- Joined
- Nov 27, 2002
- Messages
- 7,890
- Reaction score
- 756
NHTSA (National Highway Traffic Safety Administration) is doing studies as described in this report on CNN.
In short sheriff's deputies pull drivers over and then ask them to voluntarily submit to anonymous saliva and/or blood tests to see how many people are driving while impaired.
This data will be used to help NHTSA set goals and policies and for anyone who didn't know NHTSA is very influential in the world of EM in that they set policies for EMS and trauma systems.
Aside from the civil liberty and privacy concerns discussed in the article I have two big problems with this research:
First any study like this will be devastated by selection bias. Imagine if you had one drink or took a Benadryl that day and got "selected." Would you believe the people saying it's anonymous or would you remember an important appointment you just had to get to? Clearly the number of impaired people will be underestimated in this study. Presumably the statistical techniques and analysis they mention will amount to trying to compensate by inflating the positive number but then that's really just a guess. Consequently any result will be highly dubious.
Second this study is unnecessary and wasteful. What difference will it make if the number of impaired drivers is 5% or 10% or 30%? If it's 5% will NHTSA declare they are ok with that and nothing else needs to be done? That's highly unlikely. Since NHTSA will always work to reduce the number of impaired drivers the study is unnecessary. I think it's an example of a bureaucracy doing things because they always have and needing to justify their budget.
In short sheriff's deputies pull drivers over and then ask them to voluntarily submit to anonymous saliva and/or blood tests to see how many people are driving while impaired.
This data will be used to help NHTSA set goals and policies and for anyone who didn't know NHTSA is very influential in the world of EM in that they set policies for EMS and trauma systems.
Aside from the civil liberty and privacy concerns discussed in the article I have two big problems with this research:
First any study like this will be devastated by selection bias. Imagine if you had one drink or took a Benadryl that day and got "selected." Would you believe the people saying it's anonymous or would you remember an important appointment you just had to get to? Clearly the number of impaired people will be underestimated in this study. Presumably the statistical techniques and analysis they mention will amount to trying to compensate by inflating the positive number but then that's really just a guess. Consequently any result will be highly dubious.
Second this study is unnecessary and wasteful. What difference will it make if the number of impaired drivers is 5% or 10% or 30%? If it's 5% will NHTSA declare they are ok with that and nothing else needs to be done? That's highly unlikely. Since NHTSA will always work to reduce the number of impaired drivers the study is unnecessary. I think it's an example of a bureaucracy doing things because they always have and needing to justify their budget.