Skip to main content
SearchLoginLogin or Signup

Diverting people who use drugs from the criminal justice system: A systematic review of police-based diversion measures

Published onApr 28, 2022
Diverting people who use drugs from the criminal justice system: A systematic review of police-based diversion measures


Background. Previous reviews of the effectiveness of measures to divert those who use drugs from the criminal justice system have focused mainly on post-conviction or post-sentence programs and report mixed results. The present systematic review synthesizes evidence on the effectiveness of police-based diversion measures in reducing criminal offenses and other harms related to drug use and then summarizes evidence from qualitative studies to identify facilitators and barriers associated with the implementation of such measures.

Methods. Eight databases were searched to find evaluations of police-based diversion measures for drug-related offenders. Twenty-seven studies were identified. The vote-count method and the Maryland Scientific Method Scale were used to assess the impact of police-based diversion measures. Themes related to barriers or conditions facilitating the implementation of these measures were extracted from qualitative studies.

Results. Evidence from quantitative studies indicates that in general police-based diversion measures are effective in preventing criminal offending and show promising results for improving participants’ health and diminishing social costs as well as costs associated with processing drug-related offenses. There was insufficient evidence to draw conclusions about the effect of police-based diversion measures on drug use, drug accessibility, or changes in participants’ socioeconomic conditions. Findings from qualitative studies suggest that program acceptance by police officers, constructive intersectoral collaboration, clear eligibility criteria, and timely access to services seem to facilitate the implementation and delivery of police-based diversion measures.

Conclusion. Police-based diversion measures can be effective in preventing drug-related criminal offenses and harm. Additional research is needed to evaluate their effect on participants’ socioeconomic conditions and drug use as well as drug accessibility.

Key words: police-based diversion; drug use; systematic review; harm reduction, criminal offending


Drug use creates an important economic burden for the criminal justice and public health systems (Mazerolle et al., 2006; Rehm et al., 2006). While most jurisdictions have implemented prohibition and deterrence-based programs, evidence for the effectiveness of the “get tough on crime” approach is scarce at best. Reviews suggest that police operations have limited success in restricting illicit drug sales and markets (Mazerolle et al., 2006; Sherman, 1990) and some studies have suggested that intensifying law enforcement activities to limit drug use may have harmful health and social impacts (Kerr et al., 2005; Wood et al., 2004). Several experts have questioned the effectiveness of the prosecution-incarceration-supervision nexus in preventing drug use, the criminal recidivism of those convicted of drug offenses, or health-related problems associated with drug use (Green & Winik, 2010; Werb et al., 2011).

The calls, for initiatives to divert those who use drugs to social services rather than dealing with them within the criminal justice system, are increasing (Rouhani et al., 2019). Previous reviews have, however, focused mainly on post-conviction or post-sentence measures (e.g., Brown, 2010; Hayhurst et al., 2015), and therefore provide limited insights into the potential of diversion programs initiated at the police-level.

The present study synthesizes the evidence presented in evaluation studies of diversion measures initiated at the police-level and aimed at preventing criminalization and other harms associated with drug use. Our systematic review summarizes evidence found in quantitative and qualitative studies to respectively report the impact of these programs and to identify the barriers and facilitators associated with their implementation and delivery.

Police-based diversion programs

The call for alternatives to traditional sentencing emerged in the 1990s in the wake of a growing number of offenders being prosecuted and incarcerated for drug-related offenses (Harrison & Scarpitti, 2002) and increasing recognition that drug use and possession was not only a law enforcement problem but a public health issue (Warner & Kramer, 2009).

Diversion measures were designed to provide appropriate treatment, offer alternative means of managing drug-use offenses, and prevent drug-use and related criminal offenses (Goetz & Mitchell, 2006).

Drug treatment courts appear to be the most common type of diversion measures. Individuals who have a substance-related problem associated with criminal activities are offered the choice of entering a judicially supervised treatment programs rather than being incarcerated and charges are often dismissed upon completion of the program (Gottfredson et al., 2003; Wilson et al., 2006). However, the effectiveness of such diversion measures is not clear (Brown, 2010; Hayhurst et al., 2015; Stevens et al., 2022; Wilson et al., 2006; Wittouck et al., 2013).

While diversion can take place at several levels of the criminal justice system, previous reviews have mainly focused on post-conviction (e.g., drug courts, and other treatment alternatives to imprisonment) or post-sentence (e.g., California Proposition 36 or other services for probationers) alternatives (Hayhurst et al., 2015; Brown, 2010). According to the Sequential Intercept Model, diversion can take place at five intercept points: “(1) law enforcement and emergency services; (2) initial detention and initial hearings; (3) jail, courts, forensic evaluations, and forensic commitments; (4) reentry from jails, state prisons, and forensic hospitalization, and (5) community corrections and community support services” (Munetz & Griffin, 2006: 545). While the Sequential Intercept Model was initially developed to divert people with mental illness from the criminal justice system, it also applies to people who use drugs, especially since several “clients” from the criminal justice system have co-occurring disorders (e.g., alcohol and drug addiction as well as mental health problems).

Diversion measures at the first point of intercept – during encounters between police and people who use drugs – have been implemented in a growing number of jurisdictions (Stevens et al., 2022). Commissions for the Dissuasion of Drug Addiction (CDDA) are one example of police-based diversion measures articulated mainly around prevention and harm reduction (Hughes & Stevens, 2010). Portugal has drastically changed its approach to dealing with people who use drugs, to decrease drug-related offenses and the consequences associated with drug use. In July 2001, possession of any drug for personal use was decriminalized (de jure decriminalization), making drug use an administrative infringement rather than a criminal offense (Hughes & Stevens, 2010). Rather than arresting and charging individuals for using drugs, police officers direct them to one of the 18 local CDDAs, where their drug use problems and needs are assessed and the appropriate administrative sanction (e.g., community services, fine, interdiction against being places or with some people) if required, is determined, and contact with relevant services facilitated (Félix et al., 2017a).

Police-based diversion programs have also been implemented in jurisdictions where drug use is still a criminal offense. Inspired by the harm reduction approach, Law Enforcement Assisted Diversion (LEAD), which creates a de facto decriminalization of drug use, has been introduced in Los Angeles, Atlanta, New Orleans, and other cities in the United States1 to improve public safety by meeting the different needs of people who use drugs. Police officers in participating jurisdictions can offer individuals suspected of low-level drug and prostitution offenses the opportunity to participate in LEAD (Clifasefi et al., 2017). If they agree, after arrest, but prior to booking, eligible participants take part in a one-time diversion program. They are then referred to a case manager who connects them to services that help meet their stated goals and basic needs (e.g., shelter, food, housing, clothing) (Collins et al., 2019). Abstinence is not a condition for remaining in the program. LEAD involves a high level of coordination between program leaders, case managers, and the Office of the District Attorney to ensure that “all contacts with LEAD participants going forward, including new criminal prosecutions for other offenses, are coordinated with the service plan for the participant to maximize the opportunity to achieve behavioral change” (LEAD National Support Bureau, 2021: no pagination).

In Australia, police-based diversion programs rest on a mixture of de jure and de facto decriminalization (Hughes et al., 2016). Since the introduction of a national agreement to offer diversion to minor drug offenders in 1999, the number of police-based diversion programs has significantly increased in Australia (Hughes & Ritter, 2008), The most common type of police diversion programs is “cannabis cautioning”, where minor cannabis offenders are diverted away from the criminal justice system into education or treatment programs (Shanahan et al., 2017).

