This link is for those who want to reduce dieting to "calories in < calories out". The linked paper shows that the source of the calories affects both sides of the equation (or the inequality if there is a net shift).
The thermogenesis effect is relatively small, but it is combined with naturally reduced calorie consumption on an otherwise unrestricted high protein diet, which is attributed to satiety, i.e., feeling full [0].
[0] Skov AR, Toubro S, Ronn B, Holm L, Astrup A: Randomized trial on protein vs carbohydrate in ad libitum fat reduced diet for the treatment of obesity. Int J Obesity 23:528–536, 1999
Had to look up where this bit from the abstract...
> Body temperature was slightly higher on the high protein diet (p 0.08 after the dinner meal).
...was discussed in the article:
> At 2.5 hours after the dinner meal, the change in body temperature for the HP diet was nearly 40% higher than that for the HC diet (0.8 and 0.5 °F respectively, p = 0.08).
So, for n=10, they observed a 0.3 degree F difference in the rise in body temperature with a p value of 0.08? Maybe there is something about about bio-sadistics I am failing to grok, but this seems like a pretty tiny effect size, with a relaxed standard of statistical significance, measured in an awfully small sample.
On the contrary , to me it's good to see a relaxed p-value cutoff being reported, as opposed to self censorship for being barely on the wrong side of the completely arbitrary 5% cutoff. Occuring in the 8% tail is still unlikely to occur by chance, enough to call for replication with a bigger sample.
