News

Back in November, EID detailed why a study claiming fracking is causing a rise in premature births in shale regions of Pennsylvania actually fell spectacularly short of supporting that claim. Now, a top-tier biostatistics analysis expert has issued a sharp criticism of that study, offering even more insight into the numerous flaws the EID debunk pointed out.

Dr. Tony Cox — who is a clinical professor of biostatistics and informatics at the University of Colorado-Boulder, as well as President of Cox Associates, a Denver-based applied research company specializing in quantitative health risk analysis, causal modeling, advanced analytics and operations research — has penned a letter to the editor in the journal Epidemiology that essentially confirms that the study’s claims lack merit.

Cox takes particular issue with the authors’ claim that “This study adds to limited evidence that unconventional natural gas development adversely affects birth outcomes,” pointing out the study’s methodology was completely inappropriate to draw causal conclusions:

“First, ‘unconventional natural gas development adversely affects birth outcomes’ is an unwarranted causal interpretation of associational results: in general, ‘the associational or regression approach to inferring causal relations — on the basis of adjustment with observable confounders — is unreliable in many settings” (2). Without explicit causal analysis (e.g. Granger causality tests, causal graph models), claiming that associations provide evidence for a causal conclusion is unjustified.”

EID highlighted in its original debunk that the researchers also failed to analyze baseline data (despite the fact that it was readily available) and took no environmental samples. Instead, the authors chose to use modeling, which might have been somewhat acceptable — had they not botched their modeling completely, in Cox’s estimation:

“Exposure metrics were not validated and estimation errors were ignored. The authors constructed surrogate exposure metrics using well depths and distance, ignoring relevant geology and whether homes are up- or down-gradient from the well. View Full Article