While several studies have found positive outcomes associated with initiatives such as LEAD and CDDAs (Collins et al., 2017, 2019; Félix et al., 2017a; Félix & Portugal, 2017b; Shanahan et al., 2017), concerns have also been expressed. First, these programs involve a coordinated collaboration between community partners and legal authorities, and it has been suggested that police officers may be reluctant to participate in activities seen as social work (Skogan & Hartnett, 1999; Worden & McLean, 2018). Second, citizens and other stakeholders may also be reluctant to participate in such programs if police officers are not seen as the legitimate “brokers” of such services (Goetz & Mitchell, 2006). Third, some studies have found that those involved in the operation of diversion programs, especially police officers, think that diversion measures trivialize drug use and could lead to increases in drug consumption (Worden & McLean, 2018), especially if alternatives to prosecution are not supported by campaigns promoting health and responsible drug use (Hughes & Stevens, 2007). At last, some concerns have been expressed about a possible net-widening effect (Stevens et al., 2022). The diversion measure could bring more people into contact with the criminal justice system than the “treatment as usual” condition. In fact, Christie and Ali (2009) reported a significant increase in people being prosecuted by the court following the introduction of the Cannabis Expiation Notice Scheme in South Australia.

Problem statement and the present study

As emphasized by Munetz and Griffin (2006), diversion can take place at various stages. Their Sequential Intercept Model shows the importance of conducting a systematic review of diversion alternatives offered by law enforcement agencies. Previous systematic reviews not only provide mixed evidence about the effects of diversion programs, but they have mainly focused on post-conviction or post-sentence measures (Brown, 2010; Hayhurst et al., 2015; Wittouck et al., 2013).

One exception is the work of Stevens et al. (2022) who have conducted a realist review of alternatives to criminalization for simple drug possession. Their review focused on contexts in which alternatives to criminalization operate, mechanisms likely to be triggered by these alternatives and their potential outcomes. Their main objective was not to critically appraise the quality of the evidence nor to conclude about the effects of these programs, but to propose a program theory by addressing issues relevant to end users (Stevens et al., 2022).

To complement the findings of previous reviews, we conducted a systematic review of police-based alternatives to the criminalization of drug use or possession. Our contribution to the actual literature is twofold. First, we synthesized the evidence about the effect of police-based measures for six categories of outcome: (1) health-related harms, (2) drug use, (3) criminal recidivism, (4) socioeconomic conditions, (5) drug availability, and (6) social costs and costs associated with the implementation of alternatives. Second, we reviewed results from qualitative and descriptive evaluations to identify barriers and facilitators associated with the implementation of diversion initiatives. The combination of both types of finding permits to determine if effective measures were successful at addressing barriers and facilitators associated with program implementation and service delivery.


This systematic review focuses on police-based diversion measures. The essential feature of these measures involves “the police initiating and leading the intervention and the [people who use drugs] receiving a diversionary scheme to avoid a criminal record and any consequences that may result from continued formal contact with the criminal justice system” (Wilson et al., 2018: 10). These interventions are offered at an individual’s first point of contact with the criminal justice system and provide an alternative to formal charges or court proceedings (Barberi & Taxman, 2019). Police-based diversion measures are often combined with programs such as educational sessions or treatment (Wilson et al., 2018) and involve community members and the social services of justice providers (Tallon et al., 2017). Hence, this review solely focuses on diversion measures that offer an alternative to a formal arrest (i.e., measures where arrest and criminal prosecution were only replaced by an administrative fine were not included).

The literature usually distinguishes between three types of police-based diversion measures. First, with pre-arrest diversion measures, police officers advise or order a suspect to participate in treatment or take part in a prevention program (Belenko, 1999). Second, pre-booking diversion occurs when an arrest has been made but formal charges are lifted if the suspect agrees to participate in a treatment or prevention program (Broner et al., 2002). At last, police-based diversion measures can be initiated by a self-referral. People show up at the police station and are then directed to community services and treatments (Schiff et al. 2017).

The present systematic review was conducted following PRISMA guidelines (Page et al., 2021). Our review focuses on results about the impact of police-based diversion measures and qualitative or descriptive findings documenting facilitators and barriers associated with their implementation. This strategy “allows the integration of quantitative estimates of benefit and harm with qualitative understanding from people’s lives” (Thomas et al., 2004: 1012). Both sets of evidence – qualitative and quantitative – can then be matched to determine whether effective police-based programs address these barriers and facilitators.

Eligibility criteria, information sources and search strategy

The search for evaluations of police-based diversion measures was conducted between July and December 2019. Key words such as “diversion”, “drugs”, “psychoactive substances”, “decriminalization”, and “evaluation” were used to search eight databases on EBSCOHost: (1) Medline; (2) Health Policy Reference Center; (3) PsycINFO; (4) SocINDEX; (5) Social Work Abstract Political Science Complete; (6) Public Affairs; (7) Criminal Justice Abstract; and (8) Index to Legal Periodicals Retrospective. Erudit was searched for potential French publications. Google and Google Scholar were also searched as well as registers of best practices (SAMHSA, Blueprint, Office of Juvenile Justice and Delinquency Prevention). A steering committee was created at the outset of the study, and they also provided the research team with information about relevant studies. Following comments from reviewers, other studies were also considered.

To be considered for review, studies had to include either qualitative or quantitative results related to the evaluation of a police-based diversion measure. Measures had to be designed to prevent harms associated with the criminalization of drug use, providing an alternative to arrest or conviction (e.g., referral to education sessions, guidance services, or treatment). Studies also had to provide sufficient information about the alternative measure in comparison to treatment as usual. Pre-booking, pre-arrest and self-referral diversion measures involving both de jure or de facto decriminalization were considered. Think tank reports, studies about drug trafficking, or studies published in languages other than French, English, or Spanish were excluded.

Selection and data collection processes

Two authors screened the title, summary, and key words of each publication and only those dealing with police-based diversion measures were retained. These studies were then assessed to make sure they presented the results of an evaluation. Data were extracted about the place where the program was implemented (e.g., country, state, or city), characteristics of the sample used (e.g., criminal history, mental health problems or other social vulnerabilities), analytical strategy, and results. Four qualitative and four quantitative studies were initially coded by two raters. Coding grids were then compared, and conflicts were discussed and resolved. The remaining studies were coded by a single rater.

Distinct strategies were used to code results from quantitative and qualitative studies. Main themes in qualitative studies were identified (e.g., perceived effectiveness, conditions and challenges associated with the implantation of programs) and their frequency in each study was computed. All estimations (relationships between the diversion measure and an outcome) from quantitative studies were coded as well as p-values, types of coefficients, and confidence intervals when available. The quality of research designs in quantitative studies was assessed on the Maryland Scientific Method Scale (SMS) (Farrington et al., 2002), which uses a 5-point system to rate the internal validity of study design (Table 1). Level 5 corresponds to randomized control trials, which have the strongest internal validity. Level 3 research designs, which include before and after measurements of comparable control and treatment groups, are considered the minimum standard from which conclusions can be drawn about the effect of a program on a particular outcome (Sherman et al., 1997). Research designs with scores of 1 or 2 have the weakest internal validity. 

[Table 1 about here]

Synthesis methods for quantitative evaluations

Most of the studies in our sample had a score of 1 or 2 on the SMS (Table 3), making it unfeasible to conduct a meta-analysis. Wide variation in types of outcomes made statistical summarization impossible (Petticrew & Roberts, 2008). The strategy proposed by Sherman et al. (1997) was therefore used to synthesize outcomes and determine whether police-based diversion measures were effective, not effective, or promising in preventing health problems, drug use, drug-related offenses, social costs, drug availability or improving social conditions. The vote-count approach was used to summarize the number of outcomes that were favorable, neutral, or unfavorable for a particular measure, regardless of their statistical significance. An outcome was considered favorable if, for instance, participants were less likely to recidivate following participation in the program. An increase in the recidivism rate was considered an unfavorable outcome while an absence of change was a neutral outcome. Scores on the SMS and “votes” were combined to determine what worked, what didn’t, and what was promising (Sherman et al., 1997). Effective measures had to meet two conditions: (1) outcomes were evaluated as favorable in at least two studies with a score of 3 or more on the SMS and (2) most outcomes, based on the vote-count method, were favorable for the measure. The measure was deemed ineffective if at least two studies with a score of 3 or more reported results unfavorable to the measure and most outcomes, based on the vote-count method, were neutral or unfavorable. A measure was considered promising when one study with a score of 3 or more on the SMS reported results favorable to the measure and most outcomes, based on the vote-count method, were favorable. In other situations, it is impossible to determine the effect of the measure on the set of outcomes considered.


