International approaches to drug law reform: research

5. Summary of key themes

5.1 Overview of criminalisation and harm minimisation

Research on drug criminalisation generally indicates that criminal sanctions for drug use or possession tend to exacerbate harm or undermine efforts at harm minimisation. For example:

• Injection safety:
  Studies looking at barriers to using a new needle for every injection in large, diverse urban settings have consistently found that fear of law enforcement encounters is associated with people:
  • Sharing syringe and injection equipment (Wagner, 2010) (Wagner, Simon-Freeman, & Bluthenthal, 2013) (Bluthenthal, 1999) (Wood, 2017) (Pollini, 2008) (Flath, 2017)
  • not accessing new needles, from reduced exchange program participation (Wood, 2017) (Beletsky, 2014) (Bluthenthal, 1997) (Davis, 2005)
  • feeling reluctant to carry new needles (Flath, 2017) (Sarang, Rhodes, & Platt, 2008) (Grund, 1995)
• Treatment access:
  When evidence-based and delivered well, treatment is known to be a cost effective means of reducing drug harms.
  • A 2015 study in England used estimates of opioid use in the general population, the extent of treatment provision, and the number of deaths related to opioid use, to develop a counterfactual model which estimated that treatment prevented 880 excess opioid-related deaths each year between 2008 and 2011 (White, et al., 2015).
  • Evidence from the US looking at the cost-benefit ratios of treatment programmes consistently show that their benefits reflect a large return on investment. A study of California treatment outcomes found a benefit to cost ratio of 7 to 1, largely attributed to reductions in subsequent criminal activity (Ettner, et al., 2005). This is consistent with an earlier literature review of 18 benefit-cost studies, which found that benefits exceeded costs, with benefit–cost ratios of 1.6 to 26 (Cartwright, 2000).

Considering this, it is important to ask what impact prohibition has on treatment engagement. An international survey of drug users found that those from countries with a strong prohibition-based drug policy reported a far greater propensity to seek help following the introduction of more permissive policies (Benfer, et al., 2018). The main reason for the change in help-seeking behaviour cited was the reduced fear of criminal sanctions.

• Incarceration:
  A meta-analysis of studies from six countries found that after imprisonment, drug users were at three to eight times more risk of drug-related death in the first two weeks after release, and remained elevated for a further two weeks (Merrall, et al., 2010).
• Justice system strain:
  Decriminalisation reduces the demands on the criminal justice system, through less demand on police, courts and prisons. For example, total law enforcement costs in California dropped from $17 million in the first half of 1975 (before decriminalisation) to $4.4 million in the first half of 1976 (after decriminalisation). Such savings can contribute towards replacing criminal intervention with well-resourced social and treatment intervention instead.

Additionally, research from New Zealand estimated that more than half of the revenue from all drug trafficking is reinvested in criminal activity, and that around 20-28% of this funds other organised crime activities such as extortion, fraud, pornography, illegal poaching and weapons trafficking (McFadden Consultancy, 2016).

• Social integration:
  Decriminalisation improves employment prospects and relationships with significant others for those detected with drugs, and evidence from a number of countries shows that decriminalisation can lead to improved social outcomes. For example, individuals who avoid a criminal record are less likely to drop out of school early, be sacked or to be denied a job. They are also less likely to have fights with their partners, family or friends or to be evicted from their accommodation as a result of their police encounter (Drug Policy Modelling Programme, 2016).

Moreover, research indicates that decriminalisation does not lead to significant increases in drug use or other crimes (Babor, et al., 2010).

• Decriminalisation has no or very small effects on rates of drug use. Research from across the globe has consistently found that decriminalisation is not associated with significant increases in drug use.
• In instances where just cannabis has been decriminalised it has not led to increases in use of other drugs such as ecstasy or heroin.
• Research has shown that decriminalisation does not lead to increases in crime through perceptions of weaker laws. People who do not receive a criminal record are much less likely to engage in future crime or have subsequent contact with the criminal justice system, even when previous offending history is taken into account. There is also no evidence that decriminalisation will lead to other types of crime, such as supply or drug-related crime.

