5 ms·

Interesting idea (congrats on launching) and interesting discussion here. Reading this thread has popped up some questions in my mind and I'm curious to hear t

by nlh 7y ago

Interesting idea (congrats on launching) and interesting discussion here. Reading this thread has popped up some questions in my mind and I'm curious to hear thoughts: Why do we (aka society aka the police) test for specific substance quantities at all? Really, I think, what we care about is whether someone is able to competently perform the task of driving at a given moment. If you are unable to perform that task, then you should be removed from the road (and, possibly, punished). Frankly, I don't really care whether you're drunk, high, tired, distracted from texting or even if you're just a plain 'ol bad driver. The net result is the same - you're an increased risk to others on (and off) the road and your risk should be mitigated. This gets to the root of a number of related issues - the fact that different bodies react to different levels of chemicals in different ways, etc. etc. If you're able to function at a given moment, fine by me. If your driving is compromised for whatever reason (as demonstrated by a field sobriety test), then off the road you go. Thoughts / corrections / slap-downs?

sergiomattei 7y ago
You can test and arrest for objective things, not subjective ones. "Who can or can't drive" is a subjective criteria. Objective criteria is someone under the influence of x or y substance.
- stevenhuang 7y ago
  What's subjective about a field driving test? The reason is simply that it's impractical to do so. It's easier to run a quick BAC test, look at a number, and call it in than supervise a road test. So you can imagine there being false positives where the person fails the test but would otherwise drive safely. Don't forget the true intent of these tests are to determine impaired driving. The criteria of "being under the influence of something" has historically been a useful analogue for that, but the question is if we can do better somehow by testing for impaired driving directly and not 2nd order metrics.
SannTek 7y ago
Excellent point. The best solution to impaired driving would be a 100% accurate impairment test. The problem comes with designing and implementing such a test. Let's start with what they have today, the standard field sobriety test. It is a pretty good test of impairment (with accuracy range between %75 to 90% depending on the study). However, there are several problems with the SFST. First of all, it requires police officers to be well trained and exercise their judgement roadside. This introduces the possibility for bias and the costs of training can sometimes be prohibitive. Secondly, the SFST cannot be administered in all conditions. A snowy day is a really bad time to be impaired, but also really difficult conditions for running a SFST. The alternative is some kind of device that could detect the same (or different) markers of impairment as the SFST. The problem is making a device that is as accurate as the SFST, is portable, and easy to use is very difficult. If you look at attempts to date (like the Ocular DAX), these device tend to be really unwieldy and not even as accurate as the SFST. It seems like a technology that needs a little more time (maybe a decade?). Overall though, definitely an avenue to explore.

by nlh 7y ago

sergiomattei 7y ago

You can test and arrest for objective things, not subjective ones. "Who can or can't drive" is a subjective criteria. Objective criteria is someone under the influence of x or y substance.

stevenhuang 7y ago
What's subjective about a field driving test? The reason is simply that it's impractical to do so. It's easier to run a quick BAC test, look at a number, and call it in than supervise a road test. So you can imagine there being false positives where the person fails the test but would otherwise drive safely. Don't forget the true intent of these tests are to determine impaired driving. The criteria of "being under the influence of something" has historically been a useful analogue for that, but the question is if we can do better somehow by testing for impaired driving directly and not 2nd order metrics.

SannTek 7y ago

Excellent point. The best solution to impaired driving would be a 100% accurate impairment test. The problem comes with designing and implementing such a test. Let's start with what they have today, the standard field sobriety test. It is a pretty good test of impairment (with accuracy range between %75 to 90% depending on the study). However, there are several problems with the SFST. First of all, it requires police officers to be well trained and exercise their judgement roadside. This introduces the possibility for bias and the costs of training can sometimes be prohibitive. Secondly, the SFST cannot be administered in all conditions. A snowy day is a really bad time to be impaired, but also really difficult conditions for running a SFST. The alternative is some kind of device that could detect the same (or different) markers of impairment as the SFST. The problem is making a device that is as accurate as the SFST, is portable, and easy to use is very difficult. If you look at attempts to date (like the Ocular DAX), these device tend to be really unwieldy and not even as accurate as the SFST. It seems like a technology that needs a little more time (maybe a decade?). Overall though, definitely an avenue to explore.