Approach to Supplementation
I enjoyed reading this discussion thread and seeing the many different perspectives, all of them well reasoned. Without going into detail, my approach is more one of broad base than minimalist, with one important difference.
Model
With the exception of rapamycin (which I’m putting on hold for the moment) and metformin, my model is to take low doses of substances at levels that are generally higher that might be obtained from a diet but lower than is generally recommended as a healthspan enhancer. I take these supplements indefinitely until a consensus research trend suggests otherwise.
As an example, I supplement my intake of most of the well researched polyphenols (all types) at doses close to what I might get from my diet but only with great diligence and not for all of them combined. I used to take megadoses of some supplements but have backed off in recent years. I used to take 2,000 mg./day vitamin-C, sometimes more for example, but now take 1,000 mg. or ess per day.
The one exception to my general decision-rule is supplementing where I believe or know I have an outsize risk. I take a specialized prostate supplement for BPH, and DIM for E2 balance.
Evidence
Having taken supplements diligently for more than 50 years, I have no way of being sure that they are helping me. An article of faith is involved beyond the research. I have two observations that support inferring that they are beneficial in an outsize way. On a few occasions, I have not taken supplements for more than the day or two break I have always taken one or twice a month. On those occasions, my overall energy and sense of well being drops significantly. However, the explanation could be more one of habituation than long term benefit. Placebo is a possibility as well but much less likely over these decades. Also, I am a low placebo reactor. The second reason for inferring benefit might be stronger. The men in my family line from grandfather, to father, to uncles, to cousins (maybe 10 in all), have a remarkably consistent profile of health, fitness, energy and wellness. So far, my profile is at least a standard deviation and maybe two above the others. Everyone in the family has observed this exception.
Sources
My go-to sources for guidance are the NIH and other research stacks, Life Extension researchers, Kamal at Examine.com, and ConsumerLab.com. Increasingly, I’m finding the three AI LLMs I use to be a useful time saver. It is fairly easy to work around their shortcomings once you get the hang of it. More than once, an LLM has dug out an inference I missed due to its strength of considering all available sources more-or-less simultaneously.