Re:Low-Salt Diets May Pose Health Risks, Study Finds
by Swizec on Monday August 18, 2014 @01:56AM
(#8191560)
Scientists do in fact report how confident they are in their findings. Read original sources, most of them say things like "in X% of observed cases there was a Y% of Z over the course of T with confidence level of P".
But that sells a lot less clicks than "SCIENCE SAYS LOW SALT IS BAD FOR YOU OMG"
Re:Low-Salt Diets May Pose Health Risks, Study Finds
by colechristensen on Monday August 18, 2014 @02:32AM
(#8191613)
The truth of the matter is that the general public shouldn't be reading scientific publications, because without the context of the intended audience, reasoning about what you read is incredibly difficult or impossible. This is doubly true for 'science journalists' who so very rarely are educated to be able to correctly translate and very often distill a paper into an incorrect and baiting article with only the tiniest shreds of truth.
The confidence rating you mention has little value to the general public, and even among scientists, the value of 'p-value' is questionable and the matter of occasional debate.
http://annals.org/article.aspx?articleid=712762#ThePValueFal...
Re:Low-Salt Diets May Pose Health Risks, Study Finds
by Semaphor on Monday August 18, 2014 @05:01AM
(#8191900)
examine.com does that, they list level of confidence (A - "Robust research conducted with repeated double blind clinical trials" to D - "Uncontrolled or observational studies only") with a percentage value for consensus and a rating of the magnitude of the effects.
Independent body who runs a well-controlled double blinded study? A+
Biased organization who runs an analysis of questionable observational data with barely passible statistical significance? F
They already do this for cancer therapies. NCCN gives rating that describe how much evidence exists for a given therapy.