Fruit And Vegetable Intake: How Many Servings/Day Is Optimal?
Вставка
- Опубліковано 8 лип 2024
- Join us on Patreon! / michaellustgartenphd
Discount Links:
NAD+ Quantification: www.jinfiniti.com/intracellul...
Use Code: ConquerAging At Checkout
Green Tea: www.ochaandco.com/?ref=conque...
Oral Microbiome: www.bristlehealth.com/?ref=mi...
Epigenetic Testing: Trudiagnostic.pxf.io/R55XDv
Use Code: CONQUERAGING
At-Home Blood Testing: getquantify.io/mlustgarten
Diet Tracking: shareasale.com/r.cfm?b=139013...
If you'd like to support the channel, you can do that with the website, Buy Me A Coffee:
www.buymeacoffee.com/mlhnrca
Conquer Aging Or Die Trying Merch! my-store-d4e7df.creator-sprin...
__________________________________________________________________________________________________
Paper referenced in the video:
Fruit and Vegetable Intake and Mortality Results From 2 Prospective Cohort Studies of US Men and Women and a Meta-Analysis of 26 Cohort Studies
pubmed.ncbi.nlm.nih.gov/33641... - Наука та технологія
35 servings/day is outstanding! My stomach can't handle even 10/day.
It doesn't have to be 35 servings/d, but we can all replicate the approach, i.e. using objective biomarkers to determine what/how much is optimal.
Thanks Michael. Always great video. 👍
Thx Michael. Look forward to observing your next lab results to help validate expected biomarker changes.
Ha, or not-if my biomarker data is worse, then it's re-evaluate and see what's significantly correlated with the biomarkers, to try to get closer to the ideal diet (for me) at the n=1 level.
This is a very interesting post. Some much it feels to unpack. A number of points strike me immediately:
- massive quantities of fruit and veg; not how the average person eats at all. Closer one might argue to evolutionary dietary origins. The takeaway message given the average persons diet - eat as much fruit and veg as you can (because no matter what, they’ll never reach these values)
- secondly, I was very surprised by the fact that of the two components, fruit and vegetables, the former if anything, seems to contribute more toward healthy biomarker development
- some advocate for eating fruit only seasonally but this seems to blow that myth entirely
- another common belief that appears to be false is the glycemic effect of fruit - since the blood glucose correlation is negative if anything
- this presentation questions the argument that dietary sugar even of fruit origin causes diabetes
Excellent work
Thanks Kevin. I wouldn't say that necessarily more is better-for some people, more may not be better. To help evaluate how much is optimal at the n=1 level. I think we can all benefit from regular blood testing.
@Kevin Eakins
Actually, you can read the data as if the bad effect of to much sugar in fruit appears above 50 g sugar, or so
I separate between fruit and berries consumption, since I consider berries much better than fruit.
They have half the sugar for more on the good stuff. I think you can eat twice as much of them. So 500 g of berries, or 250 g of fruit.
One way to stress-test the correlations would be do significantly lower your fruit and vegetable intake to approx. 5 servings a day and check your blood again after 3 months.
The net correlative effect of my fruit intake is positive on the biomarkers, so that suggests that dramatically reducing it below my average intake will be bad for biomarker changes, so I won't do that. In contrast, I've cut my vegetable intake from 20 to ~14 servings/d, so we'll see how that impacts the biomarkers (or not) on the next blood test next week.
Dear Mike, many thanks for this video. This is great information! Do you have a blog post about your meal plan and timing for the different foods or could you share this information? I struggle to include a lot of fruits in my daily diet.
E.g. I do not know if I should eat fruits first thing in the morning, after meals or between meals (as an afternoon snack). And which fruits can be combined and which should not.
Would be nice to hear just how you do it (no need of explanation), so I may can pick up something for my own daily diet routine.
Hi Paulo, my latest diet info is in this video from September:
ua-cam.com/video/ue-mzz1bm3E/v-deo.html
I just blood tested today, so another diet update will be in a future video. Note I post my daily diet on Patreon, so that may offer value:
www.patreon.com/MichaelLustgartenPhD
Which fruits and vegetables are best and could you make a practical video on how to include so many servings?
