Slashdot Mirror
Machine Learning Used To Predict Military Suicides
HughPickens.com (3830033) writes David Wagner writes that a predictive computer model using machine learning methods is helping to identify soldiers in the United States Army most likely to commit suicide. Computers combed through data on more than 40,000 soldiers who'd been hospitalized for mental health problems looking at 421 variables on each soldier drawn from 38 military data systems. Using a method known as "machine learning," the researchers identified roughly two dozen factors that are most important in predicting soldiers most likely to commit suicide. The soldiers most likely to take their own lives were men with past suicidal behavior and a history of psychiatric disorders and criminal offenses, including weapons possession and verbal assaults. Soldiers with hearing loss also faced heightened risk — a strong indicator that they had suffered a head injury. So did enlisting in the Army after age 27, most likely because those soldiers had already experienced trouble finding their way in life. "There's this group that comes to the Army later in life — they're smart, they have skills, they tend not to be married and they have no career or have left a career to join," Dr. Kessler said. "We don't know why they should be at higher risk, but they appear to be."
Murray Stein, co-author of the new study, found that among soldiers recently discharged from psychiatric hospitals, more than half of suicides were committed by just five percent of patients. "The most impressive thing is that they identified this high-risk group in the hospital, and by just focusing on one in 20 of them, you're really dramatically improving your ability to predict," says Dr. Mark Olfson, a professor of psychiatry at Columbia University who was not involved in the study. "Clinicians don't do a very good job predicting suicide risk, even though we think we do."
Murray Stein, co-author of the new study, found that among soldiers recently discharged from psychiatric hospitals, more than half of suicides were committed by just five percent of patients. "The most impressive thing is that they identified this high-risk group in the hospital, and by just focusing on one in 20 of them, you're really dramatically improving your ability to predict," says Dr. Mark Olfson, a professor of psychiatry at Columbia University who was not involved in the study. "Clinicians don't do a very good job predicting suicide risk, even though we think we do."
The soldiers most likely to take their own lives were men with past suicidal behavior
More surprising tales from the site that gives you news for nerds, and US government propaganda.
"...more than half of suicides were committed by just five percent of patients."
The proportion that weren't committed by patients are referred to as "homicides".
Political debates have me rolling my eyes so much I think I got optical whiplash. I should sue. - Foamy The Squirrel
If "more than half of suicides were committed by just five percent of patients", then all the suicides were committed by just ten percent of patients. Assuming at most one suicide per patient.
Why doesn't machine learning predict suicides any longer?
Stop speaking of machine learning as if it's a new kind of black magic. I know it sounds better than "using a mathematical algorithm" or "performed statistical analysis", but to me it sounds as ridiculous as the "quantum whatever" of the 90's. Seriously, ML is being hyped beyond reasonable.
Video of some good progressive thrash music
Some of the factors associated with higher risk were expected, like previous suicide attempts, a history of using weapons and symptoms of severe traumatic brain injury, such as hearing loss. Others were less so, like a higher I.Q., and being older than 26 at enlistment.
"There's this group that comes to the Army later in life — they're smart, they have skills, they tend not to be married and they have no career or have left a career to join," Dr. Kessler said. "We don't know why they should be at higher risk, but they appear to be."
Isn't that sort of obvious?
When I look at the world today . . . I see Obama threatening to use dictatorial powers . . . which will cause the Republicans in Congress to shut the government down . . . while ISIS is making a mighty romp . . . and no Muslim countries seemed to be concerned about that . . . Putin, is, while just being Putin.
And, and and more and. I'm trying hard do decide why I am not committing suicide myself.
Oh, I need to teach a class at TU Darmstadt in a couple of weeks, and need to create a presentation. Never mind.
Schroedinger's Brexit: The UK is both in and out of the EU at the same time!
JavaScript includes Java
No wonder you posted that garbage anonymously.
"Transparent" is a shit show that trades on every stereotype going. A man in drag is NOT a transsexual.
If someone's been searching for PTSD, Major Depressive Disorder and Suicidal ideation, maybe they're at risk?
People at risk tend to isolate themselves during bouts of depression (not just "the blues"), so they turn to the internet.
"Transparent" is a shit show that trades on every stereotype going. A man in drag is NOT a transsexual.
Verbal assaults? Oh no, Cpl Wilson yelled at me and called me names. True, I shat in his kit bag for seven days running, but he's clearly suicidal for raising his voice. Plus he hurt my feelings so he's a sociopath as well. Best lock him up before he hurts someone.
Fucking pussies.
