Abstract
Linked article:
2. Background
For the last decades, drug addiction has become an increasingly worrying problem throughout the Western World. Drug-addicts have been disproportionately involved in criminal activities, making drug-addiction, beyond public health concerns, a formidable challenge to public order.
In Switzerland, for example, burglaries and robberies increased by several hundred percent during that period (Killias, 2001). International comparisons (Killias & Ribeaud, 1999) suggest that the extent of involvement in property crime among addicts of any kind of hard drugs is about 10 times higher than among non-users. Thus, the increasing crime trends between 1970 and 1995 may reasonably be seen as a side-effect of increasing drug use. In response to this phenomenon, numerous programs have been set up to provide drug addicts with narcotics (e.g., heroin prescription programs) and substitution drugs. The intended goals of such treatments have been:
(1) to improve drug users’ quality of life, reducing the risks of overdose or contagious diseases, controlling the quality of drugs available on local markets, preventing marginalization and improving social integration, (2) to diminish social costs of drug addiction, (3) to reduce drug-related offences and protect public order. It is assumed that drug addicts commit many predatory offences mainly to finance the purchase of drugs, and that criminality will decrease once drugs are supplied to addicts through official channels. The same effect should be observed if drug addicts are supplied with products (such as methadone) that suppress physical effects of withdrawal and, indirectly, the immediate need to consume drugs, (4) to reduce public order problems of all sorts. If addicts obtain drugs through official channels, they should spend less time in the search for drugs, which means that they have more time available for legitimate earnings and will less concentrate in places where addicts and dealers regularly gather (e.g., “needle” parks in Switzerland).
Many researches have studied the effects of drugs prescription programs on criminal behavior among participants. We shall review these programs and try to find out whether they have been effective in reducing criminality.
3. Objectives of the review
The review will aim at evaluating the effects of drug prescription and substitution programs on criminal behaviour among participants. To be included in this review, studies have to assess the effects of drug substitution on re-offending. If this review reveals significant effects of such programs on criminality, the results could have implications for crime and justice as well as for drug policies. For example, if the results of our meta-analysis support the conclusion that treatment using heroin reduces criminality, medical prescription of heroin should be recognized as an option in the treatment of severely addicted drug users with high criminal involvement.
4. Methods
4.1- Criteria for inclusion and exclusion of studies in the review
Randomized studies, quasi-experimental studies and before-after comparisons on the effects on re-offending of drug substitution programs will be included. Interventions can be court-ordered or unrelated to any involvement of the criminal justice system. Only interventions based on substitution programs (using e.g. methadone and/or opiates as substitution drugs) will be considered. Possible effects beyond re-offending will not be considered, in particular eventual medical outcomes. Studies published in any language after 1960 will be considered. A coding protocol has been prepared along the guidelines of the Campbell Collaboration (see attachment). Moreover, our review will comply with the current standards of meta-analysis (e.g. as specified in Practical Meta-Analysis by Lipsey & Wilson). Details relating to the techniques used will be determined once the exact number of studies meeting the eligibility criteria is known.
Types of studies
Studies meeting level 2 or higher on the scale developed by Sherman et al. (1997) will be considered.
This includes:
One-group, pre/post studies: studies comparing individual delinquency rates before, during and following treatment. To be eligible to our analyses, studies must include the prescription of a drug (e.g., methadone, heroin). Multi-group comparison group studies, including both true experimental studies (randomized designs) and quasi-experimental designs: studies comparing delinquency rates among subjects of an experimental group before, during and following treatment to those of a control group (with or without random assignment). As in the previous paragraph, studies are eligible only if the treatment group undergoes substitution therapy. As a control group, we will consider any group undergoing an alternative treatment or no treatment at all (including placebo control). For example, if the treatment group is being treated with heroin as a substitution drug, the control group may remain untreated or receive any other substance as a substitution drug (e.g., methadone), or undergo abstinence therapy with or without psychotherapy, detoxification, etc. Macro level studies: studies assessing the impact of drug substitution at the macro (i.e. city, regional) level. In order to be eligible, such studies would need to measure the impact of the program on delinquency at the city/regional level, using police, court or survey data.