Which FV are best is likely variable between people, but I place great emphasis on carotenoids. For ex., see this video:
ua-cam.com/video/Hvk1wvDmlms/v-deo.html
Also, a full diet breakdown is in this video:
ua-cam.com/video/ue-mzz1bm3E/v-deo.html
Thanks. It may be a stretch depending on quality of data, but it seems that when comparing right directtion vs wrong direction (e.g. glucose vs Lymphocye) it would be worth factoring in how much the shifts change the hazard ratios for given markers. Or maybe in your case the shifts are so small as to have no discernable efftect on individual HRs. Would probably be more telling in someone just starting with fairly bad blood panel.
Hey Peter, that's tough to determine at the n=1 level. Also, I don't think it's possible to say that 1 biomarker is more relevant than another at the n=1 level. For example, is glucose more important than hs-CRP? To me, they're all important, and looking at the net effect provides value.
Thank you Dr L for sharing your information. Great work! Looking forward to the lab results!
Question: How many times per day do you eat? Do you have a time period? Just curious… Thank you so much!
Thanks JBen. I don't count how many times that I eat per day, but I definitely have an eating window, between 530 AM to before 3PM. Also, a 6 to 7hr fast (or mostly fasting, with little food/water) before bedtime at 930 PM. I don't do that so much for any purported health benefits, but it limits nighttime awakenings to pee with all the FV that I eat.
Awesome! Thank you for replying back to me. I have similar lifestyle choices as you, and I’m always trying to get the best out of each day. Again, thank you for your work and thank you for sharing it.
Have they in any way corrected for the healthy user bias? (like income, exercise & sleep etc)
Do you have any thoughts on Broccoli Sprouts (sulforaphane), seems to be a lot of hype around it. Would be interesting if you track sulforaphane and understand effects on biomarkers.
I haven't made a video about it yet, but I do have a blog post on my website about sulforaphane:
michaellustgarten.com/2014/08/13/restoring-boiling-induced-sulfurophane-depletion-in-broccoli-with-mustard-powder/
Thank you for sharing! Curious if you also track caffeine intake? I just learned vis a vie Modern Healthspan that caffeine may suppress some forms of chronic deleterious immune system responses.
Hey Matthew, although cronometer records my daily caffeine intake, I don't track it, as I consume approximately the same amount every day, with green tea as the primary source. I also get relatively smaller amounts from cacao.
@@conqueragingordietrying1797 might be interesting to boost it a bit and see if numbers change... once you are done varying other numbers of course. btw did you see the recent iAge series on Modern Healthspan? He interviews a researcher from Stanford who has created an aging clock based around immune cells. And he plans on selling access to this clock to the public.
@@surfreadjumpsleep Ah, I wish I could. I'm caffeine sensitive, which means that I have to limit coffee intake because even if I drink it in the morning, it can mess with my sleep onset at night. So I rarely drink coffee (in addition to my daily green tea).
Yep, I saw the MH iAge series, and read the paper-unfortunately most of the immune biomarkers aren't commercially available, but hopefully that will change in the near future.
Could you make a video sharing your diet?
Hey Veekz, a full diet breakdown is in this video from September:
ua-cam.com/video/ue-mzz1bm3E/v-deo.html
Michael where are your daily NUT and SEED consumption levels as these are also considered fruit ?
Nice
Can you get Lymphocyte (%) by diving total number by total number of immune cells? Or is it a different test?
Yes, absolute lymphocytes/total WBCs
Is there a possibility that the correlation of FV with ALT is due to pesticides? Did you notice any change when you switch to organic for example? A quick Google search indicates a connection.
That's doubtful (for me), as I eat all of my FV organic. I switched a long time ago, and it was a gradual process, so I didn't notice any major health effects.
I am very curious about vascular age if, in addition to bio markers based on blood analysis, one can analyze the calcification in the arterial system and heart valves can be added to the longevity scores. It has been my experience that one can have a high longevity score (10 years or more phenotypic age below chronological age), but still have a disparate cardiovascular age. You thoughts?
