Random drug testing: Increasing risk of detection
While surveillance and identification of people who use drugs while incarcerated are key objectives of random drug testing (Bruno et al., 2016), preventing or deterring drug use is arguably an even more important and long-term objective of the practice. Deterrence theory is premised on the notion that as perceived risks or expected costs of an action increase, individuals should become less likely to engage in that particular action. At the most fundamental level, the deterrence doctrine is designed to inhibit criminal activity by fear of the certainty, severity, and celerity of punishment (Nagin, 2013), which contemporary scholars largely agree on is a perceptual process (Apel, 2013). Two crucial elements of deterrence are especially important in the context of random drug testing: certainty and perceptions thereof. First, as Beccaria (1764) pointed out, not all three components of the deterrence doctrine are equally effective at preventing crime. Beccaria (1764, 73) writes, crimes “are more effectually prevented by the certainty, than the severity of punishment.” Accordingly, over two centuries later, it has been well-documented through empirical evidence that certainty of punishment is the stronger deterrent than the severity of punishment (Loughran et al., 2016; Nagin, 2013). Second, in the mid-1980s deterrence theorists began to understand deterrence as a perceptual theory, which underscored that the perception of the threat of sanctions was more powerful in explaining behavior than the true certainty or probability of punishment (Geerken & Gove, 1975; Paternoster et al., 1983). In other words, if an individual is weighing the benefits of committing a crime against the potential costs of that act, it is the potential offender’s subjective belief about the certainty of punishment that matters, meaning that simply increasing detection cannot have any effect if potential offenders are not aware of the change.
The goal of deterring substance use and related behaviors in prison through increasing random testing has a strong theoretical grounding. Ideally, increasing the frequency of drug testing would increase incarcerated persons’ perceptions of the certainty of detection (being tested and failing a drug test after use), and thereby increasing expected sanctions and preventing drug use in prisons (Hawken, 2016). While in community correctional settings drug testing with clearly communicated probabilities of detection has been associated with reductions in substance use (Hawken, 2018a; Kilmer & Midgette, 2020; Midgette et al., 2021), evidence of the efficacy of U/A in institutional corrections is sparse.
One of the key objectives of randomized drug testing is to deter incarcerated persons from using drugs (Bruno et al., 2016). Specifically, the theoretical mechanism associated with deterrence is that the certainty, celerity, and severity of the threat of being sanctioned given a positive test should prevent incarcerated persons from using drugs. While we were able to examine the relationship between testing rates and positive drug tests, we were not able to directly test any of the mechanisms that underpin the deterrence doctrine. A number of studies have studied deterrence in the context of random drug tests among probationers, such as evaluations of Hawaii’s Project HOPE (Hawken & Kleiman, 2009) and Delaware’s Decide Your Time (DYT) program (O’Connell et al., 2011). Early evaluations found that the relationship between the certainty, celerity, and fairness of sanctions reduced failed drug tests and arrests (Hawken & Kleiman, 2009). Subsequent evaluations demonstrated reductions in drug use (Hawken, 2018b; Humphreys & Kilmer, 2020) but a mix of null and negative results over effects on traditional criminal justice outcomes, such as recidivism, also exist (Lattimore et al., 2016). We advocate for research to disentangle the mechanisms associated with increasing the effectiveness of random drug testing in prisons.
Prior empirical analyses demonstrate that a certain consequence should be clearly communicated to maximize the deterrence impact, regardless of the severity of the actual consequence (Nagin, 2013; Zimring & Hawkins, 1973). If in fact, as scholars have argued for decades, deterrence is a communication process, for changes in rates of random drug testing to be effective as a deterrent, incarcerated persons must be able to accurately perceive the probability of random drug testing and internalize the changing levels of risk. Though there is evidence that individuals are rational updaters when it comes to synthesizing their own experiences with crime and criminal justice contact (Anwar & Loughran, 2011; Lochner, 2007; Matsueda et al., 2006), it is not clear if incarcerated persons can accurately perceive subtle changes in rates of random drug testing. This is especially suspect over coarse periods of time like month-to-month differences across the entire correctional facility. Therefore, while there is important natural variability in the rates of testing in facilities month to month, it is unclear if incarcerated persons can assess these changes, and therefore be deterred by them.