Human Beings Are Not Computers

I find a lot of things I like in the New York Times. This article resonated with me as few others. First, there is the purpose of the human profiled.  Changing medicine into data science?  God save us all.

Sometimes I feel the best thing I do for a patient is to be human.  Just to have the pretension (a pretension which I do not take lightly) of being one human being in a room with another human being, trying to make them feel better.  This does more, I think, to make most of my patients “better” than all of the pills I have spent years studying about. All those years studying normative use of medications on large populations of humans.  And they work enough to please the powers that be.

There are some  people doing “meta-analysis.”  They generally find things, throwing together research studies, that might be bringing us closer to an absolute truth. About 1/3 of depressed people can get better with a single antidepressant.  About 1/3 seem to get better with the “placebo affect.”  Even larger percentages get better in studies of anxiety.  Right now, there is a lively debate in the literature about whether people can get better with a placebo even when they know it is a placebo.

Whether or not you are convinced by any of the above, they are enough evidence to contest the fact that medicine is, or should become “data science.” Things like mechanical engineering and perhaps theoretical physics may be data science, but the human sciences, like medicine, to which I have deliberately decided to devote my life instead of “hard” sciences like physics, decidedly are not.

Perhaps the most blatant example among patients whom I have treated was a precious young Mexican immigrant couple, whom I interviewed in Spanish.  Both of them were on tiny daily doses of lorazepam, brand name “Ativan,” a medicine like Xanax, maybe a little less addictive, but still in that class of “benzodiazepines,” anti-anxiety drugs. Their problem had been “depression.” Benzodiazepines usually make this worse.  But they said it had cured them and they were asymptomatic. They had a life of grinding poverty in Mexico.  They knew life could only get better in the States.  They were very happy and proud that they had been able to make it to the states.  America was wonderful and American Medicine was wonderful, and the American medicine had cured them, so life was perfect and they were both happy and asymptomatic with the great American medicine.

I was not stupid enough to take them off it and give them antidepressants.  They were well and without complaint, so I kept them on the same medicine.  I saw them a month or so later and they were still fine. This is not, and should not be made into data science.  In the moment that I saw these patients, all of the careful data taken by scientists who had spent many years researching minutiae and disease classifications of anxiety and depression collapsed into nothingness.  All that remained was the person and the illness and something higher. That “higher” thing is the force of belief.  I use a lot of “Emotional Freedom Technique,” or “EFT” with patients.  I ask my patients how they believe the world works.  I will go to and try to think like whatever their belief system is.  It can be Buddhism” or spirits in trees reminiscent of Druid beliefs or things I have not heard of.

Most of the time people tell me they are Christian.  I ask them if they think Christ can work through a crazy old lady Jewish psychiatrist.  They generally say “yes.” Then I tell them what my grandmother said, even though she did not seem to know she was kind of quoting George Bernard Shaw at the time. “The God you believe in cures everything.  Doctors just collect the money.” Of course it is now all governments and insurances but I usually do not linger on that.

Then they join me in tapping on a simple, easy to repeat sequence of acupuncture points, forgiving or detaching from who or whatever caused their troubles and saying “I can do all things through Christ who strengthens me,” and they start saying what is on their mind, kind of like Freudian free association at warp speed, and they start feeling better. They learn how to do it and find videos and references on the internet and keep notebooks and repeat the sequence and the insights by themselves and generally get better. It’s not just pills and its not just computer collected data.  It is people with broken souls, who may never in their lives have reported a trauma or abuse that might seem so slight to be subtle or laughable, but that seared a hole in somebody’s soul when a person in authority said something like “never tell anybody about this…”

Healing comes when these issues are brought to the fore. It changes neurotransmitters as surely as any drug ever invented. It would be easy if this were the only thing that bothered me in the profile in this New York Times article. The young man who wants to change medicine into data science was diagnosed bipolar. He had been using drugs.  Drug abusers can have and mimic every symptom of bipolar illness.  Depressants make a person depressed.  Stimulants make a person manic. I doubt these are the same people who have hereditary bipolar illness.  For that illness, there is a “final common pathway,” a consistent DNA sequence. It seems that the only people who are curious about their DNA sequences are some kind of super “body hackers” who are more worried about whether insurance companies are going to use this information to increase their premiums than to get their life prognosis.

The question with a patient like this is what kind of bipolar illness we are dealing with.  Is it genetic?  You can guess from family history, confirm with DNA testing.  I have yet to hear from an insurance plan that pays for this. This kind of bipolar seems to burn out in the 50’s, so somebody could need medications to stay out of manic and depressive episodes. If someone’s bipolar illness is “drug-induced,” and this is clinically indistinguishable from the other kind, “the genetic.”  But if it is drug induced, it ought to treated with sobriety, and is likely to burn out much faster, although it is not possible with data I have reviewed to predict exactly when.

In many quarters, “drug experimentation” is looked upon as a rite of passage.  It affects brain cells in ways similar to serious mental illness that could quite easily land someone in the hospital, or surprisingly easily, in the morgue. Medical science has its share of human factors, but psychiatry is the head-on interface of science with human factors.  Make me practice with computer-generated data to satisfy a payer makes helping a psychiatric patient more difficult, not more easy.

I am not stupid enough to throw out the computer completely, for costs and their consideration already dominate the system which I am reluctantly locked into by being a physician.  Use previously existing and validated scales, previously studied analyses of human factors, to make and computer work together within the context of human truths already determined.