Results are reported in three sections. The first section presents the study selection process and the characteristics of police-based diversion measures. The second section provides a synthesis of outcomes from quantitative evaluations. This section also discusses whether police-based diversion measures were effective, not effective, or promising in dealing with drug-related problems. The third section presents themes found in qualitative evaluations. These themes are articulated around two main dimensions: (1) perceived effectiveness of the measure and (2) conditions for its successful implementation and program delivery.

Study selection and characteristics of diversion programs

Figure 1 shows a flow chart of the study selection process. After removal of duplicates, 1,894 records were screened and 1,765 were excluded because they were not associated with an intervention, information about the measure was not available, or they were related to post-booking diversion measures. Careful assessment of the title, abstract, and key words of the remaining 132 records led to the exclusion of 56 additional records. The complete text of 76 publications was read and 27 were retained for review (12 peer-reviewed journal articles, one book chapter, two dissertations, and 11 technical reports).

[Figure 1 about here]

Table 2 presents the main characteristics of police-based diversion measures in the 27 studies retained. Eleven different diversion measures had been evaluated in these studies, which had been implemented in three countries: United-States (n = 17), Portugal (n = 5), and Australia (n = 5). See online supplementary material for a summary of each study.

Ten studies evaluated LEAD and five evaluated CDDAs in Portugal. Three assessed the Cannabis Cautioning Scheme in Australia. All other diversion measures were evaluated once. Nine of these other measures rely on de facto decriminalization rather than de jure, showing that legislative changes are not always necessary to implement police-based diversion measures. CDDAs and the Cannabis Cautioning Scheme involve de jure decriminalization.

Column 2 of Table 2 lists the actors involved in the delivery of the diversion measure. While the available information varied across studies, inter-agency collaboration appears to be a central feature of these alternatives. In most programs the police officer makes the initial contact, but the Gloucester Police Department’s Angel Program and the Safe Passage Initiative involve self-referral.

Column 3 (Table 2) provides information on the referral process, evaluation process, and service delivery. Health and social service professionals are often involved in assessing the needs and program eligibility of drug users and delivering a variety of services, including education, guidance, counselling, and treatment for substance use disorders (including medication-assisted treatment in some cases). In programs such as the Gloucester Police Department’s Angel Program and the Safe Passage Initiative, volunteers provide assistance and transportation to clients. In several diversion measures, the state’s attorney or equivalent is involved as charges are lifted once participants have successfully completed the program.

As seen in column 4 of Table 2, people must satisfy criteria for referral or enrollment in alternative measures. While eligibility criteria vary greatly between measures, two were common to several programs. Potential candidates must have possessed drugs only for their own consumption to participate in SHORT, the Cannabis Cautioning Scheme, the Adult Civil Citation Program, LEAD, the Drug Diversion Pilot Program, Queensland Police Diversion Program or CDDAs. People with a history of violent offending are not eligible to participate in LEAD, Safe Passage Initiative, Eau Claire County Pre-Charge Diversion, Queensland Police Diversion Program or SHORT.

As column 5 shows, some programs target the use of specific drugs while others deal with any drug use. CDDAs, LEAD, MDDP, and SHORT are offered to all offenders, regardless of type of drug used. Queensland Police Diversion Program, Eau Claire County Pre-Charge Diversion Program and the Cannabis Cautioning Scheme are offered only to cannabis users. Diversion programs offered by the Gloucester Police Department and the Safe Passage initiative specifically target opioid users.

Synthesis of outcomes associated with police-based diversion measures

Nineteen studies assessed the effects of police-based diversion measures. These 19 studies provide the results of 20 quantitative evaluations that resulted in 159 estimations. Outcomes were classified in six categories: (1) health of people who use drugs; (2) drug consumption; (3) socioeconomic conditions; (4) criminal recidivism or new contacts with the criminal justice system; (5) drug availability; and (6) social costs and costs associated with case management.

Table 3 shows the number of times an outcome was evaluated in investigating the effect of police-based diversion measures for each category of outcome. Scores on the SMS were then reported for each category of outcome and the classification of these outcomes (i.e., favorable, neutral, or unfavorable) was tabulated. Most evaluation studies have a score of 1 (n = 8; 40.0%) or 2 (n = 7; 35.0%) on the SMS. Two and three have scores of 3 and 4 respectively on the SMS; no randomized control trial was found. The 159 outcomes were generally favorable to the measures (n = 122; 76.7%) while 17.6% (n=28) were unfavorable. Close examination of the distribution of outcomes, however, suggests that it is important to consider outcome categories before drawing any conclusions about the effect of police-based diversions measures. Most outcomes are related to criminal offending and subsequent contacts with the criminal justice system.

[Table 3 about here]

Table 3 shows that health-related outcomes were measured 13 times in five studies (Félix et al., 2017; Hughes & Stevens, 2010; New Mexico Sentencing Commission, 2018; Yablon, 2011; Hales et al., 2004). Outcomes were related to: (1) drug-related deaths (n = 3); (2) health-threatening behaviors (n = 2); (3) mental health (n = 2); (4) and calls or visits to emergency services (n = 6). Except for one unfavorable effect, all effects were favorable (n = 12/13; 92.3%). Two studies had scores greater or equal to 3 on the SMS (Yablon, 2011; Felix et al., 2017) but a favorable outcome was reported for only one of these (Felix et al., 2017). Police-based diversion measures evaluated in these studies therefore qualify as promising for improving the health of people who use drugs since (1) all outcomes of studies with scores between 1 and 2 on the SMS are favorable to the measure and (2) one evaluation with a score of 3 or more on the SMS shows a favorable outcome.

Six studies evaluated the effects of police-based diversion measures on 17 outcomes related to drug consumption (Hughes & Stevens, 2010; New Mexico Sentencing Commission, 2018; Schiff et al., 2017; Shanahan et al., 2016, 2017; Hales et al., 2004) and assessed the impact of CDDAs in Portugal, LEAD, Queensland Police Diversion Program, and the Australian Cannabis Cautioning Scheme. Most outcomes were related to drug consumption habits (n = 15) while others looked at abstinence (n = 1) and days of methadone treatment (n =1). Most outcomes (n = 11; 64.7%) suggest that diversion measures positively influenced drug consumption among participants. Since all studies had scores below 3 on the SMS, it was impossible to draw any conclusions about the impact of police-based diversion measures on drug consumption.

Four studies evaluated the effects of police-based diversion measures on participants’ socioeconomic conditions (Clifasefi et al., 2017; New Mexico Sentencing Commission, 2018; Shanahan et al., 2016; Hales et al., 2004). Outcomes in these evaluations were associated with: (1) licit sources of revenues (n = 6); (2) perceived quality of life (n = 1); (3) housing (n = 3); and (4) quality of social relationships (n = 2). All outcomes were favorable to the intervention. However, all studies had a score of 2 on the SMS, so it is not possible to draw conclusions about the impact of the measures examined on the socioeconomic conditions of participants.

Table 3 shows that 14 evaluations reported 87 outcomes about the effect of police-based diversion measures on recidivism or any new contact with the criminal justice system (Kopak, 2019; Baker & Goh, 2004; Callister & Braaten, 2016; Collins et al., 2017, 2019; Cotti & Haley, 2014; Cunningham & Stone, 2011; Dole & Freeman, 2018; Félix et al., 2017; Hughes & Stevens, 2010; New Mexico Sentencing Commission, 2018; Shanahan et al., 2016; Yablon, 2011; Hales et al., 2004). The effect of diversion on criminal recidivism was measured principally by the number of new arrests, charges, or convictions, without necessarily distinguishing between types of offense (n = 63). Seventeen outcomes were associated with new drug-related offenses (e.g., possession, consumption, or distribution of drugs). Seven outcomes appraised the effect of the diversion measure on incarceration. Most outcomes (67/87; 77.0%) were favorable to the measure according to the vote-count approach. Five outcomes were neutral (5.7%) and 15 (17.2%) were unfavorable. Four studies have a score equal or greater to 3 on the SMS (Yablon, 2011; Felix et al., 2017; Collins et al., 2017, 2019). These four studies reported 26 outcomes, 88.5% of which were favourable to the measure. Three of these studies found a decrease in criminal recidivism (Felix et al., 2017; Collins et al., 2017, 2019). Since two studies or more with a score equal or greater to 3 on the SMS report favorable outcomes, and most outcomes from level 1 and 2 studies are in favor of the measure, police-based diversion measures evaluated in these studies can be considered effective in preventing criminal offending or future contacts with the criminal justice system.