Within a legalised system, there are different purchaser/end-user regulatory options including, for example, age and place of use restrictions like those that most jurisdictions currently use for alcohol. From least to most strict, these options include:

Unlicensed sales model	Regulation focuses on standard product descriptions and labelling. Where appropriate, food and beverage legislation (dealing with packaging, sell by dates, ingredients etc.) may be adopted.
Licensed premises model	Licensee is responsible for restricting sales on the basis of age, intoxication and hours of opening. Licence infringements may be sanctioned by a sliding scale of fines, loss of licence, or even criminal penalties. Licensees can be held partially or wholly liable for how their customers behave, similar to the licensed premises model applying to the on-premises sale of alcohol. Various controls exist over the venue and (in particular) the licensee.
Licensed sales model	May put various combinations of regulatory controls in place to manage the vendor, the supply outlet, the product and the purchaser. These controls may be supported by changes to police, customs, trading standards, and health and safety policies and practices, similar to the off-premises sale of alcohol.
Pharmacy sales model	Pharmacists are trained and licensed to dispense prescriptions, although they cannot write them. They can also sell certain generally lower risk medical drugs from behind the counter of licensed pharmacy venues, usually with conditions such as buyer age, level of intoxication, quantity requested, or case-specific concerns relating to potential misuse. Pharmacists are overseen by regulatory legislation, managed by various agencies and are subject to a clearly defined enforcement infrastructure.
Prescription model	The most tightly controlled and enforced drug supply model - drugs are prescribed to a user by a qualified and licensed medical practitioner, and dispensed by either a licensed practitioner or pharmacist from a licensed pharmacy or other designated outlet. The process is controlled by a range of legislation, regulatory structures and enforcement bodies, which guide, oversee and police the prescribing doctors and dispensing pharmacists. They also help determine which drugs are available, in what form, where, and under what criteria.

summarised from Rolles, 2010

One negative consequence of decriminalisation that has been reported in research from Australia is 'net widening'. Net widening occurs when more people are sanctioned after a reform than before, due to the greater ease with which police can process minor drug offences. The extent of this depends on the specific choice of policy design and how the reform is implemented (for example, whether the consequences for non-compliance are more severe than the original offence; the extent of police discretion) (Bryan, Del Bono, & Pudney, 2011).

5.2 Drug consumption rooms

To reduce harm and prevent overdoses, many countries including Denmark, Switzerland, Germany, Spain, Norway, Australia and Canada have established drug consumption rooms or safe injecting facilities over the last 30 years . Drug consumption rooms are typically professionally supervised healthcare facilities where drug users can use drugs in safer and more hygienic conditions. As of 2018 there were over 90 DCRs in over 60 cities across 10 different countries, with several more jurisdictions working towards them (EMCDAA, 2018).

A growing body of scientific evidence shows that consumption rooms have an impact on both improving health and reducing death by overdose among clients who use these facilities (Bravo, De la Fuente, Brugal, Barrio, & Domingo-Salvany, 2009)(DeBeck, et al., 2011) (Kinnard, Howe, Kerr, & Marshall, 2011). For people who use drugs, unsafe drug intake often involves unhygienic or incorrect injecting, which cause both injury and infection. A systematic review found that supervised consumption at a facility reduces the risk of fatal overdose and disease transmission (Pardo, Caulkins, & Kilmer, 2018).

Several qualitative studies highlight the benefits of DCRs, noting that they can address various contextual risks associated with public injecting by:

• enabling safer injection practices (Kerr, Small, Moore, & Wood, 2007),
• providing refuge from street-based crime (Fairburn, Small, Shannon, Wood, & Kerr, 2008),
• mediating and facilitating access to healthcare and social resources (McNeil & Small, 2014) and
• delivering education regarding safer injection practices which is more likely to be accepted among clients than it would be from other sources (Fast, 2008).

