Slashdot Mirror
New Brain Scans Can Spot PTSD
Neuroscientists think they may have found a scientific method to identify post-traumatic stress disorder (PTSD) using a brain imaging method called magnetoencephalography (MEG). In the test study, the scientists studied 74 vets with PTSD and 250 civilians without and were able to spot the PTSD sufferers with 90% accuracy. "MEG machines are a fast, sensitive and accurate way to measure electric activity in the brain. Whereas CT scans and MRIs record brain signals every few seconds, MEGs can do it by the millisecond, catching biomarkers and brain activity that the other tests inevitably miss. The study could be a breakthrough for the military, who've been scrambling to address a surge in post-traumatic symptoms among newly returning vets. Right now, troops are evaluated by mental health experts, but diagnosis is a crap-shoot: symptoms can take years to show up, and vary from person to person, even among those exposed to the same traumas. The Pentagon's already been pushing for more objective, systematized diagnosis tools, like portable at-home sleep monitors and genetic testing to detect PTSD vulnerability. They've even launched a program to create stress-mitigating pharmaceuticals."
"scientists studied 74 vets with PTSD and 250 civilians "
Is it possible that they aren't spotting PTSD but a wiring from a soldier? I am a civillian with NO military experience, but I do hang around several soldiers and police officers. Each group has similar mannerisms and they have ALL had similar experiences within that group (basic training for the green guys, the academy for the blue ones). I see a good chance that this new scan could be picking that up.
A more valid group would be:
- some vets without PTSD
- some vets with PTSD
- some civillians with PTSD
- some civillians without PTSD
Of those four groups some significant correlation would be helpful too. For example a set of soldiers from Afghanistan with and without PTSD. A set of civillians that had been through the same or similar trauma (say armed robbery or 9/11 or plane accident). Breaking it down by age would also be useful, a Vietnam veteran who has had a few years either with or without treatment would be a lot different than a recent return home from the Sandbox.
IF they can really achieve 90% detection rate (and that's a huge if), then that'd be hugely impressive.
FYI, Mammograms only have roughly an 80% chance of detecting breast cancer. http://breastscreening.cancer.gov/data/performance/diagnostic/rate_age_time.html
Rates of detection of other cancers are often much worse. And rates of early detection (as opposed to self-report) of mental illnesses are worse again.
Anybody who would use the test, in its current form, as an automated "confirm/deny" mechanism deserves a punch in the face. However, the establishment of a fairly reliable correlation between a psychological condition that could previously only be diagnosed subjectively("subjectively" in the sense that the patient has to introspect and report symptoms for a diagnosis to be made, not in the "Oh, its just subjective, it must just be you" sense) and an objectively measurable electrical phenomenon is pretty cool.
It is more or less a commonplace, for anybody not deeply in the grips of some metaphysical or dualist theory, that psychological phenomena must have material correlates; but actually hunting them down and making them useful is fairly tricky work.
Not for nothing, but I went 20 years without being diagnosed. I understand the attempt at humor, but it isn't funny to me.
PTSD is not just some dude screaming "incoming" and diving under the table at Thanksgiving. It has a measurable effect upon every decision in a person's life. Just stepping into a restaurant is a tense and fearful experience that can last for hours. Hours in which I must scan every person coming into the room for potential threats, monitor the exits, sit so that no one can approach me without my knowing. My family knows that to catch my unawares is, at the very best, a chance to have me yell at them as if they'd done something wrong by entering the same room without announcing themselves.
When I head into wallmart, I have to plan my route to minimize the stress, I use weird checkout methods, like buying a box of vitamins and then doing my checkout at the pharmacy to avoid the lines where I am vulnerable to attack. For 20 years I thought that everyone looked between the parked cars as they walked down the street, planned the move to cover in case there was gunfire, looked at every window and rooftop for snipers. I didn't realize that what I took for survival instinct was way beyond what almost everyone else did to safe guard themselves. I have been emotionally removed from my daughter's entire life, I have no emotional reaction to the suffering of others, as I instinctively believe that it is their fault for not being ready to deal with whatever the situation.
If this technology can help get people diagnosed and in a proper treatment regimen, then it is a worth while venture. A 10% miss beats 40% (http://ajol.info/index.php/ajpsy/article/viewFile/30263/30480)[pdf]
I for one, salute our new MEG Overloards
To launch headlong into the most expensive, unworkable, unreasonable, ill-thought, entirely not-researched solutions. I'm not talking so much about the brain scan here; Harvard came up with the idea that PTSD could be detected in the brain, along with bi polar disorder and a few other conditions which might have detectable pathologies, and it's a very good idea.
What's a very bad idea is the notion that PTSD can be stopped, or at least mitigated with chemical or mechanical tools. Once you already have PTSD, or indeed any disorder on the dissociative scale (Howell, Chu), then there is some good evidence that pharmaceuticals can help mitigate some, but not all of the symptoms. Currently those pharmaceuticals fall into two major classes: sedative-hypnotics and atypical antipsychotics. Sedative-hypnotics, particularly benzodiazepines, cause massive problems with the creation of short term memories. Atypical antipsychotics have a host of horrible side effects, from flattened affect to tardive dyskenisia-- which is the permanent, uncontrolled flexing of small muscles, like facial tics, thumb-wiggles and circles, and shuffling gait. In short, they should not be used for any condition which is not treatable by any other means.
Also, there is no evidence (as has been noted by Harvard, at least) that there is any sort of genetic pre-disposition to the development of PTSD, or any other environmentally caused dissociative disorder. That is a dead end.
Furthermore, there is also absolutely no evidence that the pre-dosing of atypical antipsychotics or sedatives have any effect whatsoever on the development of PTSD, and in fact in both cases may very probably result in soldiers with far less willingness to shoot the enemy in the face than the military requires.
The fact is that the Pentagon is and has always created an enormous mess out of the minds that manage to survive their plans, and there's simply no way around it. It would be really great, I think, to come up with some way to make war more palatable for the people who have to be in it, but somehow the very root of the notion seems disingenuous.
War is fucking hell.
Sure, My university has a subscription. Let me give you the numbers:
Test subjects:
All 75 test group members had confirmed PTSD as the primary diagnosis, using the standard structured clinical interviews for PTSD. There were many variations as to cause of PTSD some from combat others from before they became soldiers. 69 test subjects were male, and 5 were women.
Control group:
250 members from the general public in the same age range as the test subjects. 151 men, 49 women.
Complete nurological histories, and multiple interview examinations were performed to help exclude general public members with latent PTSD.
The test with the paramters used by the team had the following results:
72 true positives.
2 false negatives.
31 false positives.
219 true negatives.
According to the paper this is .001 .765
97.3% Sensitivity
87.6% Specificity
92.4% Accuracy
Chi squared-statistic: 189.8
P value:
phi coefficient:
odds ratio: 254.3
Stylish sheet to fix many problems in Slashdot's D3: https://gist.github.com/801524