Friday, November 19, 2010
(Rough Draft. I would appreciate any input.)
Would TSA be a good medical screening test?
As a family practice physician, every year I have a few patients ask me to order a whole body MRI, "just to make sure I don't have any cancer". I inform them that I don't recommend these tests for otherwise healthy people. This invariably leads to a long discussion about what makes a good screening test in medicine. Something every medical student learns is that a good screening test needs to be sensitive, specific, cost effective, safe, and widely available. A whole body MRI does not meet most of these criteria.
Being a frequent flier, I have recently been preoccupied with the increasingly invasive of "security" at our airports. This got me wondering - would the TSA's new security measures meet the criteria for a good medical screening test? If terrorism was a disease, and the new virtual strip searches and crotch grabbing pat-downs were a test for that disease, would any of the medical associations recommend it for universal screening. In order to determine this, let me briefly review what makes a good screening test.
The first attribute of a good screening test is that it has to be sensitive for what it is testing for. In other words, if I you have the disease a sensitive test will come out positive (not many false negatives). Although the new naked body scans and "third base pat-downs" may appear extremely sensitive, there is no objective evidence that they could have actually identified even the "underwear bomber". Also, given the fact that since it's inception 9 years ago, TSA has identified and prevented exactly zero terrorist attacks on planes, their tract record is 0% sensitivity.
Secondly, a screening test has to be specific. When a specific test is positive, the person really has the disease (not many false positives). So far, TSA has also been very unspecific. Every person they have frisked, illegally detained, or otherwise prevented from catching their flight HAS NOT been a terrorist. So they also have a 0% specificity in identifying terrorists.
Third, a good screening test should have a reasonable cost. For example, if every primary care doctor ordered a complete body MRI on every single one of their patients, it would bankrupt the health care system overnight (if Obamacare hadn't done that already). It would also take time and resources away from more efficient screening tests, as well as make MRIs less available for those who really need them). The annual cost of TSA screening is more than $200 million and they have yet to apprehend even one terrorist. Meanwhile, cargo, arguably a far more dangerous threat than passengers, is going completely unchecked. TSA = not cost effective.
Next, a good screening test has to be safe. Checking some one's cholesterol levels - safe. Doing open heart surgery on every person over 50 to feel their coronary arteries to see if they're clogged - not safe. I'm choosing not to discuss the safety of the high-tech strip search scans because I am personally not too worried about the small amounts of radiation used. But we must also consider emotional and physical discomfort when determining the safety of a test. More and more Americans are expressing their discomfort in knowing that in order to get on a plane, they may need to let a complete stranger see images of them naked, or have someone touch their private parts. Since TSA has failed to provide any demonstrable benefits of their security procedures, I believe their safety risks outweigh their benefits.
Lastly, a good screening test needs to be widely available. That's one thing TSA's got. Thanks to our over-zealous Department of Homeland Security, nudie and touchy-feely screening are more widely available than most of us want.
When I can't talk a patient out of getting a whole-body MRI, I at least warn them that they are much more likely to find benign lesions than anything life threatening. Unfortunately, you can't always tell if a lesion is benign from an MRI, and many people may need to go through painful, dangerous surgeries and procedures just to find out that what the radiologist saw was nothing they needed to worry about. So, what does TSA find in their extensive screening processes? Juliet Lupidos of Slate notes, "From Nov. 8 to Nov. 14... agents found six 'artfully concealed prohibited items' and 11 firearms at checkpoints, and they arrested six passengers after investigations of suspicious behavior or fraudulent travel documents. (Those figures are close to the weekly average.) It's not clear, however, whether any of these incidents represent attempted acts of terrorism or whether they were honest accidents. (Whoops, forgot I had that meat cleaver on me! Or, I had no idea flares weren't allowed!)" In other words, TSA is mostly finding a bunch of "benign lesions".
Some would argue that TSA really acts as a deterrent for terrorists, and that you can't really know how many terrorist attacks have been prevented by TSA looking down our underwear. But that reminds me of a Sesame Street skit I saw when I was 5, where Ernie was holding a banana in his ear, and Bert asked him, "Why is that banana in your ear?" "I use this banana to keep the alligators away." "Ernie, there are no alligators on Sesame Street." "Right! It's doing a good job, isn't it Bert!" Statistically, TSA can claim the same efficacy at stopping terrorists as holding a banana in your ear, and I'd rather hold a banana in my ear than have someone, who hasn't even bought me dinner, touch my penis.
Now back to the MRI scan. I order MRIs almost every day, but only for people who really need them, and only of the body part I'm interested in. I determine who and what to MRI by listening to patients' personal histories, family histories, and current complaints. I then di a careful, focused physical examination. From this information I cab determine who is most likely to benefit from an MRI. I guess you could call me a "medical profiler". It's too bad Homeland Security can't take a lesson from the medical field.