However, as noted above regarding criminalisation, the success of these services relies on people's willingness to engage with them. Fear of apprehension by the law can be a significant deterrent to accessing services, so to be effective, DCRs should be accompanied by support from policing to ensure service users are not criminalised, and that police actively refer people who use drugs to the service.

Evidence on consumption rooms' impact at a population level is sparse, but some evidence from ecological studies suggests that, where coverage is adequate, drug consumption rooms may contribute to reducing drug-related deaths at city level (Poschadel, Hoger, Schnitzler, & Schreckenberger, 2003) (Marshall, Milloy, Wood, Montaner, & Kerr, 2011). A study in Sydney showed that there were fewer emergency service call-outs related to overdoses at the times the safe injecting site was open (Salmon, Van Beek, Kaldor, & Maher, 2010).

5.3 Cannabis law reform

For decades, research on the impact of cannabis decriminalization has shown that, in a variety of jurisdictions in Australia, Europe, Canada and the United States, decriminalization does not cause a significant increase in consumer demand or ease of access.

• Data from four Canadian cities showed that under prohibition, cannabis users generally, even in times of easy access, moderate their cannabis use such that it does not interfere with their lives or lead to adverse health consequences. These patterns appear to continue under decriminalization (Duff, et al., 2012).
• The literature on depenalisation of cannabis possession in various states of the United States, the Netherlands, Portugal and Australian states finds that reducing penalties has either no or only small effects on prevalence of use (Home Office, 2014) (Hughes, 2016) (MacCoun, 2010).
• Bryan et al (2013) found there was no evidence that decriminalisation of limited cultivation in South Australia and Alaska substantially changed consumption.
• At the same time, decriminalization decreases related social problems (such as criminal records and their impact for people) as well as enforcement and judicial costs (see, for example: (Single, 1989) (Lenton, et al., 2000) (Room, 2008).

In relation to young people specifically, a recent review of 38 countries shows no significant increase in cannabis use amongst adolescents living in liberalised states (Stevens, 2019). Consistent with the results of previous researchers, the most comprehensive empirical study to date found:

• there was no evidence that the legalization of medical marijuana encourages marijuana use among youth.
• Moreover, the estimates showed that marijuana use among youth may actually decline after legalization for recreational purposes (Anderson, Hansen, Rees, & Sabia, 2019).
• Some of the evidence suggests legalisation may reduce underage consumption in some instances, due to fewer illegal suppliers.

There is also some evidence that liberalising cannabis laws causes some people to substitute away from higher harm drugs such as opioids or alcohol. In US states that have legalised medicinal cannabis, there is early evidence of some people substituting from opioids to cannabis, and lower attendant health burdens:

• death certificate analysis showed that states with medical cannabis laws had lower opioid analgesic overdose mortality rates compared with states without such laws (Bachhuber, Saloner, & Cunningham, 2014).
• This persisted when excluding intentional overdose deaths (ie, suicide), suggesting that medical cannabis laws are associated with lower opioid analgesic overdose mortality among individuals using opioid analgesics for medical indications.
• Similarly, the association between medical cannabis laws and lower opioid analgesic overdose mortality rates persisted when including all deaths related to heroin, even if no opioid analgesic was present, indicating that lower rates of opioid analgesic overdose mortality were not offset by higher rates of heroin overdose mortality (Bachhuber, Saloner, & Cunningham, 2014).
• A study of substance use trends amongst US college students in the ten years ending 2018 found that recreational cannabis legalisation was linked to decreased binge drinking among students age 21 and older. Recreational marijuana legalisation was also associated with increased sedative misuse among minors (Alley, Kerr, & Bae, 2020).
• Dragone et al. (2019) estimated that cannabis legalisation increased cannabis consumption by about 2.5 percentage points, a decrease in alcohol consumption of 2 points and a decrease in other drug consumption of about 0.5 points. This change in consumption was associated with a change in criminal behaviour, with reductions in rape of between 15 and 30 per cent and theft of between 10 and 20 per cent.
• In their review of the evidence Anderson and Rees (2014) find that "studies based on clearly defined natural experiments generally support the hypothesis that marijuana and alcohol are substitutes." Increasing the drinking age seems to result in more marijuana consumption, for instance, and pot smoking drops off sharply at age 21, "suggesting that young adults treat alcohol and marijuana as substitutes." This supports other findings that legalizing marijuana for medical use has been associated with a drop in beer sales and a decrease in heavy drinking.
• Earlier research found that enacting medical marijuana laws is associated with a 13 percent drop in traffic fatalities. They posit that this effect is likely due to a combination of the fact that marijuana impairs driving ability much less dramatically than alcohol does, and the fact that alcohol is more likely to be consumed outside the home, resulting in more driving under its influence (Anderson, Hansen, & Rees, 2013).