Yes, that's an important point. I track blood pressure as a related metric, but not as rigorously as the other biomarkers. For now, I'm at ~125/70, which isn't bad, but I'd prefer data that is closer to 100/60. I'm working on that, and I'll probably make a video about it sometime in the future (later vs sooner, though).
There is an odd piece of data I came across. Someone was looking at the biomarkers of some hunter-gatherers. One detail stood out. They had arterial calcification. But, for whatever reason, they did not have cardiovascular disease. Heart attacks, strokes, etc were rare among them. This indicates that there is something else going on with cardiovascular health.
@Abdel-Ilah BENAHMED Thanks Abdel-Ilah BENAHMED. Even at the Dr's office, it's about the same. Weight loss can affect BP, so I'm working on getting back to being leaner, which should optimize BP. Stay tuned!
@@conqueragingordietrying1797 1st let me state I so admire your work. It is very meaningful.
At age 76 and vegetarian for 55 years, I went vegan in 2006. I exercise 2 hours daily, and practice CR and eTR, and normal BP is in the LT 120s/over 60s. Nighttime and some daytimes resting HR in the high 40s, Still, had some aortal calcification probably form dairy. I bring this up as you have made among so many excellent videos some on cholesterol where you monitored increased intake. I keep mine at zero. I have no genetic markers for cholesterol risk. I shar3e this as you are young and many of your readers are young (ish). As we all are monitoring risk factors and desire to optimize longevity and down regulate risk factors, I would find it prudent to be on the lookout indications of vascular ageing (Circ Res. 2018 Sep 14; 123(7): 849-867.
doi: 10.1161/CIRCRESAHA.118.311378
PMCID: PMC6248882
NIHMSID: NIHMS1502302
PMID: 30355080
Mechanisms of Vascular Aging
Zoltan Ungvari,1,2,3,4 Stefano Tarantini,1,2 Anthony J. Donato,5,6 Veronica Galvan,7 and Anna Csiszar1,2,3)
@@RobertCHood Good points, thanks Robert.
Is this on raw veggie, juiced or cooked veggies? Does cooking actually lose nutrients?
Raw and cooked. I don't juice, but make smoothies instead. To minimize nutrient loss during cooking I drink all of the remaining fluid.
What are the metrics for the types of FV? i.e. Carbs, Protein, Fat, type of Fat, Total fructose / glucose.
How can you be so sure as to weight all the biomarkers equally? I would think the true effect of the biomarkers would be some weighting scheme and probably include exponential versions of the same biomarker variables. Is that how PhenoAge does it?
It's a major challenge to say that any biomarker (or set of biomarkers) is more important than others. For example, is glucose more important than markers of liver or kidney function? I don't think so. Maybe I'm wrong, so if people think they have a better approach, go for it, and the proof will be with improved health over time, and potentially, greater longevity! I'm more than happy to modify my approach if someone comes up with a better method.
@@quantifiedmax Hey Max, I'm more interested in getting them all to be simultaneously optimal, rather than determining whether they're equal, or not.
@@conqueragingordietrying1797 I respect a thought process that says in the absence of knowing how to determine which variable is more important than another, I treat them all equally (weight = 1). However, I was just thinking maybe the PhenoAge or Aging.ai have some coefficients available, such that those parameters are better than equal weighted ones.
@@imtryinghere1 Ah, but those weights are based on large (n>10,000) population-based studies. Whether the weights are as accurate at the n=1 level is very debatable. For now, I treat them all equally.
@@conqueragingordietrying1797 Who cares. Are you saying you are different than the mean estimated population of 10,000 observations? Why not take those parameters, in the absence of data otherwise? or are the parameter/coefficient values not available publicly?
Instead of just direction, each marker has a coefficient in the regression model. I.e. delta one marker is more powerful than another...
Yes, I understand that's possible. However, at the n=1 level, there's no way to know which variable has the biggest impact on health. I aim to optimize all organ systems simultaneously, with no weak links, so I score them equally (for now).
@@conqueragingordietrying1797
Sir, forgive me. Of course you are quite correct. Please accept my apologies for such a simplistic suggestion.
