Ultra-low-carb eating has appeared for about a century already to be the most effective treatment for both overweight and diabetes (both linked to the more general metabolic syndrome), beating all drugs and other interventions by wide margins (on this, see Taubes’ groundbreaking Good Calories, Bad Calories). However, the establishment has resisted or ignored (if not memory-holed) this information for general application primarily based on separate claims about lipid profile risk and heart disease. The conventional view has been that these risks outweigh the benefits, so other treatments are likely to be better on balance.
Turning this view completely on its head, in this 1 July 2014 conference presentation, cholesterol researcher Jeff Volek explains how his and related carefully controlled research over the past two decades indicates that ultra-low-carb eating appears to also be the best known intervention for improving lipid profile markers, properly interpreted. He focuses particularly on issues with the measurement, context, and interpretation of LDL-C, and by the end appears to have left the strong conventional view concerning low-carb and lipid profiles in the dustbin of failed scientific claims.
[Gets going at about 1.05.]
See my Evolutionary Health page for more perspective and selected references.