Hacker News new | past | comments | ask | show | jobs | submit login
Body on a 3-day fast: Real-time data with Basis (basishealth.io)
72 points by TheJurgen on Jan 18, 2021 | hide | past | favorite | 72 comments



> At the 3-day mark, I've hit a key milestone in cellular regeneration. My body has broken down old immune cells and generated new ones (Cheng et al., 2014).

... if you were a mouse. Here's the article being referenced: https://pubmed.ncbi.nlm.nih.gov/24905167/


There's now a number of studies on effects of fasting on chemo therapy patients, and they all make the same claim of regeneration through autophagy for some types of leukocytes.


Could you explain how this is used to support statements like:

> I've hit a key milestone in cellular regeneration

What's key about it? In what way is it a milestone? What's the expected health impact in human patients?


Well, if he was a mouse you still would have those same questions in the air until you do the proper blood tests, right?

Of course this is a naive, over-simplified interpretation of science findings, but it doesn't mean the science it's inspired by is bad.

One thing that I couldn't find any research on is what are the effects of repeated fasting cycles (and that's how most of people do it now). Traditional fasting is something you do few times a year, and then you have a recovery period afterwards. Can't find that research now, but I remember reading a paper saying that it takes about 2 weeks of normal food intake to get back to the initial white blood cells counts. What happens if you don't let the body recover and instead create a new calorie deficit stress seems unclear?


> Of course this is a naive, over-simplified interpretation of science findings, but it doesn't mean the science it's inspired by is bad.

I am complaining entirely and exclusively about the interpretation, I’m sure the science is fine.


If so, why not cite those?


Because it's easy to google it, and I don't have that much free time... these links are on a first page of google results when you search it:

Effects on stem-cells: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4102383/

Reducing toxicity of chemo therapy: https://jeccr.biomedcentral.com/articles/10.1186/s13046-019-...

clinical trials on humans: https://clinicaltrials.gov/ct2/show/NCT01175837


I will add a non comprehensive list of relevant papers at the end of the post for anyone who wants to look further into the research.


So... is it good to break down old immune cells?


It's not about just randomly breaking the cells, it's a normal process called autophagy where damaged cells are being recycled to be replaced by new ones. This happens all the time, but it seems that fasting can increase it significantly. The leukocyte count only temporarily falls, and then gets back to normal. This is extremely beneficial to chemo therapy patients because chemo damages cells, suppressing the immune system as a side-effect and also makes people feel very sick. This can reduce these effects.


I’d guess so, but I’d also guess your body is doing it all of the time


I love how there are exactly zero technical details on the website about how this Basis thing works and what it is. It's chock-full of marketing copy and unnecessary animations instead. But from the early access form I can gather that it's going to have a monthly subscription (of course, it's 2021 after all) and probably need proprietary consumables.


What would you like to know?


What the device is and how it works.


Basis as a system works with multiple passive, continuous inputs. The goal is to help you track your health without thinking about tracking your health - where many people fail. So to use Basis today (get the minimum benefit) you would need to use a glucose tracking biosensor (that we provide) and have an activity tracking wearable (e.g. Apple Watch, Oura ring, etc). It’s possible to use Basis with your own continuous glucose sensor (if you are diabetic let’s say) and already get those regularly. Depending on the peripherals/wearables you have you could also feed your water intake and temperature in the data as well.


> you would need to use a glucose tracking biosensor (that we provide)

It is an "Abbott's Libre Sense Glucose Sport Biosensor"[1] or a different sensor ?

[1] https://www.libresense.abbott/en/home.html


I don't believe in these over-engineered 'quanitified self' solutions. Just get yourself a simple step counter and go for one or two walks a day and try to up the step count by a bit every time and don't eat like an asshole. It's worked for me for a long while now.


> don't eat like an asshole

Man, it feels like you're taking a shot at me. I've averaged my 10k steps a day for almost a year now (actually averaged close to 11k in 2020 on a whole,but didn't get consistent until mid-March), and, while I have lost weight, I just can't get my diet under control. It's quite frustrating, honestly, and stems form my lack of desire to meal prep.


You don't have to meal prep per se, but I find having a plan helps. Buying and preparing food when you're hungry will make you eat more of the wrong stuff than if you had a plan.

