Send email for updates
Is treatment the best way to overcome dependent substance use, or can we suppress it more cost-effectively by credibly threatening sanctions? Influential researchers have argued that some US programmes (evaluations of which are collected together in an Effectiveness Bank bulletin) show many dependent individuals stop using substances if non-use is enforced through intensive monitoring and swift, certain, but not necessarily severe sanctions. Rather than mandating treatment, these programmes directly mandate abstinence on penalty of sanctions like a day in jail or temporary interruption of professional practice. Among the researchers were top White House drugs advisers. Their perspective found its way in to US anti-drug policy and is now gaining ground in Britain.
‘Test and sanction’ schemes can be thought of as lying at the ‘heavy’ and more coercive end of contingency management programmes, which punish people or deprive them of rewards when they use substances in ways we don’t want them to, and reward them when they behave as we want. Commonly rewards like shopping vouchers or a more relaxed treatment regimen are offered if the patient avoids use of the targeted substance(s) or complies with therapy, and withheld if they do not – not so drastic as a spell in jail or threatening your livelihood, and usually intended to reinforce rather than replace conventional therapy.
As in contingency management schemes, to maximally affect behaviour it is thought test-and-sanction schemes should impose consequences swiftly and inevitably on the infringement, strongly linking the two in the offender’s mind. In turn these advantages depend on the intensity of testing and the relatively minor nature of the sanctions. As a report on a scheme in Hawaii put it, “Severity is the enemy of swiftness and certainty, because a severe penalty will be more fiercely resisted and requires more due process to support it.” Importantly too, minor sanctions mean the offender remains in a position to get swiftly back on track without too much damage having been done, rather than being forcibly parked in a dead-end which is either irrevocable (like permanent loss of professional status) or cannot be reversed for months or years (like an extended prison term).
Though this hot topic focuses on evidence, there is also a strong ethical component to the acceptability of test-and-sanction schemes, especially in societies which classify dependent substance users as suffering from a medical or mental disorder. Despite being charged with supporting international, sanctions-based, drug control treaties, the United Nations Office on Drugs and Crime nailed its colours to the mast in the title of its document on this issue: From coercion to cohesion: Treating drug dependence through health care, not punishment. It points out that “The conventions encourage the adoption of a health-oriented approach to both illicit drug use and drug dependence”, and argues that since “Drug dependence is a health disorder ... punishment is not the appropriate response”. Rather than coercion coming first and treatment being reserved for non-responders, the document argues for treatment as the first-line response with coercion reserved for offenders who reject this opportunity.
Running this search shows the UK also has a considerable history of implementing testing-based programmes for offenders, though generally as a way of monitoring progress and as part of a criminal justice element ‘gripping’ offenders while treatment exerts its effects.
The US programmes challenge this subsidiary role, elevating testing and sanctions to the primary role,
Results challenge the view that relapse is an essential feature of substance dependence
and sometimes relegating treatment (if available at all) to those unable to comply without it. Results are said to challenge the view that relapse is an essential feature of substance dependence, and to demonstrate that the key to long-term success lies in sustained changes in the environment in which decisions to use or not use are made. If this rewards substance use, it is likely to continue, but the drinking and drug use of many dependent individuals stops if the environment not only prohibits use in principle, but enforces this through intensive monitoring of substance use and meaningful consequences.
The face validity of these approaches and the persuasiveness of advocates has yet to be matched by rigorous and positive research findings – not unusual in the criminal justice sector, where allocating offenders to different types of punishment at random can fly in the face of ethical practice and attract legal challenge.
One prominent scheme has however been tested in a randomised trial – but one which seems never to have been published in a peer-reviewed journal. Hawaii’s HOPE programme randomly and frequently drug-tests substance-using offenders to promote compliance with a probation requirement not to use illicit drugs. Each violation results in a brief jail stay of typically just a few days; continued violations attract longer sentences. There is no attempt to universally impose treatment; probationers are ordered into treatment only if they continue to test positive for drug use, or if they request it. HOPE’s goals are reductions in drug use, crimes, and imprisonment. Those goals were shown to have been achieved, both in an initial pilot programme among high-risk probationers, and in a trial which randomly allocated 493 probationers to HOPE versus probation-as-usual. They had been selected by their probation officers as being at the highest risk of failing probation, and averaged about 17 prior arrests. In the follow-up year 21% of HOPE probationers were rearrested compared to 47% allocated to usual probation. HOPE’s lead was just as substantial in the tally of drug-free urine tests, days not spent in prison, and in avoiding revocation of the probation sentence.
Though moving in the direction of US examples, the UK still relies largely on conventional sentencing and treatment. British courts do have Drug Abstinence Orders at their disposal, which can require certain drug-related offenders, including those dependent on illegal drugs, to remain abstinent from drugs and to prove it by submitting to testing. In contrast, similar provisions for alcohol are reserved for non-dependent drinkers. For these offenders, the Alcohol Abstinence and Monitoring Requirement allows for electronic monitoring (for example, by ankle bracelets which detect alcohol secreted in the drinker’s sweat) as well as more conventional testing. As for drugs with the Drug Abstinence Order, drinking contravenes the court order and means the offender can be recalled to court to face a possible sanction.