What ment to say is each of the components of the regression model used to predict the outcome variable (age) have value that weights the contribution to the outcome variable Y.
If the regression model equation is published in the original research the by simple inspection of the regression model equation the "powerful" vs weak contribution to model is easy to visualize.
Each variable contribute a % to the predict power of the model.
Stepwise regression uses these values in creating the regression equation models and are usually included in the statistics output when the model is created. Often it is even discussed in the original paper in a statement like:
"Blood pressure explains 90% of the predict power of the model"
Am I correct, sir?
No apologies needed, @@drrutledgemd!
Yes, I understand how it works, but note that's based on large studies in other people. Whether the population data will have similar weights at the individual level is unknown (although it's reasonable to assume they'd be similar).
I'd rather err on the side of caution and treat markers of each organ system with equal scores, rather than giving preference for a few over others. If I'm wrong, and others use that approach, they'll have better long-term health and/or lifespan. If not my way may be the right path. There's no way to know until then, unfortunately!
@@conqueragingordietrying1797
Sir,
My apologies for bothering you further.
Please feel free to ignore the following long post on selecting how to choose dietary and lifestyle intervention changes in upcoming studies.
If I understand correctly you are assuming that the relative value/predictive power of the components of the model that you are using are (as far as you can tell) roughly equivalent and that for example 4 in the “right direction” is better than 3..
Maybe not.
Forgive me, I am an old trauma surgeon and offer you an example where the individual components were recently shown not to be equivalent.
Small Example:
Article summary, (edited)
Glasgow: The Power Is in the Motor, October 9, 2022
Glasgow: The Power Is in the Motor: Abstract & Commentary
Improving the Glasgow Coma Scale score: Motor score alone is a better predictor. J Trauma 2003;54:671-680.
The Glasgow Coma Scale (GCS) has served as an assessment tool in head trauma for decades
... often is used to rapidly assess neurologic status as a predictor of survival and severity of injury.
... study compared the predictive power of components (motor, eye, verbal).
... the relationship between GCS score and survival was not linear.
One single component of GCS, the motor score component
** is related linearly to survival AND also has ** almost all of the predictive power of the entire score!
I.e. There was one very powerful component of the score that had almost all of the predictive power of the entire score!!!
=> Now to your wonderful and very powerful work:
Question #1: Are some of the components of your predictive models more powerful than others?
Answer #1: Almost certainly!
Questions #1x: What are the most powerful predictors? What factors make minimal impact? What if Blood Pressure/Glucose/BUN etc. explained 90% of the outcome variance?
Knowing the power of the various contributing factors I think would be of tremendous importance. Yes?
Question #2: Can we discover the power of the different components of the model?
Answer#2: Yes, of course...
either by
#2a: reviewing the published equation for the linear model coefficients (is the equation free to view online? (simple, I will volunteer to do it if refer me to the publication)
(again I can volunteer to do this if you like, but it's a bit better to just look it up. ;-)
Or by
#2b: brute force (tedious but doable)
If the predictive model is free and online you can simply add different data and measure by direct inspection the Change in the predictive age values
Iterative technique to determine the power coefficient of the components of the score model
Example:
Step 1 input your sample data => get outcome value: age
Step 2 input sample data but change one value by some amount (i.e. 10-50%)
Step 3: Record the delta age prediction,
i.e. example: 10% lower serum glucose => 10% lower predicted age
Result for an arbitrary number of variables you will have the relative contribution to your predicted age, for example a decrease in 10% of blood glucose will result in some level of change in the predicted age. Let us say for argument's sake the age decreases by 10% as well.
Repeating the above technique for LDL cholesterol would give a similar value of
for example 50%
i.e. example: 10% lower LDL Cholesterol => 25% lower predicted age
Question #3: What is your best next modification of your diet/lifestyle model?
Answer #3: With information from above investigation modification of diet/lifestyle might be best
supported by changing the more powerful predictor of the model .i.e. LDL cholesterol might be predicted to have greater impact than glucose or if the power were reversed than a reverse strategy might be best.
