Bryan Johnson, founder of Braintree and Kernel recently (a year ago) embarked in what seems one of the most (or perhaps, the most) ambitious program to slow down (and attempt to reverse) aging: Blueprint.

On its face, Blueprint seems like a prescription for good health and longevity: diet, exercise, and supplements. Hence many have focused on the specific supplements that Johnson is taking. Andrew Steele, author of the primer to the longevity field Ageless did a video going over the key points of Blueprint, pointing that the evidence base behind some of the supplements he takes (like spermidine), while promising, is not super robust yet. Others have started to copy Blueprint as is. The whole thing is open source, in the Blueprint website one can see not only what Johnson eats and how he works out but also what his biomarkers are and how they have changed over time. One could quibble with whether this iteration of Blueprint is optimal or whether some of the supplements or experiments he is doing are not really doing that much, but the multiple measurements he has taken (here for the latest update) do point in the right direction (skip to here for a summary).

Some on the internet have been getting mad at Johnson, either criticising the cost of the program or just getting mad at the prospect of someone trying so hard, and perhaps being too impatient; Bryan Johnson aims to reverse his biological age substantially but as of today he does not look under 40 (here, here).

I think everyone should be greatly thankful for Blueprint: He's spending all that money and giving away his results and most of the methods for free to everyone, and he is in fact very healthy for his age.

Even the point that Steele and others have made that some of the interventions he is trying are not backend with publicly available robust evidence is not a very strong critique: The point of Blueprint is to experiment and generate that very same data. Johnson might be doing some weird looking light therapy that he mentions in some of the videos as trying, but he's not committed to that; if his skin doesn't get bette I'd guess I'd drop it.

A better critique of Blueprint would be to criticise the way he is measuring his health, but I haven't found many critiques of this sort. One could criticise the use of epigenetic clocks on their own, but the rest of the measurements have gone undisputed.

Less of a critique, but more of a request would be to know how exactly the loop is closed between measurement and action. For example in my own case I do blood tests periodically. Recently my iron was too high, so now I know I need to reduce my intake of iron or add calcium or curcumin to my diet (that reduce iron absorption). That's an easy one because the metabolism of iron is well understood that the lever to change that variable is very clear. The same is true if my iron levels were low. But for supplements like BroccoMax (sulphorephane), or garlic, what's the right variable to measure?

Working in Blueprint's advantage is the fact that it asks for a fixed daily routine: It's the same exercise and the same diet. Holding this much constant probably helps with the ongoing self-experimentation. However, it is still tremendously challenging to optimize a supplement regime given all the variables in play: which supplement, when to take it, and at what dose.

Part of the current iteration of Blueprint is a set of known anti-aging prescription drugs (rapamycin, acarbose, metformin; and iirc in one of the videos he mentions a non-feminizing estrogen he might be adding, this is probably 17-a-estradiol). This is the set of drugs that have been shown to work in the gold standard mouse lifespan study Interventions Testing Program from the NIA that I summarized here in my Longevity FAQ.

Over at the rapamycin forum (yes there is one such forum) some wonder if the unarguable success of Blueprint is mostly because of rapamycin (the most promising anti-aging drug we have so far), and one person proposed a cheaper version of Blueprint.

This all misses the point to some extent: If one reads the website and watches the videos and listen to the podcasts, Blueprint is not a specific set of foods to eat or exercises to do. What Blueprint actually is is explicitly stated at the beginning of the Blueprint website: it's a philosophy of self-improvement by restricting decisionmaking and empowering our best selves to make those choices, coupled with letting AI (or experts) make those choices for us. It is not telling us to eat that exact amounts of those foods and take those exact supplements, Blueprint is supposed to be personalized.

The usual go-to example to explain the motivating philosophy of minimizing choice he uses is food cravings at night, where he would order food he would then regret eating. So he decided to not make that decision in that context. It's the same as not walking into a supermarket while hungry.

Blueprint for all?

Copying Blueprint would mean having the same set of things measured and having a team formulating custom diets, supplements, and workouts. This is very expensive, and the reason why Johnson spends so much on Blueprint.

But one can probably get most of the benefits of Blueprint from a simpler set of interventions. In fact even if someone were to do Blueprint, one should do the simple interventions first and establish a baseline to see if later interventions also work.

This set of interventions being good diet, sleep and exercise. The goal of this mini-Blueprint would be something that can be measured simply: feeling well rested, and having a defined musculature (a combination of the right amount lean body mass and low body fat). Johnson's is at ~7% body fat but even 15% is already excellent and nontrivial to achieve for men. The way to get there is caloric restriction coupled with exercise and enough protein. The usual recommendation I've seen for protein is 2 gr per kg of lean body mass. Johnson takes 93/(160.7lbs~72kg)=1.28. Each person will need more or less.

