In other words, what happens when a population suddenly stops taking fluoride in their drinking water, like Juneau's citizenry did?
Now, thanks to a recent study led by first author and public health researcher Jennifer Meyer from the University of Alaska Anchorage, we've got new insights into the subsequent effects.
In the study, Meyer assessed Medicaid dental claim billing records for two groups of children and adolescents aged 18 or under.
One of these groups represented what the researchers call "optimal" community water fluoridation (CWF) exposure: 853 non-adult patients on behalf of whom Medicaid dental claims were filed in 2003, years before the fluoride cessation began in 2007.
The other group was made up of 1,052 non-adult patients from families who similarly met Medicaid income requirements, and who made the same kind of dental claims almost a decade later, in 2012.
[...] "By taking the fluoride out of the water supply... the trade-off for that is children are going to experience one additional caries procedure per year, at a ballpark (cost) of US$300 more per child," Meyer explained to KTOO News.
Reference: Jennifer Meyer, Vasileios Margaritis & Aaron Mendelsohn, Consequences of community water fluoridation cessation for Medicaid-eligible children and adolescents in Juneau, Alaska, BMC Oral Health, https://doi.org/10.1186/s12903-018-0684-2
(Score: 3, Informative) by FatPhil on Monday December 16 2019, @09:59AM (1 child)
> Just like no one would only look at 2003 vs 2012 instead of all the data. If they didn't get the result they wanted there they would have paid for more years, etc.
You cannot say that with any certainty. They only looked at one before/after pair, and the data from those years was clear. They did not p-hack, as they did not look at any other years.
It is you who does not understand p-hacking. You have to throw away data that's inconvenient in order to p-hack, they did no such thing. And even if they'd bought 3 years of data from each side of the change, which they didn't, and thrown away the least convenient 2 of each, which they didn't, their p-values would still have been significant (e.g. p<0.004, and p<0.001).
You're making a claim of scientific fraud - that's a serious claim - you need serious evidence to back that up, and you have none (as there is none).
Great minds discuss ideas; average minds discuss events; small minds discuss people; the smallest discuss themselves
(Score: 0) by Anonymous Coward on Monday December 16 2019, @10:08AM
That is wrong, look it up.