Evaluation of different roadside drug tests. ROSITA. Deliverable D4

This part of Rosita consisted of the evaluation of the on-site devices for urine, oral fluid and/or sweat in 8 countries.
The study was performed on 2968 subjects, 92% of them male.
The need is so big that in some countries police officers will settle for imperfect devices, although we strongly advise against using any of the present oral fluid devices for benzodiazepines or cannabis detection.
Police did not have major objections to collecting specimens. There was a majority of countries that favoured oral fluid as a matrix.
On site testing gives police confidence, saves time, and money. In general, the use of roadside tests offered the following advantages in the enforcement of drug-driving laws, both in countries with an impairment-type law and in countries with per se laws: It gave confidence to the police officer. Without an on-site tool to confirm his impression, a police officer will be more reluctant to press charges. After some time, thanks to the immediate feedback, he rapidly increases his skill at detecting drugged drivers; The publicity that accompanied the use of roadside tests (e.g. in Finland) was considered (by the police officers) to have a preventive effect, because people thought they could be controlled everywhere (while in fact the number of actual tests used was limited).
Most of the urine devices worked well and generally served as good predictors of blood concentrations. Sampling urine was no problem if appropriate facilities were present. For the different drugs (amphetamines, benzodiazepines, cannabinoids, cocaine and opiates), several on-site devices met our analytical criteria for the reliability of analytical results.
Oral fluid and sweat are promising specimens and in some cases are better than urine but more research and development will be needed. Sampling of oral fluid and/or sweat are much better accepted by the drivers and the police officers. For some drugs, with reference methods, there was a better agreement between oral fluid and blood than between urine and blood. The oral fluids devices that were tested were not satisfactory for use at the roadside either in terms of ease of use, duration, sample volume needed, sensitivity and reliability (accuracy of 50-81 % for the different drugs in comparison to blood). For sweat, only one device is available and relatively good results were seen for some drugs, but more studies are needed to determine if external contamination and the later appearance of drugs in sweat are an issue.
The technology is changing rapidly and more accurate, more sensitive, easier to use devices are expected in the near future. Many development efforts are under way, and new devices and improved versions of the devices that we tested here, are expected soon. Further studies will be needed to evaluate them. (Extract of the publication)