In Britain interest has centred on using SCRAM alcohol-detecting anklets manufactured by Alcohol Monitoring Systems Ltd, which effectively monitor drinking continuously and 24 hours a day illustration. Widely used in the USA, but it seems never rigorously evaluated, the anklet was one of the strategies employed by the US state of South Dakota’s 24/7 Sobriety programme, an exemplar for test-and-sanction strategies. It started as a court order for drink-driving offenders, who were subject to immediate 24-hour imprisonment if found to have drunk alcohol, but was extended to other criminal justice situations to enforce abstinence. At least during the sentence, recidivism reductions and compliance among programme participants were on the face of it impressive, but the programme has never to have been benchmarked against an adequate comparison group.
It seems the closest we have come to an adequately benchmarked evaluation was a US study which recruited drink-drivers sentenced to wear a version of the SCRAM anklet piloted in London below. Explained further in the commentary on the London study, the study identified an individually matched comparison set of convicted drink-drivers. However, the analysis was silent on whether SCRAM wearers overall were reconvicted less often. Instead it focused on the relative conviction reduction among a subgroup of repeat offenders who wore the anklet for at least three months. Fewer than half as many as in the comparison group were reconvicted, but the analysis no longer retained the reassurance of a matched sample, because no equivalent subgroup could be identified among comparison offenders. The study as a whole was vulnerable to differences between offenders (or circumstances) for whom courts ordered the anklet and those for whom they did not, differences which may have contributed to the results, regardless of the anklet.
Inspired by US examples, the UK coalition government in power until 2015 committed to funding a trial of alcohol-detecting bracelets for serious drink-related offenders. That mantle was later taken up by the Major of London, whose office in 2014 announced a pilot during which anklets would be fitted to offenders in four London boroughs to enforce an Alcohol Abstinence and Monitoring Requirement.
Offenders were eligible for the anklet if they had committed a drink-related offence but were not dependent on alcohol. Over on average 75 days wearing the anklet, 92% complied with their court order and did not drink. Whether there were longer term effects is unknown. Evidence of a bounce-back to offending after anklet removal among some US drink-driving offenders makes this a concern.
Publication in 2016 of the pilot’s results led to a decision to make the anklet-based sentence available across London, and formed part of the evidence base for the UK government’s plan to introduce sobriety as a court-imposed community order to reduce alcohol-related reoffending.
The Mayoral pilot in London was not the only time the SCRAM anklet had been trialled in the UK, but the other study was very different, involving male student drinkers. They volunteered for a study which for a fortnight randomly allocated them to be instructed not to drink either wearing or not wearing the anklet. If by the end of the first week the anklet revealed drinking, the student concerned was phoned and reminded of the instruction not to drink. Another set of students wore the anklet, but were told to drink as normal during the two weeks.
When drop-outs were assumed to have drunk alcohol, there were no statistically significant differences in drinking between students told not to drink, regardless of whether they had been allocated to the anklet. This result reflected grater drop-out among anklet-allocated students, sometimes because the device made them feel “uncomfortable, anxious, or dehumanized”. Among remaining students, far fewer drank if they were wearing the anklet, but this result cannot be relied on as an indicator of the overall impact across all students.
In addition to convicted offenders, testing has also been used in the UK to control offenders who wish to avoid prosecution and instead agree to be cautioned. After piloting, the new ‘Conditional Caution with Sobriety Conditions’ became nationally available to courts as one way to deal with low-level alcohol-related offending. In lieu of prosecution, the offender agrees not to reoffend and to completely abstain from alcohol for a specified period on days when they are likely to offend as a result of drinking, enforced by regularly breath-tests. Failure to comply could result in prosecution for the original offence.
Roll-out was approved despite the unconvincing results of the pilot programmes. Run from May 2012 in five areas of England, the pilots “highlighted a general lack of understanding of the process”. What made the Home Office admit this was the fact that just 10 out of 92 offenders who qualified for the new cautions were given them, largely because (despite the intention being to make them acceptable) 68 refused the option. Of the 10 who did start the new court orders, just six completed them.
A distinctive feature of the US programmes is the strong leverage available to sanction substance use and reward abstinence: in physician health programmes, removal from practice and ultimately the loss of one’s license to practice medicine, versus continuing to practice in a prestigious and well paid profession; in programmes for offenders, immediate brief imprisonment versus freedom.
Wider application of such programmes hinges on finding or engineering this leverage, and on having both the legal authority and the resources to swiftly and certainly sanction transgressors. Without leverage, programmes risk simply siphoning non-compliant offenders into conventional penal sanctions; without sure sanctions, the programme exists only on paper and can safely be ignored by offenders.
Test-and-sanction programmes also require considerable administrative reorganisation and commitment. In 2014 the lead researcher of the HOPE evaluation knew of “at least 40 replications of HOPE-style models on the [US] mainland”. Whether they were producing the results seen in the original was unclear, and dependent she thought on adherence to the strategy’s core principles: “swift, certain and proportionate sanctions”. From her account, it seems that generating and maintaining this approach is not straightforward and cannot be expected to work everywhere, requiring a committed judge, the cooperation of probation and other partners, effective, engaging leadership, increased efforts to detect probation violations, and substantial reorganisation of procedures. The strategy is simple – detect and punish – but putting it into effect is not.
Thanks to David McDonald of Social Research & Evaluation Pty Ltd in Australia for suggesting this hot topic.
Last revised 30 October 2016. First uploaded 29 June 2012
Comment/query to editor
Give us your feedback on the site (one-minute survey)
Open Effectiveness Bank home page
Add your name to the mailing list to be alerted to new studies and other site updates