I also found that by basically tripling my intake of vegetables helps a lot with feeling satiated. No one ever got fat because they ate too much broccoli, lettuce, and spinach.


> No one ever got fat because they ate too much broccoli, lettuce, and spinach.

Challenge accepted.


This is kinda unexpected comment on hackernews. :)


Can anyone explain the significance of these numbers, and their changes?


A personal story about fasting:

About 8 years ago, I had a particularly stubborn stomach bug, that just would not go away. After a week I went back to the doctor, and he was reluctant to put me on anti-biotics incase it was viral, and also he was generally anti-anti-biotics if your immune system could do it on its own. So he had me do a 3 day fast. Only water, and once a day a small bottle of Gatorade. Day 1 was easy, day 2 was hard, day 3 was actually easy again. On day 4, I was told to reintroduce food slowly and to eat as close to unprocessed foods as possible, so I had a bowl of watermelon for breakfast, a cup of oatmeal for lunch, then for dinner... McDonalds.

The outcome was the bug did die, but I no idea if it was from the fasting or just that it had been 10 days. But the fasting wasn't that hard, and I felt better afterward.


The gatorade sounds horrendous, to dump a ton of sugar on you when fasting. It was probably for the electrolytes, a pinch of salt would have worked without all the junk in gatorade. And then you rarely need it on just a three day fast, only if you're dizzy. And if you're dizzy it means you've drunk too much water.

Glad you're better though!

Reddit fasting sub is ok to learn more, though they have some hangups like dry fasting.


I was just following the doctor's orders. I assume he wanted me to have caloric intake along with salt, but I'm not a doctor, I didn't question him, I don't know anything about fasting, and don't even know if it was necessary.

I've done a couple of 2 day fasts since then, just to see if I can still do it without much trouble, it gets harder as I get older.


When you water fast you excrete salt with the water you're drinking. Beginners tend to drink far too much water to mentally replace the food, and so they lose a lot of salt. Gatorade when it was invented was an electrolyte replacement. Now it has a lot more sugar too. It was probably for the salt.

My longest fast was two weeks, AMA. :-)


For clarification, I take a daily electrolyte tablet with vit. C, magnesium, calcium, potassium and sodium when I fast. Generally, you want to take in less than 40 calories to stay in a proper water fast and I don’t think Gatorade fits the bill. Unless you’re doing a Fast Mimicking Diet where you’re allowed to eat up to 700 calories (I think). My longest was 5-day water fast (a few times). I found that Day 5 is better than Day 4. What days do you think are the worst in a 2-week fast?


The transition to ketosis is usually the hardest, which for most people is around day 3.

If you eat keto it isn't a problem.


A small bottle of Gatorade is 140 calories, I believe. Not nothing but also not too much to work through. Also you don’t have to chug it all in one gulp.


Why Gatorade... I assume for electrolyte? There are so many better alternatives. I had a serious electrolyte imbalance first time trying Keto diet. Took a potassium supplement and immediately got better.


I was just following doctors orders, and can only assume because it has electrolytes and calories.


I'm not biochemist but:

1) If you are not eating for some extended period of time, your body starts to have deficit of not only glucose but also proteins which are then accquired from your own muscles. E.g you are not losing only weight but muscles too.

2) There are various types of immune system cells and they all have different lifespan. If your body detects new pathogen it start's to produce various new immune system cells to fight infection. For example: B-lymphocytes serves as "memory cells" for specific pathogens so it is required for them to live as long as they can. What exactly are "old immune cells" then?

Thanks for answers :) Your idea of continuous health monitoring is exactly what I'm searching for, like for years.


For the first problem, you need to ensure that you exercise the muscles along with fasting, to nudge the body away from salvaging proteins from muscle tissue. (Not a bio chemist either, this is what I have read in most articles about intermittent fasting)


I wonder what this "biosensor" is. There are several Glucose monitoring products out there however connecting them and retrieving data (for other purpose) is usually a non trivial task.


I'm also wondering about that. Abbott isn't known for making the raw CGM data easily available. Neither is Oura.

It might be possible via Google fit/ Apple health or via nightscout for the CGM devices, but the latter would imply some legal trouble. Maybe OP can clarify some details of how the integration works in this product.


Yeah. However, if you have a libre2 you need a patched app and then nightscouts xdrip etc to get the data of the cgm into a db.

