Friday, October 26, 2012

Human MTBF

A friend asked me the other day, “why do you write?” And I got to thinking, why do I write? The answer was simple, “because it gives me pleasure.”

So I sat down to write me a story and there was nothing. The words just did not want to write themselves. The story wasn’t there. Nothing flowed. It was a series of one word followed by delete. No writer’s block, mind you, just blah.

Figuring that the limited mojo had gone out for a walk, I started with this…


(2 minutes later)


(5 minutes later)

You know the triple aboriginal thoughts of “Oh my!” “Now what?” and “We’ll see,” happened all at once. The chaotic migratory pattern of confusion has many a focus of blinding lights that are streamlets of productivity. So, I latched on to one.

What if?


What if, there was a predictor of loss of function? You know, like the one called MTBF (Mean Time Between Failure) in machines, only associated with the human. Now that would be nice, wouldn’t it? Say, the average MTBF of a modern day American is at 78 years. We could then use Baysean logic and find out which elements of us will go awry and cause an early departure from the land of the living. 

That, of course would be predicated on previously known and compiled human data and tailored to one’s deeds as a human. Say, if one were sedentary, drank a lot of soda and ate bonbons all day and it was determined that in many such individuals, this behavior caused Diabetes and through that conduit, high blood pressure and thence a stroke, followed by death from a heart attack, well that information would be a determinant of the expected longevity in your case if you were into those things, now wouldn’t it? Of course it would.


Now suppose we added another wrinkle in the form of smoking a cigarette, well then, the historical data would suggest a lower number of smokers surviving till 78 years of age because of infirmity from emphysema, chronic bronchitis and cancer. This predicate would be equally valid to reduce the likelihood of survival to the 79th year. 

And if you were to add some fine wine, or just potato released Vodka to the mix, a head and neck cancer would be the potential cause of death in many an individual. That would lead to a reduced chance of living for an additional year too- thus higher morbidity and mortality. For another rung to this ladder of survival complexity, we could add a chronic, unrecognized HPV (Human Papiloma Virus) infection in the throat, and the whole drama would certainly be played out much earlier in a lifetime. The HPV would be propelled into a state of furor, not Fuhrer ~ although at this point they are both the same damage causing agents. The inverse ratio lives; the more the number of diseases, the lesser the chances of living longer. And to a large extent that is predicated on human choice and lifestyle.

Null Hypothesis and the p-values:

In circumstances of many diseases, the rigid Algorithm of the calculator under these circumstances would suggest, the possibility, that survival at 78 years, although remote, is possible but beyond the calculated realm of mathematical reality. The past always catches up to the present eventually and limits the future. A meaningful number of p-value at 0.04, negating the null hypothesis (that this was not the norm) gives little comfort to the receiver (Only 95% of the tested population lives inside the bell curve-depending of course on the selected criteria). Essentially negating the Null Hypothesis states that that this person’s chance of survival is limited beyond "a coincidence" and the forces of his past behavior are against him (this probability is based on the cumulative data from a large cohort of individuals with similar behaviors and life-styles in the past).

Now I know what you are thinking. This writer (me) has lost it. He has gone mad. Yes, you are certainly entitled to your opinion, but guess what, the words have started flowing, and you my friend have arrived here with me. Haven't you?


We actually do have the Actuarial data to determine future outcomes of human beings. The Insurers use it all the time, but it is not that accurate or detailed ~ although it remains the best so far. What might and may already be happening as I write and you read this, IBM Watson programmers are shoveling this data, in spades, into the machine and writing billions of lines of codes for the software algorithm to project and predict the future. It is not to say that they will be absolutely accurate, but it will be an improvement.

And don't forget the genetics:

But, you say, well that may be true, yet my grandfather died at age ninety and he used to drink and smoke and was obese, how can your projections or those of the future Watsons be accurate? Ah, my dear friend, you are correct in your skepticism. There is the genetic code issue that we have left out. 

from Lew

Soon the GWAS and the assays of various Oncogenic and Suppressor gene data along with genetic and externally modulated epigenetic predictors for cardiovascular, diabetes and all other diseases, will be accrued, assimilated and digested for improved prediction. Watson of the near future will be able to mix and match and eventually beep a better digital result of survival probability ~ maybe even with a Confidence Interval (CI) of 97.5%.

“Doubt thou the stars are fire,
Doubt that the sun doth move…”

But doubt not the intent of human ingenuity and purpose.

Probability and Eventuality:

Remember probability determines the likelihood of an eventuality and not the eventuality itself. So yes, some may dodge the single bullet in the game of Russian Roulette and unhealthy living, but adding more bullets into the chamber will certainly increase the chances of a mistimed “bang!”

Russian Roulette:

So what if you empty the chamber of the gun; don’t smoke (one less bullet), don’t drink (one less bullet), don’t eat in excess (one less bullet), exercise (one less bullet) then that will add to the age to the result. Even if you happen to have a genetic anomaly on one of the pair of the chromosomes (allele), well, you will live out your `90-100 years and never know it, because you did not challenge the remaining normal (good) gene with bad habits. (It takes two mutated alleles to tango). 

50 minutes later… we have a story. I think?

What if … you live a healthier life style?

Live Long, Live Well, Be Prosperous!

No comments:

Post a Comment