"Wait. Something's happening. It's opening up! My God, it's full of apricots!"
"Soldiers with hearing loss also faced heightened risk — a strong indicator that they had suffered a head injury."
Wouldnt head injury/tbi already be one of the tracked stats? Maybe they're writing off how sucky hearing loss can be. The constant unrelenting ringing not only can take over a persons world, its also painful and very discomforting, especially if its on only one side. Its not just a Neeeee, to a lot of people it sounds like a modem chirping or like a "sssssss" steam valve constantly, or a bunch of dissonant tones. Play two notes on a piano immediately next to each other & imagine that in one ear forever (try it with headphones).
People without bad tinnitus dont know how distressing it can be. Its not like regular sound, the noise is more direct than anything you can experience acoustically,
Once corrected for age demographics (which people tallying raw numbers usually forget to do), the suicide rate in US military is lower than civilian population. Rather impressive for organization whose purpose is to kill, maim and blow up shit.
Instead of using computers to figure out who's most likely to take their own lives, how about we use machine learning to figure out why these people are killing themselves, and fix the actual problem?
An enigma, wrapped in a riddle, shrouded in bacon and cheese
It would seem that some profit taking by a psychiatrist might benefit yourself as well.
Faster! Faster! Faster would be better!
Wo-0-O-0-o-0-O-0-o-0-O-0-o-0-O-0-o-0-osh
JavaScript includes Java
JavaScript = Java + Script
QED
Filter error: Too much repetition.
Your comment violated the "postersubj" compression filter. Try less whitespace and/or less repetition in the subject line.
Unironic captcha: drunker
Why should it be illegal to commit suicide?
I don't feel that if someone wishes to take their life we have a right to stop them. It is their existence, their experience, their decision, their right. We should not encourage it, but if we want to reduce the number of people who do it, then help them with their problems, don't go hunting them down, and locking them up in white room - mattress walled asylums in straight jackets, doped up on who knows what.
It's dangerous to allow the government the ability to predict, and then potentially take action based on it, far more dangerous than simply leaving people alone, even should they choose that awful path. People are creative, if they really want to leave this world, they will.
I art more snarky, and terse than thou. I art Slashdot!
How did this get a score of 1? All the comment did was take a statistical conclusion drawn from data, create a pretend monologue for an imaginary person using this scientific conclusion as the basis, and finally insults said imaginary person(s) for... falsely accusing a second imaginary person? It's literally nonsense. He insulted his own fictional character.
Please stop bullying suicidal people. That may not have been what you meant, but it's definitely what you did.
Or alcohol.
Thank you, NRA and Congress, for giving our brave service men and women the freedom and personal responsibility to kill themselves when things get too tough.
http://touch.latimes.com/#sect...
Programs to prevent psychological problems in troops questioned
By Alan Zarembo
February 20, 2014, 7:34 p.m.
Many federal programs aimed at preventing psychological problems in military service members and their families have not been evaluated correctly to determine whether they are working and are not supported by science, says a new report commissioned by the Defense Department.
"A lot of their programs don’t have any good data behind them," said Kenneth Warner, a professor of public health at the University of Michigan who led the Institute of Medicine committee that produced the report. "We remain uncertain about which approaches work and which ones are ineffective."
At the same time, some proven interventions are not being used, the committee found. Researchers said there was ample evidence to suggest that limiting access to personal firearms on military bases would reduce suicides. About 60% of service members who take their own lives do it with guns — usually their own.
"Means restriction has been demonstrated to work," said David Rudd, a psychologist and suicide expert at the University of Memphis who served on the committee.
But in 2011, Congress prohibited the Defense Department from regulating legally owned personal firearms and ammunition on military bases.
http://iom.edu/Reports/2014/Pr...
Preventing Psychological Disorders in Service Members and Their Families: An Assessment of Programs
February 20, 2014
Among the small number of DOD-sponsored reintegration programs that exist, none appears to be based on scientific evidence. The committee was unable to identify any DOD evidence-based programs addressing the prevention of domestic abuse. More recently, the services have implemented a number of prevention interventions to address military sexual assault, yet a DOD review found that critical evaluation components needed to measure their effectiveness are missing.
The committee also found that environmental strategies with strong evidence of effectiveness are underutilized, such as restricting access to lethal means such as personal firearms to prevent suicide or homicide in domestic violence cases or placing restrictions on the sale of alcohol to reduce substance misuse.
In place of these proven approaches, the committee typically found interventions such as campaigns, Internet tools, or in-person events with no evidence for their effectiveness at preventing the targeted problem.
fuck everything. there is no point...