There is some evidence that prohibition of cannabis has led to the advent of synthetic cannabinoids, which have been formulated to mimic the effects of natural cannabis but have much greater harms. The incentive for this innovation arises because of their benefits in a prohibited market place: "They are easy to purchase, relatively inexpensive, produce a more potent high and don't emit the typical marijuana scent. And, they are much harder to detect in the urine or blood than marijuana. Legalisation is likely to shift users from synthetic cannabis back to natural cannabis. Regulation would also reduce contamination of natural cannabis" (White C. , 2018).

• In 2013 alone, 150 new cannabinoids were identified and these new synthetic drugs have been linked to poisonings, hospitalisations and deaths. Serious illnesses due to cannabis are exceedingly rare, while those due to synthetic cannabinoid use are becoming more common, and clusters of synthetic cannabinoid overdose are associated with the newest drugs (Bannister et al. 2015).
• A recent study of Australian drug harms (Bonomo, et al., 2019) ranked the harm to users from synthetic cannabis as more than twice that from natural cannabis. An American study estimated synthetic cannabinoids were 30 times more likely to harm the user (White, 2017).

There is less in the published literature on implementation issues, but it is worth noting that several health policy organizations in Canada and abroad have warned against the perils that industry lobbying may present for protecting the public's health in cannabis law reform. One proposal for mitigating this issue is for a legal market to be operated by the state:

• "A not-for-profit cannabis authority would maintain a singular mission of protecting the public's health. It would serve this mission by supplying safe product to serve only the existing demand for cannabis. There would be no intent, or provision of incentive or encouragement of any kind to increase use among cannabis users or to induce nonusers to start using cannabis. Given the primacy of a public health mission, a long-term objective might actually be to reduce use by individuals and prevalence in the general population. The Authority would provide and promote evidence-based information on the low-risk use of cannabis, and actually encourage and support people who wanted to stop or reduce their cannabis use" (De Vallaer, 2017).

5.4 Reform for other drugs

As experience of significant law reform regarding harder drugs than cannabis is relatively recent and not yet widespread, there is less research evidence available to draw from. However, the overall trend of the literature on drug law enforcement consistently indicates that increases in punitive or prohibitionist approaches are generally associated with increased harms and, conversely, that many harms can be ameliorated when the law, and/or its enforcement, is relaxed.

• Studies using data from Florida, New York and Portugal consistently show that escalations in drug enforcement are accompanied by property and violent crime rates increasing, relative to what they would have been (Benson, 2009).
• Effectively removing mandatory minimum sentences for people convicted of a range of felony drug charges, and increasing eligibility for diversion to treatment led to a 35 percent rise in the rate of diversion of eligible defendants to treatment in New York. Although the use of diversion varied significantly among the city's five boroughs, it was associated with reduced recidivism rates, and cut racial disparities in half (Parsons, et al., 2015).