25 replies to this topic

[Zarathrustra]

To expand on how I know to I am doing: I have algorithms that convert my daily activity into five health-illness metrics.

 

I presently use a total of 32 biometrics to assess my overall illness-health metric, and also subdivide these into 12 for cardiovascular, 6 for CkD, 12 for cancer, and 3 for body composition.

 

Then my algorithm gives me a single figure for my overall illness-health metric, and similarly for each of the four of my present health concerns. Negative values are equated with increasing health, positive ones with increasing illness.

 

Examples are:

 

Overall (yesterday was -0.562): 

 

 

Cardiovascular (yesterday was -2.1255):

 

 

CKD (yesterday was -0.0508):

 

 

Cancer (yesterday was -1.3933):

 

 

Body Composition (yesterday was 1.5334):

 

 

As overall morbidity is my main concern, I also track that by combining the illness ones (yesterday was -0.5259):

 

p.s. I wished to post the graphs, but that seems to be not allowed. So if you are interested, please see attached document.

Attached Files

•  Proven anti-ageing for me – the metrics.docx   131.52KB   17 downloads

[albedo]

Against what the algorithm is tested, all causes mortality or  ..? Can you please explain better what are you aiming at? Thank you.

[Zarathrustra]

I am aiming at reducing my illnesses: hypertension, CKD3, and cancer.

 

My algorithm is tested against changes in biometrics associated with these: how my lifestyle correlates with these markers and how they change with time - I now use the last 13 years of daily records of lifestyle and various records of the biometrics (some daily, others of decreasing frequency). I change my lifestyle according to these results and test again..

 

In a sense, the mortality figures are built into the biometrics. For example, high blood pressure leads to premature death, so getting that back to about normal (120/80) is good for me. Similarly with all the other biometrics, except the body composition one. For the latter, I assume that increased muscle and bone mass along wit reduced body fat is beneficial. Whilst it may be possible to go too far with such metrics, generally what i have stated is a good thing for those about average or obese or old with sarcopenia.

Edited by Zarathrustra, 24 November 2021 - 10:27 PM.

[albedo]

I wonder if you ever thought to compare with aging metrics existing in literature and/or commercialized, say Levine's PhenotypicAge and DNAm PhenoAge or Insilico Medicine's Aging.ai or TruDiagnostic reports. These and others are based on clinically validated health biomarkers regressed on mortality data, or AI driven on similar biomarkers or epigenetics (DNA methylation) in nature. In this Forum some of these measurements are discussed in the "Biological Age" thread.

[Zarathrustra]

No, Albedo, I hadn't thought of any of that.

 

Thanks for the suggestions. I'll follow them up.

[QuestforLife]

You haven't said what your measured biomarkers are (apart from BP). I am guessing they are things like pulse, BP, HRV, weight, fat free BW; or do you take blood tests as well(I assume these wouldn't be daily)?

 

You say that you have found proven anti aging interventions using this data. Can you share any more?

 

There was an interesting paper a while back that suggested that it was useful to look at the variance in a biomarker over a given period, as well as the time it took for a fluctuation in that BM to return to its long term average. Obviously the BM deteriorated over the long term (with aging), but the increase in variance and the increase in the time it took to return to baseline was quicker and could be more actionable. They combined results from whole blood count tests for their BM (in a way they didn't disclose), but also applied it later to step counts and found it equally applicable. 

 

I thought is was a nice way of capturing resiliency change over a sensible time period; rather than looking at daily fluctuations or very long term changes.

 

 Longitudinal analysis of blood markers reveals progressive loss of resilience and predicts human lifespan limit

 

We investigated the dynamic properties of the organism state fluctuations along individual aging trajectories in a large longitudinal database of CBC measurements from a consumer diagnostics laboratory. To simplify the analysis, we used a log-linear mortality estimate from the CBC variables as a single quantitative measure of the aging process, henceforth referred to as dynamic organism state indicator (DOSI). We observed, that the age-dependent population DOSI distribution broadening could be explained by a progressive loss of physiological resilience measured by the DOSI auto-correlation time. Extrapolation of this trend suggested that DOSI recovery time and variance would simultaneously diverge at a critical point of 120 − 150 years of age corresponding to a complete loss of resilience. The observation was immediately confirmed by the independent analysis of correlation properties of intraday physical activity levels fluctuations collected by wearable devices. We conclude that the criticality resulting in the end of life is an intrinsic biological property of an organism that is independent of stress factors and signifies a fundamental or absolute limit of human lifespan.