Or you need bubble or miaomiao and again xdrip ;)

Adding a description of the setup that the author used would make the article much more substantial than it is now.


I used the system I built with my team for real time health data tracking that includes data from a continuous glucose sensor, Apple Health data (sleep, activity, heart rate, etc), urinalysis feature within Basis for the majority of this. The only data that wasn’t acquired via Basis was the blood panel.


so we were wondering which "continuous glucose sensor" and how you were able to extract the data from it.


We’re developing our own but currently using an Abbott sensor.


ah. ok thanks it would great if you could post more info on your own sensor ! much appreciated, thanks !


Off-topic, but I wonder if/when realtime glucose monitoring will ever come to the Apple Watch and other activity trackers.


Some non-invasive solutions for CGM devices are quite good already. I'd expect that the technology will be included in many higher-end health tracking devices within this decade.


Wasn't Google developing contact lenses to monitor glucose a decade ago?


I'd forgotten about that. Apparently they stopped development on it in 2018 [0].

[0]: https://www.bbc.com/news/technology-46262520


Just wondering - how has your experience with obtaining and aggregating relevant data like sleep duration and quality, heart rate, physical activity,... been and which sensors are you using? I was considering Empathica e4 + Freestyle Libre for a similar research project seeking to quantify influences on BG levels.


So I fasted 3 days about 3 weeks ago (after Christmas) ok my spouse thought I was insane but I overate so much that I had the need to swing the pendulum in the other direction - after this 3 days I ate a Pizza - no cramps; nothing; just operated as usual. So I am not so sure about the whole story...


> it’s important to break your fast with a nutritious, balanced meal that will further enhance the all-around cell and tissue health-boosting exercise you just undertook

Could you give some examples? What's safe to eat?


My fast break meal is a piece of watermelon <10 mins break> spinach, avocado and macadamia nut salad with EVOO <10 mins break> one egg omelette.

Watermelon is very easy for stomach to digest even though it has a fair amount of fructose and would otherwise not be considered optimal for first meal. The breaks are to see how stomach is reacting to food introduction. Rest of the options are made as options that activate sirtuins (restorative cellular signaling) while system is sensitive to new inputs.


I wonder how the TSH levels were a week before the 3 day fasting, I'm curious if that -2.37 delta is from only fasting or your TSH varies that much too often.


Good question. I tend to be on the higher side but after doing a 5-day water fast, several 3-day ones and IF year round I’ve seen that fasting more than 1 day has a significant impact on my* hormones.


What is the impact of fasting on hormones?


Good or bad effect on hormones?


I might have missed that from the article, but are you male or female? (I wonder wheter the hormonal response would be similar for both sex)


Male


What's the sensor being used? Doesn't show any detail on the site. I assume it only tracks glucose and the rest are done at lab?


There’s a continuous glucose monitor, ketone measurement through urine, hydration tracking, sleep tracking and less relevant to this experiment heart rate analysis in Basis. The more complex blood work was done in a lab that’s correct.


Is it possible to measure cholesterol and blood fat levels via wearables? That would be killing feature.


Agreed. Working on it


Do you develop sensors?


We do but we’re currently using an Abbott sensor


It would be interesting to see the same breakdown analysis with different diets.


Hey zenyc, we did 2 cycles of Fasting Mimicking Diet (FMD) and tracked our markers during the 5 day fast: [1] blood glucose (trending down) [2] body fat percentage (trending down) [3] ketone levels (trending up)

Here the writeup: https://www.artlapinsch.com/fmd/


What type of diet would be of most interest?


There are a few fad diets that keep coming back, for example:

* Paleo

* Keto

* Something with (only) liquids


Keto is a fad diet that was developed for legitimate reasons.

It helps in some cases of epilepsy.

Its just a tidbit I wanted to throw in, as not all diets are equal, but some diets are more equal than others.


One that stresses the body and is not a fast


How about the American diet. A week on pizzas, burgers, sodas and donuts with real time data?


Yes and compare with a healthy diet. Control as many variables as possible (eating time, exercise time and duration, etc)


Ok will post this next


Fasted for 3 days and muscle mass increased by 2.4kg. Yeah, right.


That’s a percentage increase as a byproduct of losing overall weight.


Is this Basis same as Basis watch?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: