It might just be marketing hype but the Samsung watch claims to be 97% as accurate as a DEXA scan for measuring body fat, etc. I use one (mine is a couple of years old now) and it seems pretty accurate to me.
1 Like
Looks like attenuating insulin resistance is another benefit of herring.
But the effect is greater with cod.
Findings indicate that 2.5 g/day of proteins from fish or milk may be sufficient to improve glucose regulation in overweight adults. The effects were most pronounced after supplementation with proteins from cod, herring and milk, whereas salmon protein did not affect any of the measurements related to glucose regulation.
Dose is low. 2.5 grams of protein. That is a small portion of bacalhau.
Below is a separate study on cod.
2 Likes
The paper is not compatible with any comprehensible “97% accuracy” claim:
"Significant fat-free mass (FFM) differences (P < 0.05) were observed between both W-BIA and 8-BIA when compared to DXA, though the systematic biases to the criterion were correctable. No significant difference was observed between the W-BIA and the laboratory-grade BIA technology for FFM (55.3 ± 14.5 kg for W-BIA versus 56.0 ± 13.8 kg for 8-BIA; P > 0.05; Lin’s concordance correlation coefficient = 0.97). FFM was less precise on the watches than DXA {CV, 0.7% [root mean square error (RMSE) = 0.4 kg] versus 1.3% (RMSE = 0.7 kg) for W-BIA}, requiring more repeat measures to equal the same confidence in body composition changes over time as DXA.
Specifically, the watch underestimated FFM by 3.4 kg, overestimated FM by 3.2 kg, and overestimated skeletal muscle mass by 2.2 kg on average (out of 27.2 ± 8.1 kg SMM on DEXA).
https://ajcn.nutrition.org/article/S0002-9165(23)03666-3/fulltext
2 Likes
I agree that more measurements are required with the watch to gain confidence in the readings. I usually do it daily. But you can buy the watch for about the same cost as a single DEXA scan in most cities. I’m happy with it but clearly YMMV.
1 Like