The Canadian Mothers Against Drunk Driving organization has promoted two initiatives to the Canadian government, one of which we support, and one of which we do not. The following letter was sent to MADD, explaining our position:
I understand you are promoting availability of oral fluid testing for use at roadside, which I commend. We are doing the same in Colorado, where most of my effort is concentrated. We also hope to use these devices at local station houses. We are limiting our devices for consideration to those provided by Affiniton, Alere, and Draeger, since they have fared best in DRUID and other studies. We see no need to repeat the studies that have been already done. I would welcome the opportunity to learn more about how you plan to evaluate and implement such devices in Canada.
Secondly, I understand you advocate passage of per se drug levels for the most common illicit drugs, similar to alcohol’s .08 gm/dl level. I advise against that effort for the following reasons and suggest an alternative for you to consider:
- First, one must recognize that alcohol’s per se limit was not set by scientific consensus, but rather by political decisions, based upon community values and worldwide scientific input. That explains why, even though all countries had the same scientific data available, alcohol per se levels vary from .02 to .10 gm/dl equivalent in various countries around the world.
- There is no level of an intoxicant above which everyone is impaired, and below which, no one is impaired. That’s not because we need more study, but because of natural human variability. This is true for alcohol, marijuana, methamphetamine, opiates, and any other drug you can think of. In the case of alcohol, most researchers agree that people’s sensitivity to alcohol varies by a factor of approximately ‘2’. That is, one person may be as impaired at .04 gm/dl alcohol as another person is at .08 gm/dl of alcohol. This variability is one of the reasons that per se levels are set by political decisions, not by scientific consensus.
- It also is important to realize that blood is never impaired. Neither is breath. Or oral fluid. Only the brain is impaired. Testing blood, breath or oral fluid is a surrogate to test for what’s in the brain, and because of the time it takes to equilibrate the drugs among the various body tissues, there will always be inconsistencies seen in test results. We know from autopsy studies for example, that someone can have active THC (delta-9-THC) in their brain, even though no THC can be detected in their blood.
- Sensitivity to the impairing effects of drugs varies much more than it does for alcohol. Opiates, for example, have a variability factor of over 100. A person taking opioids for chronic pain or who is on methadone maintenance may be unimpaired at levels that can be lethal to someone not habituated to such drugs. Cocaine has a very high impairment variability as well, whereas marijuana is estimated to have a variability factor of around 10 or more. This makes political decisions to set per se levels for drugs profoundly more difficult than it was for alcohol.
- The UK has recommended per se levels for various drugs and has, in my opinion, done the most comprehensive analysis ever done in order to reach their recommendations. They appear to have based their work on assumptions of impairment levels for non-habituated individuals, which must be done in order to protect the driving public. However, such levels are problematic for drivers who take some of these drugs for legitimate medical reasons, become tolerant to some of the impairing effects of the drugs, yet maintain levels of those drugs well above the proposed per se limits.
- More importantly, the work the UK has done is based upon laboratory assays where impairment and blood (or blood serum) levels are determined simultaneously. That is, a subject’s level of impairment is determined within minutes of taking a biological sample to assay for drug levels. For drugs with a slow metabolic rate, such tests can be meaningful, but for drugs with a rapid metabolic rate like cocaine (half-life of ~40 minutes) or marijuana (half-life of ~1.5 hours) the blood or blood serum level of drugs taken typically hours after a crash will not represent the drug levels at the time of the incident. Marijuana is even worse, since someone may be smoking a joint at the time of the incident and have an extraordinarily high level of active THC in the blood stream. The THC very rapidly equilibrates into the fat stores while it is also metabolizing. These two factors, metabolization and redistribution among body pools, cause the blood THC level to drop 90% within the first hour after smoking ceases. It is not at all uncommon that the level of THC will drop so low before the blood can be collected, that it drops below the laboratory level of quantification, which means it will be reported out as zero – even though the person could have had a very high level of THC at the time of the incident. The UK study suggests a per se level of 5 ng/ml of THC in blood, which may be OK if there were a way to collect the blood immediately, but that can’t happen. I recognize the US with its 4th Amendment restrictions is somewhat unique and therefore may have longer delays to draw blood than other countries, but even without legal impediments, logistics make it impossible for officers to sample blood or other bodily fluids at the moment of a crash, for example.
- Oral fluid drug levels are not the same as blood levels or plasma/serum levels. For example, a 1 ng/ml THC in whole blood may be the equivalent of 2 ng/ml of THC in plasma or serum, and anywhere from 20 to 40 ng/ml of THC in oral fluid. It varies so widely in oral fluids in part because the act of smoking marijuana puts THC in direct contact with oral fluids.
- Roadside testing devices available today are qualitative, not quantitative. That is, they tell the officer that a certain drug was present, but will not give a quantitative level. There are oral fluid sampling devices (and even breath devices) that can collect oral fluid for laboratory testing and eventual quantitative results, but then you need to realize that the results will not be available at the roadside. It may take weeks to get laboratory results back from an oral fluid sampling device.
For the above reasons, attempting to establish per se levels for drugs in blood, plasma, serum or oral fluids is fraught with so many difficulties, that reaching a useful conclusion will take an immense amount of time and effort, and most likely will never be achieved to the satisfaction of politicians who must make the ultimate decision. And during that time, you will not have legislation in place to protect the general public.
Therefore, I suggest you adopt a zero tolerance policy for all illicit drugs in drivers. Note that this is different from a per se impairment level of zero for illicit drugs. It simply says that illicit drugs are by definition illicit, they cause impairment, and therefore society has the right to impose a penalty on any driver found with such drugs in their system. Most countries already have such restrictions in place for airline pilots, commercial drivers and others in safety-sensitive positions.
Furthermore, the oral fluid qualitative testing devices that are currently available, should not be used to prove impairment, but rather should be used to identify the drugs causing the impairment that was observed and documented by the arresting officer. This is an easy step to take and requires far less rigor to justify its implementation.