User:Madprime/How do gendered algorithms relate to my "real" body composition?
My Withings scale estimates my body composition (percent fat, muscle, etc.) -- but I wonder whether the accuracy of this has drifted in past years. I identify as non-binary (trans FtX) and since fall 2020 I've been taking weekly testosterone. After a couple years targeting cis-male levels, the physiological effects include things like fat redistribution and increased muscle growth.
I'm far from fit: I've put on weight during the pandemic (BMI over 27), and don't routinely exercise. Regardless of which version -- male or female -- is "more correct", I know my "fat composition" is higher than ideal! I'd like to be losing weight and exercising more. Nevertheless, every time I step on the scale, I wonder about those composition numbers and wonder about what the "real" number is, and whether the "female" vs. "male" algorithms are more accurate.
Goals / Questions
- Try to determine "real" body composition. How does this compare to estimates?
- Try other methods for estimating body composition (beyond my scale), ask the same question.
- Repeat after time & change (weight loss and exercise). Does it affect the answers to the above?
Methods to measure or estimate body composition
Writing up some ideas here...
Hydrostatic weighing
Hydrostatic weighing is considered one of the most accurate ways of determining percentage body fat. One limitation might be that it might still be gendered! Wikipedia's article on hydrostatic weighing describes using gender to estimate residual lung volume, which is necessary. (Residual lung volume might also be determined more directly e.g. via gas dilution.)
TODO: there are a number of services for hydrostatic weighing in the area. I should reach out to them, see how doable this is, and maybe see whether they require gender as part of coming up with their assessment. It would be nice to wholly avoid this... on the other hand, it seems unlikely that my lungs have changed volume. (I might decide the "female" algorithm is probably appropriate in this case, unlikely to have been affected by physiological changes due to masculinizing hormones.) Also, I don't know whether there's much variance here; maybe I should try using more than one service.