Six studies assessed the effect of police-based diversion measures on costs associated with case management and on social and health-related costs (Baker & Goh, 2004; Collins et al., 2017, 2019; Gonçalves et al., 2015; Shanahan et al., 2016, 2017). These studies looked at the effects of LEAD, the Cannabis Cautioning Scheme, and CDDAs on 17 outcomes. These outcomes were associated with efficiency (i.e., the ratio between costs associated with participating in the measure and costs associated with the criminal processing of a case) (n = 7) and social costs (e.g., costs for the healthcare system or in lost productivity) (n = 10). Most outcomes were associated with decreased costs (n = 14; 82.4%). Other outcomes were rated either neutral (n = 2; 11.8%) or unfavorable (n = 1; 5.9%). Most studies had SMS scores lower than three. Collins et al. (2017) had a level 3 research design, and they estimate that LEAD saves the criminal justice system about US $7,000.00 per participant when compared to the “treatment as usual” condition. Based on the distribution of outcomes and scores on the SMS, police-based diversion measures appear promising in lowering costs associated with the criminal processing of drug use cases or other social-related costs.

Themes in qualitative studies: perceived effectiveness and conditions for effectiveness

Qualitative and descriptive results were reported in 13 studies and provide rich information about challenges experienced at the implementation and service-delivery stages (see online supplemental material for a detailed description of qualitative studies). Twelve themes were extracted (Table 4) and classified in two main categories: (1) perceived effectiveness and (2) facilitators and barriers to the implementation and delivery of the program.

Perceived effectiveness

Clients’ perceptions of effectiveness were reported in three studies (Clifasefi & Collins, 2016; Denman, 2018; Schiff et al., 2017); 40.9% of clients in the Angel Program and 44.0% of those in LEAD stated that they remained sober following their participation in the diversion program. Clients reported that they noticed improvements in self-esteem, relationships with others, autonomy, optimism, and quality of life. About half the clients mentioned that they had had no further contact with the police or that when contacts did occur, they were more positive. Clients interviewed by Clifasefi and Collins (2016) reported that LEAD was the first initiative that had provided services that took their needs into account.

Four studies documented how diversion programs were seen by police officers (McLeod & Stewart, 1999; Reichert et al., 2017; Worden & McLean, 2018; Hales et al., 2004). Findings suggest that police officers base their evaluation of effectiveness on whether the initiative decreases criminal recidivism. Police officers interviewed in two studies acknowledged that police-based diversion programs propose a concrete solution to a complex problem by preventing drug use and involvement in criminal activities (Reichert et al., 2017; McLeod & Stewart, 1999). Conversely, half of police officers surveyed by Hales et al. (2004) considered that police-based diversion measures do not reduce crime and drug use.

Five studies looked at the perceived effectiveness of police-based diversion programs by partners or service providers from healthcare or community agencies (McCleod & Stewart, 1999; Reichert et al., 2017; Magana, 2019; Hughes & Stevens, 2010; Hales et al., 2004). In general, service providers considered that diversion measures are a major development in supporting drug users. Two studies reported that collaboration between actors from the healthcare and criminal justice systems was related to the effectiveness of diversion programs (Magana, 2019; Reichert et al., 2017). One study, on CDDAs in Portugal, reported that police-based programs were seen as an opportunity to detect eventual health problems among vulnerable drug users (e.g., users with HIV or at risk of overdose) and offer early intervention (Hughes & Stevens, 2010).

Three studies reported potential adverse effects associated with police-based diversion measures. Stakeholders interviewed by Hughes and Stevens (2010) reported that CDDAs were associated by an increase in drug use, presumably because they were not accompanied by education and awareness campaigns. Police officers sometimes considered that diversion measures do not send a sufficiently strong message about reducing or ceasing drug use. These programs would also decrease the credibility of the police (Hales et al., 2004). Some key informants stressed that the Cannabis Cautioning Scheme in Australia was associated with a net-widening effect, that is an increase in the penalization of people who use drugs (Baker & Goh, 2004).

Program acceptability by police chiefs and officers was addressed in five studies (Reichert et al., 2017; Clifasefi & Collins, 2016; McLeod & Stewart, 1999; Barberi & Taxman, 2019; Hales et al., 2004). Numerous officers emphasized the relevance of police-based diversion measures (Reichert et al., 2017; Clifasefi & Collins, 2016; McLeod & Stewart, 1999). For instance, 86% of police officers who took part in the Safe Passage Initiative supported the program (Reichert et al., 2017). Several officers also mentioned, however, that diverting drug users toward health resources goes beyond their mandate and that this was the main reason for the low acceptability of diversion initiatives among some officers (Worden & McLean, 2018; Barberi & Taxman, 2019; McLeod & Stewart, 1999; Reichert et al., 2017). Some police officers thought that diversion programs make the seriousness of problems associated with drug use seem less important. Acceptability among officers was increased when high ranking officers were seen to believe in the potential of diversion programs (Baker & Goh, 2004; Barberi & Taxman, 2019; Reichert et al., 2017).

Themes related to training were addressed in six studies. Since objectives associated with the harm reduction approach are not always well understood by the police (Barberi & Taxman, 2019), stakeholders reported that providing full and continuous training is essential to ensure the effectiveness of diversion programs (Hales et al., 2004). Training is also seen as a way to increase acceptability among the police (Beckett, 2014; Magana, 2019; Reichert et al., 2017; Worden & McLean, 2018).

Three studies indicated that the success associated with LEAD rests, among other things, on effective communication and intersectoral collaboration (Reichert et al., 2017; Becket, 2014; New Mexico Sentencing Commission, 2018). Community partners and members – including supportive peers and caregivers – increase the effectiveness of diversion programs (Barberi & Taxman, 2019; Reichert et al., 2017; Becket, 2014). Informing police officers about the effectiveness of diversion programs was also seen as way to improve their experience and participation in such programs (Becket, 2014). Using two distinct computer systems, using a fax to send the compliance sheet, and rescheduling appointments with the former officer in charge of the diversion were seen as barriers (Hales et al., 2004).

Five studies addressed issues related to eligibility criteria (McCleod & Stewart, 1999; Worden & McLean, 2018; Schiff et al., 2017; Baker & Goh, 2004; Hales et al., 2004). All studies emphasize the importance of having clear eligibility criteria that are well understood by all those providing services to avoid discrimination or differential treatment. It is also important that those providing services are aware of and committed to making use of diversion programs. For example, Worden and McLean (2018) analyzed cases of arrest for drug use and found that only 43 (7.9%) of 543 participants who met the eligibility criteria had been referred to LEAD as only three of the 244 officers aware of the program were diverting drug offense cases to LEAD. Three studies reported that inclusion criteria should not limit access to those who seem likely to take advantage of diversion programs (McLeod & Stewart, 1999; Schiff et al., 2017; Hales et al., 2004). For example, potential clients with a criminal record are sometimes excluded from programs without sound justification. Similarly, diversion programs should not be limited to adults to prevent future drug use problems among teenagers (McLeod & Stewart, 1999).

[Table 4 about here]

According to four studies, a simple admission process with activities coordinated between all service providers seems to increase the rate of participation (Barberi & Taxman, 2019; Magana, 2019; McLeod & Stewart, 1999; Reichert et al., 2017). Some clients interviewed by McLeod & Stewart (1999) complained about the number of service providers they had been involved with.

