Here's a pretty good meta-analysis: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3970209/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3970209/ There's a lot

Here's a pretty good meta-analysis: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3970209/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3970209/ There's a lot of conflicting data. However I think the general thrust is that you should not try to lose weight unless you are accurately measuring how much lean mass you are losing. Otherwise it's very easy to lose a lot of muscle, which is probably not healthy unless you are extremely obese. Crucially your current body composition matters quite a lot. People under 10% body fat are going to start losing really important vital body mass. People over 25% body fat will lose mostly fat. But people have wildly different body compositions and a BMI of 25 doesn't mean 25% body fat. Again, accurate measurement is crucial. Personally, I focus on maximizing a variety of physical measurements. (How long I can hold various yoga poses without discomfort, how many situps I can do, how far I can bike/walk/swim before I begin to feel fatigued, etc.) I definitely think optimizing weight downward is a poor strategy for improved health.