 

https://www.nature.c...467-021-23014-1

 

 

 

[Zarathrustra]

I am happy to share. Just thought I’d be brief in the initial posting.

 

Let me say at the outset regarding the various putative ageing biomarkers, that I have come to regard these, at best, as secondary to the seemingly simplistic  ones already known for decades – resting heart rate, BP, etc – which will anyway be the reference of the gene-based ones. Why? Because if, say, a gene-based one showed a person was getting younger but was frailer, rising resting heart rate, couldn’t walk far, etc, then I’d go with the latter as being a better assessment of longevity than a gene-based one predicting longer life. Having said that, I base my life and research on direct factors – my BP, resting heart rate, hand-grip, creatine levels, CRP, etc.

 

An additional caution of the gene-based ageing measures is that these are necessarily based on averages, and any individual may/will vary from this. So a person needs to know their own ageing – probably dubiously based on this average. Whereas said individual can tell if they are walking more slowly, cannot get up off the chair quickly, etc.

 

I looked at your link to the loss of resilience as a putative predictor. Too complex for an individual to easily access.

 

However, to my own approach>

 

I measure over a hundred biomarkers, many via serum and urine analysis – as you correctly thought, not daily but about monthly. I have taken these up to about 100 times over the last 13 years, giving me fairly robust correlations.

I have reduced them to about 30, in three groups, for lifestyle action – for cardio-vascular, kidney, and cancer. I have a further 3 for body composition, done daily – muscle-mass, bone-mass and body-fat (trunk and visceral fat correlates highly with the overall body-fat, so I no longer analysis this separately). And a host of others: sleep, temperature, oximeter, morning HRV. I use an iterative approach – check the correlations with my present lifestyle; change what I do in the light of these, and test again in about a month.

 

My CV ones are: BP, resting heart rate, morning PNN50 when exercising, CRP, LDL, Total Cholesteral:HDL, Apolipoprotein-B, Migraines, Total Cholesterol, and HbA1ac.

 

My kidney ones are: creatinine, urea, cystatin, potassium, albumin, bilirubin, albumin:globulim.

 

My cancer ones are: Cxbladder, NMP22, CA19-9, CEA, CRP, AST:ALT, Neutrophil:Lymphocytes, Hb:PLT, Basophils, Eosinophils, CRP: Albumin

 

I look to see whether I’ve improved on all of these over time, and have a combined score for each health/illness group (based on normalising each variable).

 

Please see attachment for graphs and tables.

Attached Files

•  Response to QuestForLife.docx   171.27KB   11 downloads

[albedo]

...

There was an interesting paper a while back that suggested that it was useful to look at the variance in a biomarker over a given period, as well as the time it took for a fluctuation in that BM to return to its long term average. Obviously the BM deteriorated over the long term (with aging), but the increase in variance and the increase in the time it took to return to baseline was quicker and could be more actionable. They combined results from whole blood count tests for their BM (in a way they didn't disclose), but also applied it later to step counts and found it equally applicable. 

 

I thought is was a nice way of capturing resiliency change over a sensible time period; rather than looking at daily fluctuations or very long term changes.

 

Nice, tks. Would you agree that this brings at least some similarity with the view of homeostasis, allostasis and “allostatic load” which might better capture a system dynamic adaptation to stressors and measure response to perturbation?

See e.g. Sholl J, Rattan. Biomarkers of health and healthy ageing from the outside-in. In: Moskalev A, editors. Biomarkers of human aging,
healthy ageing and longevity. Springer, Cham; 2019.

