by Steve Milloy at junkscience.com
The new study the Consumer Product Safety Commission wants to rely on to ban gas stoves is classic junk science.
- It’s not actual research on children. It is a meta-analysis of previously published (and ignored) studies — a study of otherwise unpersuasive studies. The authors did a literature search for previous epidemiologic studies on gas stoves and asthma in kids and then just mixed those results together in an effort to contrive statistical significance. This is a bogus technique for a number of reasons including publications bias in the component studies — i.e., studies with null results aren’t published.
- The study results, including the component studies, are weak statistical associations — i.e., noise range correlations. The study results, likely including the component studies, are not statistically significant either.
- Asthma is an allergic disease. There are no allergens in natural gas. So the study has no biological plausibility. No one knows what causes asthma in children and so competing causes could not be ruled out.
- The claim that gas stoves are responsible for 12% of childhood asthma – an epidemiologic concept called “attributable risk” – is entirely bogus because epidemiological studies can only be used to associate exposures with disease. They cannot be used to determine risk of disease because (1) the underlying data is not representative of the population; and (2) epidemiologic studies are just statistics (i.e., correlation is not causation) and cannot be used by themselves to determine cause-and-effect relationships.
If none of that means anything to you. you need to read “Junk Science Judo: Self-defense Against health Scare and Scams.”