Michael, why would you say, that the lowest mortality was associated with 10 servings of fruits and vegetables? There was nothing of that sort in the data you presented. There was no further improvement above 5 servings per day😜(2 servings of fruits and 3 of vegetables).
Hey Krzysztof, all-cause mortality risk reduction for 4 vs 10 servings of FV was similar, I should've mentioned that. Whether 10 is better than 5 isn't the point of the video, but rather, how I evaluate my outlier dietary data when compared to meta-analyses.
@@conqueragingordietrying1797 i understand, but as someone has pointed out, you would know your n=1 correlation if you tasted that, but since your lowest intake was around 20 servings per day it is challenging. Did you check if your blood test taken closest to your lowest intake period compared to the one taken closest to your highest intake made any difference with regards to phenotypic age?
@@krzysztofh6308 For the past 3+ years, neither of the 2 biological age metrics are significantly correlated with FV, F or V intake. Maybe I need more data, no worries, I'll continue to test at least 6x/yr, at we'll see how the correlations look with more data.
Interesting, the study below showed risk reduction up to at least 10 servings. I actually confused the two studies until I read your comment. Fruit and vegetable intake and the risk of cardiovascular disease, total cancer and all-cause mortality-a systematic review and dose-response meta-analysis of prospective studies [published online February 22, 2017]. Int J Epidemiol.
How many meals are the intakes taken in each day?
I don't divide it into number of meals, but I have about a 10h eating window from 5-6AM to before 3PM.
Not striclty related to this video but to all your videos in general.
I noticed that the only big difference in my diet to yours (other than the fact that I'm vegan but your animal food intake is really low) is the amount of carotenoids. Other than tomatoes and sweet potatoes that I don't find always in market I barely eat red/orange whole food, sometimes red peppers but it's anyway a lot lower than yours.
I started using paprika instead, any info on paprika, it should cover the carotenoid intake easily
How many grams or oz is a serving?
80g
are you familiar with Ray Peat ?
There's this 82 year old lady on Anthony Chaffee's channels who looks like she's in her 40s and who's been on an all meat diet for 65 years. Could the carnivore pathway be valid to those who can stomach it?
There are many roads to optimal health-I'm not married to any dietary ideology. But, I am married to the idea of regular blood testing and tracking various metrics (diet, fitness, etc) to find the diet that works best at the n=1 level.
How the f can anyone have "over 50 servings a day"??!... much less "17"??
Good question. Though a serving size is 80 grams. I read that's the equivalent of three tablespoons of cooked greens or one fruit on average.
@@jackbuaer3828... so, a salad?
@@joshuascuteri1208 Oops, that 80 gram serving size was from a different study. I don't know what they used in this study
80g = 1 serving, that info is in the video
How does someone afford all those fruits and vegetables?
Interesting data analysis.
LDL going down with high veggies, probably why plant based people advocates for lower LDL
Is that mortality of eating fruits/veggies vs junk foods. Obviously plant based is better than junk foods or the SAD diet
Again food questionnaires should be taken with a grain of salt, how serious are people documenting their eating habits.
Interesting that the biological age score did not change. It is in conflict with the first minute of the video which states that more fruits/veggies gives you longer longevity.
Are you able to remove other big variables in your analysis such as exercise, sleep/recovery, stress, current diseases, smoking etc...
More people in food questionnaires studies does not means better data.
Given my excellent lab results, I will continue to be more animal based rather than plant based.
The purpose of these analysis is to continue to search for trends and better understand its impact on the health biomarkers and potentially on the biological age score.
I am sure there are not just one methodology given the different variables that affect us and our genetic make up.
I'm not suggesting that people eat as many FV servings/d as me, but to use blood testing to identify how many may be optimal for them.
@@conqueragingordietrying1797 what are your thoughts on my comments: Interesting that the biological age score did not change. It is in conflict with the first minute of the video which states that more fruits/veggies gives you longer longevity.
Hey @@claudecamire5913, sorry about that! The data on the 1st slide is for all-cause mortality risk, not life expectancy, but nonetheless, the idea is similar and I get your point. That my overall biological age scores are not significantly associated with FV intake can be possibly explained in 2 ways: I don't have enough data (12-14 blood tests) for that yet, whereas for the others there's more data (15-17 blood tests), or, because those scores are the cumulation of many biomarkers, the positive effects of FV intake get lost in the variability for including many biomarkers, rather than looking at them individually. Don't get me wrong, the biological age scores are important, but I don't place their value above the cumulative correlative effect of all the other biomarkers.