"Chapter 4
Biomarkers of Health and Healthy Ageing from the Outside-In
Jonathan Sholl and Suresh I. S. Rattan
Abstract Understanding the phenomenon of health is crucial for ageing research
since there is often an implicit view on what constitutes health and how to measure it.
We provide some reflections on how we might better understand and measure health,
discuss the basic biological principles of survival, ageing, age-related diseases and
eventual death, and end by tying these ideas together to rethink the nature of and
implications for healthy ageing.We defend a more positive view on health understood
in terms of various phenotypic parameters, such as robustness and resilience, and
show how it relates to the aim to support these parameters over time to achieve healthy
ageing. Together, these ideas suggest that the most effective strategies for healthy
ageing may come from the outside-in by altering those aspects of the environment that
prevent robust and resilient phenotypic responses and by supporting those aspects
which ensure such responses."

Edited by albedo, 05 December 2021 - 01:20 PM.

[Zarathrustra]

I have now looked at this, but, as you'll see from attached doc, it does not seem to add any useful action information compared with attempting to improve the base BM. That is, targeting reducing my systolic BP and resting HR has the somewhat incidental effect of also reducing the variance of these two. Attempting to use the variance as the targets for changing my lifestyle may detract from the more important ones of improving the base metrics.

Attached Files

•  image_2021-12-06_120814.png   349.66KB   0 downloads

[QuestforLife]

I have now looked at this, but, as you'll see from attached doc, it does not seem to add any useful action information compared with attempting to improve the base BM. That is, targeting reducing my systolic BP and resting HR has the somewhat incidental effect of also reducing the variance of these two. Attempting to use the variance as the targets for changing my lifestyle may detract from the more important ones of improving the base metrics.

 

That is interesting, thanks.

 

Your graphs are very zoomed out. I wonder if you took a shorter time interval, with (I am assuming) daily measurements of BP and HR - then you might see the variance as an early indicator of whether a change you had made was beneficial, i.e. before a long term trend in the BM itself was apparent. Just an idea. 

 

Another point: your results do seem to validate variance as a measure of health - and that it seems to be biomarker independent.

[Zarathrustra]

I'm not sure what you mean by 'zoomed out.

 

These are daily measurements.

 

I agree: the variance itself does indicate improvements in health. But I cannot see correlating them/targeting them would be better than that of the base metrics - systolic BP and resting HR.

[aribadabar]

I'm not sure what you mean by 'zoomed out.

 

 That they encompass too long a period while using a shorter one would pinpoint positive or negative trend much earlier/faster.

[Zarathrustra]

Thanks for the translation.

 

In tis case, I cannot see zooming in would be more informative. As I said, the base metric is, for me, more useful.

 

For many of my metrics, I do both - an overall picture from my whole time of 13+ years, and the last month or so.

[QuestforLife]

Thanks for the translation.

 

In tis case, I cannot see zooming in would be more informative. As I said, the base metric is, for me, more useful.

 

For many of my metrics, I do both - an overall picture from my whole time of 13+ years, and the last month or so.

 

I guess the advantage of looking at a second order metric like variance is that it could be metric independent.

 

For example, you have combined metrics in various categories like cardiovascular - I assume you have a reason for doing that? For example, if you did a correlation between all your metrics you'd find the ones you've grouped were more strongly correlated (I'd hope). So it might be possible to reduce down your metrics to a smaller number, providing the ones remaining were strongly correlated to the ones you have dropped.

 

If variance of a metric is not only correlated with its own specific metric, but others as well, you could just look at variance of one metric (or one metric from each of your categories). That is what the paper was looking at - they found it could all be captured that way. 

 

I'm not saying that would definitely work, but as you have the data, it would be worth trying. 

[johnhemming]

I have also been recording the basic metrics of BP and HR for now over 5 years.  At times when I go to bed my heart rate is under 50 and I have taken my BMI from 35 down to about 24.  I am, however, looking at other simple things to measure although I did have a set of bloods from earlier this year.

 

I don't know if I am able to post links, but if I am here is a photo.

https://twitter.com/...6441728/photo/1

 

 

[QuestforLife]

Well done for losing all the weight. It is depressing (to me) that we can clearly do things to make us healthier (for our age), but these things (like losing weight) seem to have negligible effect on our aging. That is why biomarkers are so important - hopefully we can find some simple metrics that correlate with actual aging. 

