Maine correlates with Per capita consumption of margarine
Correlation: 99.26% (r=0.992558)
Age of Miss America correlates with Murders by steam, hot vapours and hot objects
Correlation: 87.01% (r=0.870127)
Total revenue generated by arcades correlates with Computer science doctorates awarded in the US
Correlation: 98.51% (r=0.985065)
+1
Let’s look at the good side of this. It reduces the number of infants who are “obese!”
The group second closest to a well is also heaviest. Two data points? Worth less than a bucket of warm spit.
David
In New Zealand they have science presentations in pubs as part of greater communication.
Here is a link to a related in womb effect on babies. I found it most interesting, the presenter is very engaging.
All is not what it appears.
Regards
https://www.radionz.co.nz/stories/2018624979/raising-the-bar-obesity-in-the-womb
I knew it! I used to be 6 feet 2 inches tall and now I’m only 6 feet 1.5 inches!
Fracking is what done it !!!!!!!!!!!
As with many things, the name signals the opposite in fact. AAAS diminishes science by publishing this dreck. Obama’s ACA raised health insurance premiums. The ‘pink slime’ kerfuffle was about lean finely textured beef. ANTFA uses facist tactics identical to Brownshirts in Germany in the 1930s. And so on.
I subscribed to AAAS journal Science from the 60’s and off and on into the early 2000’s. But I gave it up when it became a leftist warmist alarmist magazine.