The problem is the research is mixed on cholesterol. It's not clear that overall LDL has any relevance to cardiovascular health.
Bart Kay made a response video to this:
"You go and tell an individual person, even if this is a causal artifact that you are showing here, which it isn't, let's just imagine for fun it's causal, and if you eat more than 2 servings of fruit and vegetables a day then you're odds of dying every year are somewhere between 0 and 1 in 10,000 different, that's not going to motivate you to do anything is it? There's no utility there. None at all. Laughable."
The point of the video is to assess with n=1 data if more than 10 servings of FV/d, which goes beyond the published literature, may be optimal for health or not. Bart can have any opinion that he likes, I'm not concerned about that.
There are 4 calories per gm of fruit and veggies. If you are eating 2000 to 3000 gms per day, that's 8000 to 12000 calories per day. That too many calories for your weight. Something is wrong with you numbers
There are 4 calories per gram of pure sugar, and FVs are not pure sugar. For example, 100 grams of carrots = 41 calories, or 0.41 calories per gram.
Fruits and vegetables can be up to ~96% water by weight. 100 grams of cucumber has 95.2 grams of water and is only 15 calories for example.
As everyone knows, correlations are next to meaningless or even worse than meaningless, as they can be misleading. This data is mostly comparing people on unhealthier diets with those on healthier diets. All we are seeing is the confounding of factors, as involving the unhealthy user effect and healthy user effect. Until we have a study of those on healthy animal-based diets (nutrient-dense, nose-to-tail, etc) in comparison to those on healthy plant-based diets, everything we say is speculation that might be useful for formulating hypotheses in doing further research but not for making scientific conclusions and dietary recommendations. Aren't we long past relying on data like this and pretending it has any greater value?
So you're saying that correlations for my own data, with up to 17 blood tests since 2018 is also meaningless?
Well, in addition to studies that you dislike, there are no long lived cultures that have lived on a primarily animal based diet. So maybe that's another clue to the riddle, maybe not.
I see the argument that you're making from the carnivore, keto, and low carb camps all of the time. Those camps seem to want to throw out all or most epidemiological data that exists on mortality and disease. That seems to me to be a foolhardy approach.
I would love to see further experiments. I guess you can do your own experiment by doing a carnivore or carnivore like diet for 4 months, take blood tests, and then do a diet with 10 or greater servings of fruits and vegetables daily for four months, and take comparison blood tests. I would think the experiment would be best run on people age 40 and older, since I am guessing that's when biomarkers start going in the wrong direction for most or all people due to aging.
@@conqueragingordietrying1797 - Not meaningless. But no more meaningful than all the data that contradicts it. There have been many studies, some in the West but most in Asia, that show meat-eaters are healthiest and to the degree that they eat more meat.
So, why cherry pick some data while ignoring others? That isn't to dismiss your anecdotal evidence, as that is an important part of the debate. But the point is that there is no clear conclusion yet to this debate. Let's be intellectually honest and humble about that.
Maybe the Carnivore, Keto, and Low Carb people can jointly do a go fund me to fund a university study, whereby they do not get to pick the university, to reduce the risk of bias.
@@jackbuaer3828 - That would simply be a false statement. Most long-lived cultures have been on animal-based diets. See: "Blue Zones Dietary Myth".
Interesting!
But I think you overestimated the effects of your own fruit intakes.
You eat a lot of berries. And that's, in my opinion, very different from fruit!
Berries has more nutients for less sugar. I think you can eat more of them then of fruit.
But you also eat Dates! And as fruit goes, they are probably too sugary to be good for you.
In my opinion, there's no need at all to restrict vegetables. And I'm not a vegetarian.
As you can see, I blame the bad effects of too high fruit intakes on the high sugar intake. That's mostly based on common sense.
But also supported by the difference between fruit and vegetables.