Eight studies highlight the importance of informing potential partners and clients about the existence of the program (Barberi & Taxman, 2019; McLeod & Stewart, 1999; Reichert et al., 2017; Schiff et al., 2017; New Mexico Sentencing Commission, 2018; Clifasefi & Collins, 2016; Hughes & Stevens, 2010; Becket, 2014). Some potential participants refused to take part in LEAD for fear of being stigmatized or judged by their peers, indicating the importance of properly informing the target group about the objectives of the program (Clifasefi & Collins, 2016; New Mexico Sentencing Commission, 2018).

The importance of developing quality relationships among partners and with potential program users was addressed in five studies (Barberi & Taxman, 2019; Clifasefi & Collins, 2016; Denman, 2018; Magana, 2019; New Mexico Sentencing Commission, 2018). Most of the 48 clients interviewed by Magana (2019) noted that service providers were empathetic, engaged, and caring. Clients used terms such as “available”, “engaged”, and “non-judgmental” to describe their relationships with police officers (Schiff et al., 2017) and indicated that they appreciated relationships based on compassion and where they did not feel they were being treated like criminals (Barberi & Taxman, 2019). Relationships with service providers that involved conflict or feeling stigmatized were identified as important barriers by some clients (Barberi & Taxman, 2019; New Mexico Sentencing Commission, 2018). Clients identified the provision of personal assistance at all stages of the intervention as well as a structured treatment strategy as among the most appreciated components of LEAD as they helped reinforce autonomy and strengthen relationships with case managers (Clifasefi & Collins, 2016; Magana, 2019).

Availability and timely access to services were found to facilitate program delivery and participants’ retention in eight studies (McLeod & Stewart, 1999; Schiff et al., 2017; Barberi & Taxman, 2019; Magana, 2019; Becket, 2014; New Mexico Sentencing Commission, 2018; Hughes & Stevens, 2010); Hales et al., 2004). Police officers note that, unlike community and healthcare services, they deal with potential clients on a 24/7 basis. This is an important point since timely evaluation and referral have been found to lead to increased adherence to treatment and higher retention rates (McLeod & Stewart, 1999). Clients also expressed the importance of being able to reach their case manager quickly (New Mexico Sentencing Commission, 2018). Diversion programs also require access to sufficient and diversified resources (e.g., housing resources, drug treatment, and mental health services) (Beckett, 2014). Reminder contacts and informing the clients about the content of the measure could increase compliance (Hales et al., 2004). Measures implemented at the state or country level need to address issues related to access to services in remote areas (Hales et al., 2004).

Policy implications and conclusion

The present study focused on two specific objectives: (1) reviewing evidence about the impact of police-based diversion measures on several categories of outcome and (2) identifying facilitators and barriers to the implementation and delivery of these measures. Our finding indicate that police-based diversion measures are effective in preventing criminal offending or other contacts with the criminal justice system. Police-based measures were found to be promising to improve health and decrease social and health-related costs associated with drug use as well as expenses associated with the prosecution of drug-related offenders.

Studies using quasi-experimental (level 3 or 4 on the SMS) or experimental designs (level 5 on the SMS) were required to conclude about the impact of police-based diversion programs (Sherman et al., 2017). As highlighted by previous reviews on diversion measures for drug-related offenses (Brown, 2010; Hayhurst et al., 2015; Wilson et al., 2006), few evaluations used such designs. Accordingly, we were unable to draw conclusions about the effect of police-based diversion measures on drug consumption, drug accessibility, and socioeconomic conditions of clients. These outcomes were only assessed in studies with scores of 1 or 2 on the SMS.

The combination of qualitative and quantitative findings in systematic reviews allows to compare ineffective and effective programs based on their ability to address barriers and facilitators (Thomas et al., 2004). While our review concludes that police-based diversion programs are effective in preventing offending and future contact with the criminal justice system, only LEAD and CDDAs have been evaluated with a quasi-experimental design. Qualitative findings were only reported in one study on CDDAs (Hughes & Stevens, 2010), and few barriers or facilitators were addressed. Qualitative and descriptive findings were found in six evaluations of LEAD. Future partners involved in the implementation and delivery of LEAD should consider the following facilitators: (1) have the support of police chiefs who believe that diversion is effective and a part of the police mandate (Clifasefi & Collins, 2016; Magana, 2019); (2) train and educate officers about diversion (Magana, 2019); (3) develop guidelines to promote effective communication between partners and intersectoral collaboration (New Mexico Sentencing Commission, 2018; Beckett, 2014); (4) have simple and clear diversion criterion (Worden & McLean, 2018); (5) establish a simple diversion and service delivery process to enhance participation and retention (Magana, 2019; and (6) offer services based on drug users’ needs (Clifasefi & Collins, 2016). Participants interviewed about LEAD were confident about its effectiveness (Magana, 2019; Clifasefi & Collins, 2016). Facilitators found in LEAD studies were also reported in other evaluations of police-based diversion measures targeting vulnerable populations (e.g., people with mental health problems) (Kirst et al., 2015; Robertson et al., 2019; Landry et al., 2018). At last, issues related to stigmatization represent a barrier. The program needs to receive positive publicity within the community. Some clients were not comfortable taking part in LEAD because of negative perceptions about the program in their community (Denman, 2018; New Mexico Sentencing Commission, 2018).

Results from our systematic review show that police-based diversion measures represent a promising feature for a public health or a criminal justice policy seeking to prevent drug-related harms. Their flexibility appears to be one of their greatest strengths. First, police-based diversion programs have been developed for specific drug types such as cannabis (Hales et al., 2004), opioid users (Callister & Braaten, 2016) or drug use in general (Felix et al., 2017). Second, police-based diversion measures are adapted to the user needs. Depending on their needs, people diverted from the criminal justice system can attend education sessions (Hales et al., 2004; Shanahan et al, 2017), be offered treatments (Schiff et al., 2017) or other types of training (Collins et al., 2017). Third, de jure decriminalization as in Portugal or Australia (Hughes & Stevens, 2010; Shanahan et al., 2017) is not always necessary as several measures involve de facto decriminalization (Schiff et al., 2017; Collins et al., 2019). Hence, diversion measures can be implemented at the country, state, or community level.

Despite promising results, additional research is highly needed on some issues. First and as previously mentioned, additional evaluations using experimental and quasi-experimental designs are needed to estimate the impact of police-based diversion programs on drug accessibility, drug use, and socioeconomic conditions of program participants. Studies with strong internal validity could also strengthen conclusions about the effect of police-based diversion programs on participants’ health, and on programs’ costs and savings (measures are considered promising for these two outcomes).

Second, additional research is needed on processes associated with positive outcomes. LEAD, for example, is aimed at connecting clients with community resources such as housing placement, medical care, legal advocacy, job training, mental health counselling, and treatment for chemical dependency (Collins et al., 2017). While LEAD itself was thoroughly evaluated, its effects on clients’ knowledge on prostitution, drug-related harms, and other legal issues were not evaluated. Similarly, LEAD seeks to reintegrate drug users into the community. Hence, LEAD’s impact on participants’ job skills or ability to avoid at-risk drug consumption has yet to be documented. Conducting such research would complement actual knowledge on program mechanisms (Steven et al., 2022) and assist in developing a sound theory of change (Mayne, 2015; Callister & Braaten, 2016).

At last, additional research is needed to precisely appreciate costs and benefits associated with police-based diversion measures. Net-widening is a major concern related to diversion measures (Cohen, 1979; Baker & Goh, 2004), especially when diversion involves conditions and target low-risk offenders. Only two police-based diversion measures offered to opioid users involved self-referral with no pending administrative or criminal charges (Reichert et al., 2017; Schiff et al., 2017). Concerns about the net-widening was only documented in one study (Baker & Goh, 2004). Some key informants interviewed by Baker and Goh (2004) mentioned that cannabis cautions had replaced informal warnings to some extent, especially for people carrying a small amount of cannabis.


Baker, J., & Goh, D. (2004). The cannabis cautioning scheme three years on: An implementation and outcome evaluation. NSW Bureau of Crime Statistics and Research Sydney.

