This Title All WIREs
How to cite this WIREs title:
WIREs Forensic Sci

Presumptive drug testing—The importance of considering prior probabilities

Full article on Wiley Online Library:   HTML PDF

Can't access this content? Tell your librarian.

Abstract Presumptive drug testing is commonly used in both the clinical and forensic fields to allow rapid identification of the presence and/or usage of drugs. Because the tests generally have a high sensitivity and specificity (often >90%), then a positive test result may be taken to mean there is a high probability that a targeted drug is present. This assumption is, however, incorrect. This paper demonstrates how, in order to assess the positive predictive value (PPV) of a test, it is necessary to take into account, along with the sensitivity and specificity of the test, the prevalence of the drug in the population being investigated. We demonstrate how an alternative, Bayesian approach to assessing the posterior probability of a drug being present mimics the conventional calculation of PPVs but, because a Bayesian approach requires case‐specific prior probabilities, the posterior probabilities are more meaningful than PPV in any one specific case. The effectiveness of presumptive test results in cases such as drink‐spiking, drug‐driving, testing of drugs during seizures and the confirmation of initial presumptive test results is explored. In order to exploit the potential of presumptive drug testing, it is important that the prevalence of the targeted drugs in relevant populations is understood but, more importantly, it is important to consider using a Bayesian approach in order to tailor results to the specific individual or drug batch being tested. This article is categorized under: Toxicology > Analytical Toxicology Jurisprudence and Regulatory Oversight > Expert Evidence and Narrative
Tree diagram of the false discovery rate of a cocaine screening test using fingermarks. This example assumes that 10,000 people living in Europe are randomly screened using the cocaine screening test. The prevalence of cocaine abuse in drivers is 0.42%, the specificity of the test is 97.5% and the sensitivity is 98.7%. Out of the 10,000 people screened, 41 + 249 = 290 positive tests. 249 are false positives so the false discovery rate is 14%
[ Normal View | Magnified View ]
Tree diagram for (a) the color test (sensitivity 0.68 and specificity 0.75) and (b) the electrochemical test (sensitivity 0.93 and 0.86) for cocaine. The example assumes that 10,000 tests were carried out with 80% of the samples truly being cocaine. Cost of screening test £1, Cost of confirmatory test £100. Each positive test is confirmed. (a) Total number of positive tests = 500 + 5,440 = 5,940; Cost of Screening tests = 5,940 * £100 = £594,000. (b) Total number of positive tests = 280 + 7,440 = 7,720; Cost of Screening tests = 7,720 * £100 = £772,000
[ Normal View | Magnified View ]
Graphical representation of the changes in the positive predictive value (PPV; the proportion of individuals with positive test results who are correctly identified as having consumed cocaine) and the negative predictive value (NPV; the proportion of individuals with positive test results who are incorrectly identified as having consumed cocaine). The sensitivity of the test is set at 0.987, the specificity as 0.975
[ Normal View | Magnified View ]

Browse by Topic

Jurisprudence and Regulatory Oversight > Expert Evidence and Narrative
Toxicology > Analytical Toxicology

Access to this WIREs title is by subscription only.

Recommend to Your
Librarian Now!

The latest WIREs articles in your inbox

Sign Up for Article Alerts