A field test of substance use screening devices as part of routine drunk-driving spot detection operating procedures in South Africa

This pilot study aimed to test four substance use screening devices developed in Germany under local South African conditions and assess their utility for detecting driving under the influence of drugs (DUID) as part of the standard roadblock operations of local law enforcement agencies. The devices were used to screen a sample of motorists in the Gauteng and Western Cape provinces. The motorists were diverted for screening at roadblocks at the discretion of the law enforcement agencies involved, as per their standard operating procedures. Fieldworkers also administered a questionnaire that described the screening procedure, as well as information about vehicles, demographic information about the motorists and their attitudes to the screening process during testing. Motorists tested positive for breath alcohol in 28% of the 261 cases tested. Oral fluid was screened for drugs as per the standard calibrated cut-offs of all four devices. There were 14 cases where the under-influence drivers tested positive for alcohol and drugs simultaneously, but 14% of the 269 drivers drug-screened tested positive for drugs only. After alcohol, amphetamine, methamphetamine and cocaine were the most common drugs of impairment detected. The results suggest that under normal enforcement procedures only 76% of drivers impaired by alcohol and other drugs would have been detected. In more than 70% of cases the tests were administered within 5 min and this is likely to improve with more regular use. It was clear that the pilot screening process meets global testing standards. Although use of the screening devices alone would not serve as a basis for prosecution and provisions would need to be made for the confirmation of results through laboratory testing, rollout of this screening process would improve operational efficiency in at least two ways. Firstly, the accuracy of the tests will substantially decrease confirmatory test loads. Secondly, laboratory drug testing can be restricted to specific drugs rather than a full panel analysis, which will reduce testing costs significantly.