Barberi, D., & Taxman, F. S. (2019). Diversion and alternatives to arrest: A qualitative understanding of police and substance users’ perspective. Journal of Drug Issues, 49(4), 703‑717.

Beckett, K. (2014). Seattle’s Law Enforcement Assisted Diversion Program: Lessons learned from the first two years. University of Washington Seattle, WA.

Belenko, S. (1999). The challenges of integrating drug treatment into the criminal justice process. Albany Law Review., 63, 833.

Broner, N., Borum, R., & Gawley, K. (2002). Criminal Justice Diversion of Individuals with Co-Occurring. In G. Landsberg, M. Rock. L. K.W. Berg and A. Smiley (Eds.), Serving Mentally Ill Offenders: Challenges & Opportunities for Mental Health Professionals, pp. 83-106. Springer.

Brown, R. T. (2010). Systematic review of the impact of adult drug-treatment courts. Translational Research, 155(6), 263‑274.

Callister, S., & Braaten, A. (2016). An Evaluation of the Eau Claire County Pre-Charge Diversion Program as Measured by Three Year Recidivism Rates.

Christie, P., & Ali, R. (2009). Offences under the Cannabis Expiation Notice scheme in South Australia. Drug and Alcohol Review, 19, 251-256.

Clifasefi, S. L., & Collins, S. E. (2016). LEAD Program Evaluation: Describing LEAD Case Management in Participants’ Own Words. University of Michigan - Harborview Medical Center.

Clifasefi, S. L., Lonczak, H. S., & Collins, S. E. (2017). Seattle’s Law Enforcement Assisted Diversion (LEAD) program: Within-subjects changes on housing, employment, and income/benefits outcomes and associations with recidivism. Crime & Delinquency, 63(4), 429‑445.

Collins, S. E., Lonczak, H. S., & Clifasefi, S. L. (2017). Seattle’s Law Enforcement Assisted Diversion (LEAD): Program effects on recidivism outcomes. Evaluation and Program Planning, 64, 49‑56.

Collins, S. E., Lonczak, H. S., & Clifasefi, S. L. (2019). Seattle’s law enforcement assisted diversion (LEAD): Program effects on criminal justice and legal system utilization and costs. Journal of Experimental Criminology, 15(2), 201‑211.

Cotti, C. D., & Haley, M. R. (2014). Estimating the effectiveness of a misdemeanor drug diversion program using propensity score matching and survival analyses. The Social Science Journal, 51(4), 638‑644.

Cohen, S. (1979). The Punitive City: Notes on the Dispersal of Social Control. Crime, Law and Social Change, 3(4), 339-363.

Cunningham, D. S., & Stone, W. E. (2011). Effects of a Drug Diversion Court on Client Recidivism. Dans B. Sims (Ed.), Substance abuse treatment with correctional clients: Practical implications for institutional and community setting (p. 135‑149). Routledge.

Denman, K. (2018). Evaluation of Santa Fe’s LEAD Program: The Client Perspective. New Mexico Sentencing Commission.

Dole, J., & Freeman, L. (2018). Evaluation of Santa Fe’s LEAD Program: Criminal Justice Outcomes. New Mexico Sentencing Commission.

Farrington, D. P., Gottfredson, D. C., Sherman, L. W., & Welsh, B. C. (2002). The Maryland Scientific Methods Scale. In D.P Farrington, D. L. MacKenzie, L.W. Sherman & B.C. Welsh (Eds.), Evidence-based crime prevention, pp. 27‑35. Routledge.

Félix, S., Portugal, P., & Tavares, A. S. (2017). Going after the addiction, not the addicted: The impact of drug decriminalization in Portugal.

Félix, S., & Portugal, P. (2017). Drug decriminalization and the price of illicit drugs. International Journal of Drug Policy, 39, 121‑129.

Goetz, B., & Mitchell, R. E. (2006). Pre-arrest/booking drug control strategies: Diversion to treatment, harm reduction and police involvement. Contemporary drug problems, 33(3), 473‑520.

Gonçalves, R., Lourenço, A., & da Silva, S. N. (2015). A social cost perspective in the wake of the Portuguese strategy for the fight against drugs. International Journal of Drug Policy, 26(2), 199‑209.

Gottfredson, D. C., Najaka, S. S., & Kearley, B. (2003). Effectiveness of drug treatment courts: Evidence from a randomized trial. Criminology & Public Policy, 2(2), 171‑196.

Green, D. P., & Winik, D. (2010). Using random judge assignments to estimate the effects of incarceration and probation on recidivism among drug offenders. Criminology, 48(2), 357‑387.

Hales, J., Mayne, M., Swan, A., Alberti, S., & Ritter, A. (2004). Evaluation of Queensland Illicit Drug Diversion Initiative (QIDDI) Police Diversion Program: Final Report. Queensland Health & Queensland Police Service.

Harrison, L. D., & Scarpitti, F. R. (2002). Introduction: Progress and issues in drug treatment courts. Substance Use & Misuse, 37(12‑13), 1441‑1467.

Hayhurst, K. P., Leitner, M., Davies, L., Flentje, R., Millar, T., Jones, A., King, C., Donmall, M., Farrell, M., & Fazel, S. (2015). The effectiveness and cost-effectiveness of diversion and aftercare programmes for offenders using class A drugs: A systematic review and economic evaluation. Health technology assessment, 19(6), 1‑168, vii.

Hughes, C. E., & Stevens, A. (2007). The effects of the decriminalization of drug use in Portugal. Beckley Foundation Drug Policy Programme - Briefing Paper 14.

Hughes, C.E., & Ritter, A. (2008). A summary of diversion programs for drug and drug-related offenders in Australia. DPMP monograph, vol. 16, Sydney: National Drug and Alcohol Research Centre.

Hughes, C. E., & Stevens, A. (2010). What can we learn from the Portuguese decriminalization of illicit drugs? The British Journal of Criminology, 50(6), 999‑1022.

Kirst, M., Pridhan, K. F., Narrandes, R., Matheson, F., Niedra, K., & Stergiopoulos, V. (2015). Examining implementation of mobile, police-mental health crisis intervention teams in a large urban center. Journal of Mental Health, 24, 369‑374.

Kopak, A.M. (2019). An Initial Assessment of Leon County Florida’s Pre-Arrest Adult Civil Citation Program. Journal of Behavioral Health Services & Research, 46, 177-186.

Landry, M., Blais, É., Chapdelaine, A., Carrier, S., & Elazhary, N. (2018). An evaluation of the implementation and effects of a mobile crisis intervention team intervening among individuals in crisis in the city of Sherbrooke (in French). Primary Health Care and Social Services University Institute.

LEAD National Support Bureau (2021).

Magana, E. J. (2019). A Process Evaluation of LEAD (Law Enforcement Assisted Diversion) San Francisco (SF). California State University.

Mayne, J. (2015). Useful theory of change models. Canadian Journal of Program Evaluation, 30(2), 119-142.

Mazerolle, L., Soole, D. W., & Rombouts, S. (2006). Street-level drug law enforcement: A meta-analytical review. Journal of experimental criminology, 2(4), 409‑435.

McLeod, J., & Stewart, G. (1999). Evaluation of the Drug Diversion Pilot Program. Drugs and Health Protection Services Branch, Public Health Division, Department of Human Services.

Munetz, M. R., & Griffin, P. A. (2006). Use of the sequential intercept model as an approach to decriminalization of people with serious mental illness. Psychiatric services, 57(4), 544‑549.

New Mexico Sentencing Commission. (2018). Santa Fe Law Enforcement Assisted Diversion (LEAD): An Analysis of the Pilot Phase Outcomes. New Mexico Sentencing Commission.

Page, M. J., McKenzie, J. E., Bossuyt, P. M., Boutron, I., Hoffmann, T. C., Mulrow, C. D., Shamseer, L., Tetzlaff, J. M., Akl, E. A., & Brennan, S. E. (2021). The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. Bmj, 372, n71.

