[dahart]

This is an insanely valid point, and I didn’t realize til I just looked it up how many people have high blood pressure. But isn’t the problem with medically approved science assuming that any given metric or recommendation works for everyone? This happens on both ends; apps marketing advice based on some amount of science without specifying or understanding who it does and doesn’t apply to, and us consumers taking the “it’s science” to mean it’s supposed to work for us, and that if it doesn’t it means the science was wrong. Scientists and papers and doctors definitely do not agree on everything dietary, and any given scientist or doctor will give you different dietary advice if you have high vs low blood pressure. (But I guess it’s good to look at the things that all doctors and scientists do agree on.) Lowering saturated fat and meat intake and increasing fiber intake actually is generally good advice for a lot of people, especially if they’re overdoing it, but definitely not everyone and most especially not everyone with specific health issues. I’m a little surprised by the comment that advice to eat low carbs feels a generation away to you. I can’t think of specific apps, I guess, the only dietary apps I’ve used track macros and calories, and don’t make recommendations. But, practically everyone I know is highly carb aware and many have tried the Atkins diet for weight loss and/or high protein for workouts, with varying levels of success. It’s gone far enough that I’m starting to hear people talk about how we’re eating way too much protein, and that people misunderstand carbs (they build muscle as well). While that might be true for many, it’s definitely not considering high blood pressure, nor, for example, celiac or Crohn’s disease. I guess we always need to remember, both when making apps and when buying them, that whether the correlation a scientist found in a study works for someone may depend entirely on how closely this person matches the study cohort, and we have to remember that assuming we can project correlations found in a cohort to a wider audience has historically been one of the best ways to get science wrong, right?