Machine learning is its own field of research and not just a method. That sentence is as informative as "Google uses machines known as "computers" to make money"
I graduated university with a BS in CS in 2008, right into the recession. Bam. I had been previously diagnosed with clinical depression, and it never really improved (part of the reason it took me 5.5 years instead of the normal 4.5 for my university). My folks lived in rural Iowa and I lived with them a bit while job hunting, but was unable to find anything. I then moved into a relative's home outside of Denver, but after six months there I still had no prospects. With my (rather large) student loan payments starting up soon, I took what I thought was my only option, joining the military.
The major reason for doing so was the steady paycheck and Student Loan Repayment program, but I had a secondary goal of finding some purpose in life (or just let someone order me around for most of it) and getting self-discipline and confidence, both of which I lacked in spades. For some stupid reason I decided to be 92G, Food Service Specialist, signed in 2009, and went to BCT in 2010. I was 25. After graduating that and AIT, I was assigned to a unit that would test out new gear/tactics. Prior to joining, I had bought into the line of our military being the best in the world. I thought this meant that the soldiers would be excellent people. After spending about 8 months with my unit, I found out that this was the furthest thing from the truth: the Army I saw as full of soldiers, both enlisted and commissioned, that were some of the dumbest, most selfish, unorganized, and petty individuals I have ever had the displeasure of working with.[1] Any attempt I made to improve things I could affect, even if small, was often met with derision or contempt no matter how I tried to address it.
From what I experienced, we were only the best fighting force in the world only because all the others were somehow even worse than us. Everything I thought I knew was wrong, I lost all faith in humanity, and eventually all hope in myself. After a few weeks of dealing with a heavy amount of this, then learning that a potential change in MOS would take years (if at all, as the military was starting to downsize), I cracked. I tried to go AWOL with the intent of killing myself soon thereafter, but thankfully some relatives talked me out of it and I was admitted to the local military psych ward by my sergeants. I spent three days there, got some new medication, and returned to my unit; 10 months later I was discharged for Failure to Adapt (Honorable Discharge; the official reason was failure to maintain physical fitness, which was true, but everyone involved knew otherwise.)
So, to answer the question, I think it comes from a lack of purpose and direction being combined with inept leadership. People who join that late in life do so with a partial hope that the military will fill that role, either directly or helping them find it. But, like me, I bet many find that their expectations were far too high and it completely destroys them to the point that they want to destroy themselves. If they had purpose and direction, knowing they would one day be able to get out under the inept leadership, or had no direction but good leadership, they would probably manage, but lacking both is just too heavy on the soul.
I'm curious if there's any correlation between MOS and suicide rate; couldn't find mention of it in either article I looked at.
[1] Not that there aren't good people in the military; there are, but I encountered them few and far between. And almost none of them were in a position of heavy influence for our little corner of the military.
The problem REALLY is not that clinicians are bad at predictions, the real problem is the near-total lack of access to effective treatment and the range of treatment needed for seriously ill people. The risk factors for suicide are well-known, and have been for a long time. But focusing on some factors and only 5% of people who are suicidal will lead to not treating the others--when all of them are in great need of effective treatment.
The truly great scandal is that the US does not have an effective mental health system; it hasn't had one since the 1980s.
The vibrant US mental healthcare system (which offered long-term and all other treatment options) was destroyed by Managed Care in the 1980s. The US had been spending 8-10% of the healthcare dollar on mental illness until then. Managed Care "managed" mental health treatment out of existence by lowering the percentage spent on mental health and addiction to 1-2% of the healthcare dollar. Today Diabetes care gets 8-10% of the healthcare dollar according to the NYTimes.
Now the largest psychiatric hospitals in the nation are the LA and Cook County (Chicago) jails. We are criminalizing the mentally ill, to the detriment of all. In Austin TX, 75% of people jailed have serious mental health or addiction problems. A staff of 9 psychiatrists is on call to evaluate them, but our local level IV public hospital has no psychiatric beds for evaluation or treatment. And there is little capacity for private inpatient care or treatment in Austin. To get state-funded inpatient psychiatric treatment in TX you have to be committed involuntarily; the state MH system is so underfunded that they refuse to accept people who want to be admitted voluntarily.
The nation is greatly harmed by the lack of an effective mental health system, 75% of US families are affected by Depression, addiction, or child abuse. This is a tragedy of great proportions.
Deborah C. Peel, MD
Boarded Adult Psychiatrist and Freudian Psychoanalyst
Founder and Chair of Patient Privacy Rights