I generated recipes for Super Veggie, Nutty Pudding, and one of the third meals (The Orange Fennel salad) described in Blueprint and that mix delivers basically the calories stated in the website (~1750kcal). The intake of some micronutrients is too low (Vitamins B12, D), but after adding the relevant supplements for those delivers a mostly balanced nutritional profile (Of note is 5x the amount of vitamin C RDA). The diet seems to contain a lot of iron (2.78x RDA) from the Super Veggie. Absorption will depend on what that iron is eaten with.

image-20230127141342282

An example Blueprint day

The cost of this is around $60 a day. If one wanted to go super cheap, perhaps one could try Mealsquares (~$16 a day). Mealsquares looks better nutritionally (and is easier to cook) but has a higher sodium level. The Mealsquares website justifies this by pointing to some research that is now known to be flawed. Huel ($13 a day) could also be an option, but like the Blueprint diet it's a bit higher in iron.

What about something simpler? Suppose you eat yogurt, berries, a handful of walnuts and pecans, protein powder, olive oil, and some vitamin supplements. That can also hit the diet's recommendations. My own diet is a mix of this and the huel/mealsquares approach but I'm nowhere near as strict when eating out.

Out of these diets, what is best? Probably depends on the person. That's what the tests are for!

Something I'd like here is some explanation of why the ingredients are what they are. Why cauliflower when Super Veggie already has broccoli? Why so much garlic (1 clove in Super Veggie + 2x2.4g in supplement + 2x1.2g of another garlic supplement)? I get why garlic is good, but why those amounts. Same for other supplements like NAC (2x1800mg through the day)

Measuring success

There are many, many measurements in the Blueprint suite. Some I think don't add much value (PBMC telomere length, or blood triglycerides). What do I think adds the most value?

  • Body fat (better than weight directly) is easy to measure (calipers, BIA, or DEXA) with methods generally available.

  • apoB (You could do a full lipid panel, or measure non-HDL cholesterol but measuring apoB directly is the best way to get at CVD risk directly)

  • hsCRP for inflammation

  • HbA1c (for long term blood glucose levels)

  • CBC

  • CMP

Two websites to get tests are Labtestingapi.com (Through Quest) and OwnYourLabs (through LabCorp). Conveniently OwnYourOwnLabs sells a prepackaged bundle at 68.25$ with these tests.

Exercise

The exercise routine is 1 hour a day, every day. It includes some HIIT (10 min per session). The whole routine is heavy on stretching and body weight exercises. I don't know if this particular workout is the best possible one, especially for everyone. The 'Bodyweight Fitness Recommended Routine' sounds good too.

Perhaps a better question is what kind of exercise one should do. Doing Barry's Bootcamp twice a day is further away from these two workouts, and I suspect this is not the best way to approach exercise. Maybe one could go here and pick one exercise per muscle group?

Could one be able to tell what the best possible workout is from tests alone? I don't know!

Something I do think matters though is having some justification for why some workout and not another. For example one could start with premises like 'it's important to strenghten all muscles' or 'it's important to strenghten muscles by doing movements that are natural for daily activities'

Back to basics: protocol adherence for all

A decent diet and exercise is not that unusual. The hard part is that in Blueprint you do the same thing, every day, without exceptions. That is unusual. Eating nutty pudding and super veggie daily might sound weird but remember that most people have the same thing for breakfast over and over. I tend to have yogurt (Siggis is my goto) with pecans and/or walnuts and a berry mix daily and it doesn't get old.

Famously, "diet and exercise" interventions have low adherence rates. So, even if we buy "do Blueprint and you'll get ripped", most people won't do that, even if they accept that it works.

One reason why adherence must be low is friction. As formulated, Blueprint looks very complex and dauting. Even the recipes are not as clear; eg. one says 300gr of black lentils but it does not say whether they are cooked or not (One has to then look for the weight ratio of cooked to uncooked to figure out what it was). The workout could potentially be made simpler and recorded. Moreover an explanation for why that specific workout could be given.

There is some acknowldegement of this concept in the "Principle 3" section of the doc where Johnson says he eased into Blueprint over 4 weeks, adding one thing at a time.

What Blueprint need is an MVP, something that is simple, affordable, and that works for most people.

One difference from 'just diet and exercise' is that Blueprint would involve a specific exercise routine and perhaps a specific set of dietary restrictions (In particular caloric restriction with optimal nutrition, CRON + low sodium + vegan). It could be made more flexible with objective thresholds: e.g. a diet that allows ice cream only if body fat is under a certain percentage.

Something going for Blueprint is that it's already a meme, and doing something together with others is a very good way of getting a new habit to stick. 'Blueprint Bootcamp' where groups of people sign up to a challenge could work. Not everyone will continue doing it, but a small fraction might.