Petticrew, M., & Roberts, H. (2008). Systematic reviews in the social sciences: A practical guide. John Wiley & Sons.

Rehm, J., Taylor, B., & Room, R. (2006). Global burden of disease from alcohol, illicit drugs and tobacco. Drug and alcohol review, 25(6), 503‑513.

Reichert, J., Gleicher, L., Mock, L., Adams, S., & Lopez, K. (2017). Police-led referrals to treatment for substance use disorders in rural Illinois: An examination of the Safe Passage Initiative. Illinois Criminal Justice Information Authority.

Robertson, J., Fitts, M. S., Petrucci, J., McKay, D., Hubble, G., & Clough, A. R. (2019). Cairns mental health co-responder project: Essential elements and challenges to programme implementation. International Journal of Mental Health Nursing, 450‑459.

Rouhani, S., Gudlavalleti, R., Atzmon, D., Park, J. N., Olson, S. P., & Sherman, S. G. (2019). Police attitudes towards pre-booking diversion in Baltimore, Maryland. International Journal of Drug Policy, 65, 78‑85.

Schiff, D. M., Drainoni, M.-L., Weinstein, Z. M., Chan, L., Bair-Merritt, M., & Rosenbloom, D. (2017). A police-led addiction treatment referral program in Gloucester, MA: Implementation and participants’ experiences. Journal of substance abuse treatment, 82, 41‑47.

Shanahan, M., Hughes, C. E., McSweeney, T., & Griffin, B. A. (2017). Alternate policing strategies: Cost-effectiveness of cautioning for cannabis offences. International Journal of Drug Policy, 41, 140‑147.

Shanahan, M., Hughes, C., & McSweeney, T. (2016). Australian police diversion for cannabis offences: Assessing program outcomes and cost-effectiveness. Canberra: National Drug Law Enforcement Fund.

Sherman, L. W. (1990). Police crackdowns: Initial and residual deterrence. Crime and justice, 12, 1‑48.

Sherman, L. W., Gottfredson, D. C., MacKenzie, D. L., Eck, J., Reuter, P., & Bushway, S. (1997). Preventing crime: What works, what doesn’t, what’s promising : A report to the United States Congress. National Institute of Justice Washington, DC.

Skogan, W. G., & Hartnett, S. M. (1999). Community policing, Chicago style. Oxford University Press.

Stevens, A., Hughes, C. E., Hulme, S., & Cassidy, R. (2022). Depenalization, diversion and decriminalization: A realist review and programme theory of alternatives to criminalization for simple drug possession. European journal of criminology, 19(1), 29-54.

Tallon, J. A., Spadafore, J. C., & Labriola, M. (2017). Creating off-ramps: Lessons learned from police-led diversion programmes. Policing: A Journal of Policy and Practice, 11(2), 214‑228.

Thomas, J., Harden, A., Oakley, A., Oliver, S., Sutcliffe, K., Rees, R., Brunton, G., & Kavanagh, J. (2004). Integrating qualitative research with trials in systematic reviews. Bmj, 328(7446), 1010‑1012.

Warner, T. D., & Kramer, J. H. (2009). Closing the revolving door? Substance abuse treatment as an alternative to traditional sentencing for drug-dependent offenders. Criminal Justice and Behavior, 36(1), 89‑109.

Werb, D., Rowell, G., Guyatt, G., Kerr, T., Montaner, J., & Wood, E. (2011). Effect of drug law enforcement on drug market violence: A systematic review. International Journal of Drug Policy, 22(2), 87‑94.

Wilson, D. B., Brennan, I., & Olaghere, A. (2018). Police‐initiated diversion for youth to prevent future delinquent behavior: A systematic review. Campbell Systematic Reviews, 14(1), 1‑88.

Wilson, D. B., Mitchell, O., & MacKenzie, D. L. (2006). A systematic review of drug court effects on recidivism. Journal of Experimental Criminology, 2(4), 459‑487.

Wittouck, C., Dekkers, A., De Ruyver, B., Vanderplasschen, W., & Vander Laenen, F. (2013). The impact of drug treatment courts on recovery: A systematic review. The Scientific World Journal, 2013, 1-12.

Wood, E., Spittal, P. M., Small, W., Kerr, T., Li, K., Hogg, R. S., Tyndall, M. W., Montaner, J. S., & Schechter, M. T. (2004). Displacement of Canada’s largest public illicit drug market in response to a police crackdown. Cmaj, 170(10), 1551‑1556.

Worden, R. E., & McLean, S. J. (2018). Discretion and diversion in Albany’s LEAD program. Criminal Justice Policy Review, 29(6‑7), 584‑610.

Yablon, D. R. (2011). The effect of drug decriminalization in Portugal on homicide and drug mortality rates. University of California, Berkeley.


Table 1. Maryland Scientific Method Scale (SMS)

  1. Criteria used to determine the effectiveness of interventions


At least two studies with a score of ≥ 3 on the SMS report outcomes favorable to the intervention, and most outcomes of studies with lower SMS scores are favorable


At least two studies with a score of ≥ 3 on the SMS report outcomes that are unfavorable to the intervention, and most outcomes in other studies are unfavorable


One study with a score of ≥ 3 on the SMS scale reports outcomes favorable to the intervention, and most outcomes of other studies are favorable to the intervention

  1. Research design characteristics and score on the SMS


Criteria used to score research designs


Correlation between a police-based diversion measure and an outcome at a single time


Outcome measured before and after the police-based diversion measure but no comparable control


Outcome of treatment measured before and after a police-based diversion measure with comparable control conditions


Outcome measured before and after a police-based diversion measure in multiple treatment and control units, and control for other variables likely to influence the outcome


Random assignment of treatment and control conditions to units.

Adapted from:

Farrington, D.P., Gottfredson, D.C., Sherman, LW., & Welsh, B.C. (2002). The Maryland Scientific Methods Scale. In L.W. Sherman, D.P. Farrington, B.C. Welsh, et D.L. MacKenzie (eds.), Evidence-based crime prevention, pp. 13-21. Rouledge, London.

Table 2. Characteristics of police-based diversion measures

Program, type of measure, and places where implemented (number of studies)

Actors involved in program delivery

Referral process, evaluation process, and services delivered

Eligibility criteria

Types of drugs – Offenses targeted

Commissions for the Dissuasion of Drug Addiction (CDDA)

Pre-booking: de jure decriminalization


(n = 5)

Police officers, health and social service professionals, members of the criminal justice systems

Drug use decriminalized, and an administrative citation issued for drug use. Following a citation, people who use drugs can be sent to CDDA responsible for adjudicating administrative drug offenses and imposing sanctions. People are evaluated by a committee of three experts. The committee determines the appropriate sanction (e.g., warning, treatment for dependent users, or fines, community services or provisional suspension of proceedings for non-addict).

  1. People with drugs in their possession sufficient for a maximum of 10 days’ consumption (Quantities vary according to drug type)

All drugs – using, possessing, or buying drugs for personal consumption

Law Enforcement Assisted Diversion (LEAD)

Pre-booking: de facto decriminalization

Seattle, WA; Santa Fe, NM; San Francisco, CA; and Albany, NY

(n = 10)

Police officers, program case manager and professionals from agencies with different backgrounds (e.g., social work, nursing, dependency counseling)

Once arrested or referred by police officers, people are brought to the police precinct and screened for LEAD eligibility. Eligible individuals are offered the option to participate in LEAD instead of standard criminal booking and prosecution. Clients are directed to appropriate community services (e.g., housing placement, medical care, legal advocacy, job training, mental health counseling, and drug treatment).

  1. Possessing less than 3 or 6 grams (varies among sites)

  2. Receptive to the program

  3. Not involved in drug trafficking

  4. 18 or over

  5. Not involved in pimping

  6. No history of violent crime

  7. Not already enrolled in a program

All drugs – (1) drug use, possession, and buying drugs for personal use, (2) prostitution

Gloucester Police Department’s Angel Program

Self-referral: de facto decriminalization

Gloucester – north of Boston, MA

(n = 1)

Police officers, staff trained to complete assessment, volunteers, and treatment providers

Police offer a no-arrest voluntary screening and referral service to people with opioid use disorder. Following assessment, police officers call appropriate treatment centers. (e.g., outpatient counselling, hospitalization, intensive outpatient, inpatient residential) The participant is transported to the service.