 

For example if we could correlate some everyday metric with methylation markers, that would be interesting. I have considerably increased my exercise endurance, without really doing any more exercise (than I was doing before). And this seems to be correlated with reductions in my methylation age. So this is promising. 

[johnhemming]

I think the weight thing was linked to taking far more Melatonin than most medics would advise.  That made losing weight a lot easier, but I still had to eat less food. (unsurprisingly)

 

I am looking for some easy practical tests of health that can be used to study changes and this is my current list (beyond weight, HR and BP)

https://www.westlake...kin-elasticity/

https://www.physio-p...t_To_Stand_Test

https://www.audiocheck.net/audiotests_frequencycheckhigh.php

Consider colour of gums – gingivitis test.

Whether people can rise from sitting on the floor to standing without using their hands (and how quickly).

To what extent people can go down onto their heels and restand.

Changes in hair colour and whether hair is regrowing where it previously was.

Changes in sleep patterns as measured on things like fitbits and apple watches.

https://www.telegrap...brain-age-test/

Grip strength – does require a device

 

What I think might be particularly useful is the high frequency hearing test.  That is a sign of hair health in the cochlear and hair health more generally is quite a good indicator of health.  It tends to be used as a proxy indicator.

 

Looking again at twitter:

https://twitter.com/...678861457432583

 

I am finding with my current protocol that hairs are growing on my scalp where previously there had not been hairs for about 20 years.  I am not particularly concerned about that either way, but I think it is a reasonably visible sign of an improvement at a cellular level.  However, I really need to do some testing on inflammation and NAD levels.  Inflammation can be seen on the gums although they are a particularly variable location it can also be seen on the skin.  NAD I think requires some form of blood test or biopsy.  The thing is that although health is, of course, crucial I am doing quite a lot of things and don't really want too many disruptions I cannot easily fit into my daily routine.  Hence a bit of hyperoxia is quite easy as I can do that whilst working on other things (or typing into online fora).

 

 

 

 

Edited by johnhemming, 10 December 2021 - 11:57 AM.

[Zarathrustra]

QuestForLife,

 

Whilst I get what you are hoping for - a single metric for any particular degenerative condition - until there is a proven one, I believe my approach is the better one. Any one of mine could be used - such as resting heart rate - but it is not known if that or any one is enough to assess the cardio-vascular system. Hence my use of several. This is the state of the quest for a single biological age metric and why I use my approach too.

 

As to using variance as a dimensionless metric, that is indeed attractive. Alas, for me, only a couple of mine lend themselves to this - those which I track daily. Those done far less frequently have inadequate numbers for this approach.

[Zarathrustra]

John Hemming,

 

Thanks for your posts. I’ll take each in turn, if I may.

 

Wow, your weight loss photo is really great. And much needed. The BMI has been found wanting (some athletes have high BMI, which doesn’t make sense). Hence waist:height is considered superior, below 0.5 being considered good. In the end, it is whether your visceral fat is low – I use a body composition monitor to determine this. However, one can be too thin/low fat. My weight came down most as I became more vegetarian/vegan (apart from when I had chemo).

 

As to resting heart-rate, I use the fitbit tracker’s measure – I suspect any similar method would work, as long as it is consistent. Below 50 is very good (assuming your blood pressure is OK); into the athlete territory.

 

I’m with you on the ageing tests, although many of them are difficult to provide a replicatable metric. The skin-elasticity one, gum colour, etc. I note my own results for some of these, but don’t use them for my degenerative ageing collation. There is a new Alzheimer’s test: https://wexnermedica...s/sage#SAGETest, which you may wish to explore.

 

I do track my sleep, considering the time in deep sleep being of particular importance.

 

I track my grip test, and, having found it poor, I do a grip exercise – which has improved my grip about 12%.

 

My hearing is OK, but with tinnitus, I just live with it. It doesn’t seem to be getting worse.

 

I track things like inflammation but regular blood tests, and having had it go way high (50), it is now down to about 1. Ginger, fruit, and arginine drive this down the most for me.