  1. No active arrest warrants

  2. No acute medical safety concerns

Opioid – no offense as people with an opioid use disorder must show up (self referral) at the police station

Safe Passage Initiative

Self-referral: de facto decriminalization

Lee County, IL

(n = 1)

Police officers (but people who use drugs must make the initial contact, e.g., by contacting the substance abuse hotline, Facebook Page, emergency department or talking to an officer), nine treatment providers, state attorneys, and volunteers (for transportation)

People who use drugs must show up at police station to complete an intake form and verify eligibility prior to transport to a treatment facility. Services offered include detoxification, outpatient care, intensive outpatient care, partial hospitalization, inpatient/residential services, and medication assisted treatment. A hotline is available to help clients find recovery support group meetings.

  1. No record of violent crime

Opioid – no offense as people with an opioid use disorder must show up (self-referral) at the police station

Adult Civil Citation Program

Pre-arrest: de facto decriminalization

Leon County, FL

(n = 1)

Police officers, health professionals working for an agency

Full behavioral-health assessment and elaboration of a personal intervention plan. Services are offered to prevent certain problems associated with criminal activities: mental health problems, substance abuse, and general life stressors. Participants must complete 25 hours of community service and pay $350 fees (services are offered even if the person cannot pay). Charge is lifted (including the arrest record) after successful completion of the program.

  1. 18 and older

  2. Cooperates with the police officer

  3. Minor offense

  4. No criminal history

  5. Contacts the agency offering the program within 7 days.

Not specified – possession of under 20 g of marijuana, possession of drug paraphernalia, underage possession of alcohol, non-domestic simple battery, and non-domestic simple assault

Eau Claire County Pre-Charge Diversion1

Pre-booking: de facto decriminalization

Eau Claire County, WI

(n = 1)

Police officers, program coordinator, district attorney’s office

Person meets with district attorney to establish eligibility. One-time education class covering topics such as legal and social ramifications of citations/criminal violations is offered to low-risk drug users instead of criminal processing. Charges are not prosecuted, and arrest record is not made publicly available if the person pays restitution and program fees upfront and remains offense-free pending completion of the class.

  1. First offense, or

  2. Low risk based on criminal history, and

  3. Pay restitution and program fees

Cannabis, disorderly conduct, drug/paraphanalia possession, theft

Misdemeanour Drug Diversion Program (MDDP)

Pre-booking: de facto decriminalization

Winnebago County, WI

(n = 1)

Police officers, district attorney’s office, health professionals, volunteers

Individuals referred by a law enforcement officer for a drug-related misdemeanor have the option of completing the MDDP. The 12–18-month program includes paying a $250 fee, frequent urine tests, intense supervision, finding a job, guidance, education, and drug tests. Participants had to remain crime free and abstinent for at least 6 consecutive months. Charges are dismissed after the successful completion of the program (i.e., participants respected all conditions).

  1. 17 years and older

  2. Pled no contest or guilty

  3. Referred by the police to the district attorney for a drug-related offense

All drugs – misdemeanor drug charge or misdemeanor charge where drugs are suspected to be an issue

System of Healthy Options for Release and Transition (SHORT)

Pre-booking: de facto decriminalization

Travis County, Austin, TX

(n = 1)

Pretrial officers and drug court intake officers, members of the public, assistant district attorney and court staff

A committee evaluates the person soon after arrest for eligibility. Assessment of addiction is conducted. The 12-month program includes random drug tests (sobriety is required), case-management meeting, education classes, treatment referrals, graduated sanctions as necessary (for rule violations), payment of fees, and court appearance. Charges are withdrawn following completion of the program.

  1. Committed a non-violent offense

  2. Arrested for possessing a small quantity of drug

  3. Drug addiction

  4. Agrees to be evaluated

  5. Signs a contingency contract agreement with the court

All drugs – non-violent offenses committed by people showing signs of addiction

Cannabis Cautioning Scheme

Pre-booking: de jure decriminalization


(n = 3)

Police officers, public agencies offering different services

Following a caution (official warning) for drug use or possession, people who use drugs are referred to education and/or treatment rather than arrested. No criminal record is recorded if the participant complies with the caution.

  1. Possesses a small quantity of cannabis for personal use

  2. No other offense

  3. No history of drug-related offenses (not in all territories)

  4. Admit the offense and accept the caution

  5. Goes through assessment (not in all territories)

Cannabis – drug use or possession

Queensland Police Diversion Program

Pre-booking: de facto decriminalization

Queensland, Australia

(n = 1)

Police officers, accredited service providers (education and assessment)

First time drug offenders with no history of violence are offered an alternative following an arrest for a minor drug offense related to cannabis. Police officers have the obligation to offer diversion if the person meets the inclusion criteria. Offenders who accept diversion must attend a 1-2 hour education and assessment session. Charges may be laid if the person does not attend the session. Referrals to other services are made if necessary.

  1. Admit to committing a minor drug offense

  2. No prior experience with diversion

  3. No history of violence

  4. Not previously been offered diversion

Cannabis – drug use or possession

Drug Diversion Pilot Program

Victoria, Australia

Pre-booking: de facto decriminalization

(n = 1)

Police officers, Department of Human Services, and government funded drug treatment services

Upon arrest, eligible PWUD who accept the caution (official warning) are referred to a direct line to schedule an assessment within 5 working days. Following treatment, the caution is expiated.

  1. Adult or juvenile offenders

  2. Admit the offense

  3. First drug-related offense

  4. Voluntary participation in the program

All drugs other than cannabis – drug-related offenses (specific offenses not specified)

  1. Some information was collected on the following websites to complement information found in the publication (Callister, 2016). (last consulted on 19 September 2021). (last consulted on 19 September 2021).

Table 2 is based on 26 studies since one qualitative evaluation (Barberi &Taxman, 2019) was conducted among police chiefs who offered any type of diversion program for people who use drugs. The name of the programs was not provided.

Table 3. Distribution of outcomes and scores of diversion programs on the Scientific Method Scale (SMS)

Types of outcomes

# of evaluations

Study scores on the SMS

Distribution of outcomes






# of times the outcome was measured

Favorable N (%)

Neutral N (%)

Unfavorable N (%)

All outcomes








122 (76.7)

9 (5.7)

28 (17.6)









12 (92.3)

0 (0.0)

1 (7.7)

Drug use








11 (64.7)

3 (17.6)

3 (17.6)









10 (100.0)

0 (0.0)

0 (0.0)

Criminal recidivism








67 (77.0)

5 (5.7)

15 (17.2)

Drug availability








6 (42.2)

0 (0.0)

7 (53.8)

Case management or processing, social and health-related costs








14 (82.4)

1 (5.9)

2 (11.8)

Table 4. Themes founds in qualitative evaluations

Denman, 2018

Schiff et al., 2017

Clifasefi & Collins, 2016

Worden & McLean, 2018

McLeod & Stewart, 1999

Reichert et al., 2017

Magana, 2019

Hughes & Stevens, 2010

Baker & Goh, 2004

Barberi & Taxman, 2019

Beckett, 2014

New Mexico Sentencing, 2018

Hales et al., 2004

Themes related to the perceived effectiveness of the measure

Perceived effectiveness by participants




Perceived effectiveness by police officers and chiefs





Perceived effectiveness by community partners and other care providers






Perceived backlash effects




Themes related to facilitators and barriers

Program acceptance by the police






Full and continuous training






Effective communication and intersectoral collaboration






Clear eligibility criteria






Simple admission process with coordinated activities





Informing potential partners and beneficiaries










Developing quality relationships with partners and program users






Timely access and service availability







Figure 1. Flow chart of study selection process

No comments here
Why not start the discussion?