 

I have used hyperoxia at times, but not now as it correlates badly with some of my metrics – especially for my CKD.

 

[Zarathrustra]

QuestForLife – second post today

 

I’m with you on the uncertainty of knowing whether what we do improves our biological age. Whilst methylation seems to have a lot going for it, to my knowledge it is still not universally accepted as an unambiguous indicator. The other problem, for me, is that it is somewhat expensive. So on those two grounds, I pass on this.

 

As I’ve posted before, it seems to me that in the absence of a proven ageing metric, then the equally important ones of resting heart rate (RHR), BP, grip-strength, etc are adequate for now – if not superior (that is, if, say, a methylation metric gave a result that I was getting younger but my RHR, grip-strength, etc were getting worse, then I’d go with the latter as a better indicator of my likelihood of living longer.

 

All that said, your improved endurance looks good in itself, regardless of your methylation status.

 

[johnhemming]

It seems clear that there are a number of ways in which cells deteriorate.

 

For example for a non-exhaustive list

 

There is the methylation of DNA

Then there is the availability of stem cells and length of telomeres

There is additionally chronic inflammation which kicks of CD38 and puts pressure on NAD+ availability and energy availability  (and the knock on effect on sirtuins)

Additionally there is the calcification of soft tissue and the deterioration of bone structure.

 

I am not myself persuaded that all of the above different types of deterioration can be fixed by one intervention.  I am using a range of interventions in various things.  Logically I would wish to measure them separately, but that can be quite hard to do although I am experimenting with a methylation test.

 

I think there is good evidence for Melatonin and things like the TRIIM trials (which have their similarities), but I think that can deal with Methylation and the inflammation.  I think varoxia is needed to deal with stem cells and telomeres. (and angiogenesis if that is needed). Menaquinones help with Calcium although again Melatonin helps with that as well.

[Zarathrustra]

I agree with much of what you write John. Hence my own way of measuring myself in many ways - over one hundred, though I only use about 33 for daily correlations.

 

As I mentioned, until there is unarguable proof of many of the cellular interventions, I'll mainly stick to the already proven ones - such as LDL, CRP, etc.

 

I have tried some of the touted supplements for slowing/reversing ageing. Melatonin correlates poorly for some of my metrics (Resting HR, neutrophil:lymphocyte). Fisetin seemed to have no effect. Nicotinomide Riboside similarly gave mixed results. 

 

As against those, my Resting HR, hand-grip, etc have generally improved - so I think I'm on the right track. I suppose I'll only be sure if I stay alive another thirty years or so - tho' the odds are against it. But 85 isn't bad.

[johnhemming]

Melatonin only works if you end up not undermining your endogenous Melatonin.  Practically you need to take it at stages during the night.  You also need to avoid kicking off the Cortisol Awakening Response unless you want it. I have a view that as soon as the blood serum levels of Melatonin start going down sustantially the CAR starts and you wake up.   That means just popping a pill is probably not going to help.

 

It depends upon how good sleep is.  You can take it when you wake up or just before that.

[QuestforLife]

Melatonin only works if you end up not undermining your endogenous Melatonin. Practically you need to take it at stages during the night. You also need to avoid kicking off the Cortisol Awakening Response unless you want it. I have a view that as soon as the blood serum levels of Melatonin start going down sustantially the CAR starts and you wake up. That means just popping a pill is probably not going to help.

It depends upon how good sleep is. You can take it when you wake up or just before that.

So how do you avoid reducing endogenous melatonin production? I have an interest in this as my son is autistic and struggles to go to sleep without a low dose melatonin supplement.

Also not clear on what you mean regarding when you dose (before you wake up?)

[Zarathrustra]

My sleep is OK, even good. And with the poor correlations with my other biometrics, I feel I don't need melatonin.

[johnhemming]

>So how do you avoid reducing endogenous melatonin production?

 

If people don't wake up then a slow release melatonin might help.  It is generally older people that are short on melatonin.   I wake up a lot during the night and take it then.

 

However, generally what I am talking about is taking melatonin when people wake up during the night.