The three different types of studies will be analyzed separately.
All of the studies taken into account must assess the effects of drug substitution programs on re-offending.
The criteria of inclusion are deliberately broad because we are concerned about the number of studies that may be available in this field. If a reasonable number of studies (more than twenty) can be identified and located we shall consider restricting the review to studies meeting higher methodological standards. At the documentary stage, however, it seems safer to include as many studies as possible.
Types of programs
To be included in this review, studies must report effects of drug prescription and drug substitution programs on criminality among drug users. By drug substitution program we understand a programs that includes the prescription of substances rather than a program based on drug abstinence. The prescription must imply substances considered as substitutes for illegal drugs, for example, methadone or buprenorphine as a substitute of heroin. This excludes the prescription of drugs such as tranquillizers or antibiotics, frequently prescribed to drug users. On the other hand, we have also considered the medical assisted prescription of heroin. This does not mean that heroin is a substitute of heroin, but that the uncontrolled consumption of heroin in the streets is replaced by the prescription of a controlled dosage of heroin, adapted to the user's needs.
Programs that do not include prescription of any substance will not be considered, such as programs based exclusively on, for example, psychotherapy, detoxification, etc. Only interventions based on substitution therapy (using e.g. methadone and/or opiates as substitution drugs) will be considered.
Types of outcome measures
The key variable will be re-offending as measured by reconviction data, police records and studies on self-reported delinquency. Drug possession and consumption, although an offence in most countries will not be considered as a measure of re-offending. To the extent studies address the effects at the macro level, any conventional outcome measures (statistics, crime victims surveys etc.) will be considered.
To assess improvement at the individual level, we shall look at prevalence rates (or percentage of people who re-offend) as well as incidence rates (or number of offences committed per person) during standardized pre- and post-intervention periods. Prevalence rates inform on how many persons are diverted from criminal activity by prescription of substitution products, whereas incidence rates allow assessing whether less offences are committed as a result of the program. It is important to make this distinction since a given program may reduce the number of offences without affecting the number of offenders. Possible treatment effects beyond re-offending, such as medical outcomes or effects of such programs on drug markets, will not be considered.
Types of participants
Population: Drug-addicts (e.g. heroin addicts, cocaine addicts), adults and adolescents, males and females.
4.2- Search Strategy for identification of relevant studies
Relevant studies will be identified through abstracts, bibliographies and databases such as Campbell Crime and Justice Group (C2-SPECTR), National Criminal Justice Reference Service (NCJRS), Harms Reduction Journal, Journal of Substance Abuse Treatment, National Treatment Agency for Substance Misuse (NHS), National Treatment Outcome Research Study (NTORS), Drug and Alcohol Dependence, Drug and Alcohol Review, Drug and Therapeutics Bulletin from the BMJ group (DTB), International Journal of Drug Policy, Central Committee on the Treatment of Heroin Addicts (CCBA), Journal of Clinical Psychopharmacology, Criminal Justice Abstracts (CJA), Déviance et Société, JSTOR, Criminal Justice and Behavior (CJB), Criminologie, the Germany literature (Heroinstudie.de-www.heroinstudie.de/H-ReportP2engl.pdf) and the www.drugscope.org.uk.
The keywords that will guide the search for reference databases and bibliographies are the following: drug addiction; drug prescription; substitution programs; controlled trial; re-offending; heroin; methadone; opiates, treatment programs; drug abuse; drug addict, heroin prescription, property crime, cocaine abuse; dexamphetamine; cocaine substitution. In addition, the following combinations of keywords will be used: substitution program + re-offending; heroin + treatment programs; heroin + substitution program; heroin + methadone; opiates + treatment; opiates + substitution; heroin + property crime; substitution programs + property crime, cocaine abuse + dexamphetamine.
Further studies will be located through contacts with experts in countries where relevant studies might be located. A few years ago, the reviewers have been involved in the evaluation of the Swiss heroin trials and have, through that experience, become familiar with several international centres in charge of evaluations in this field.
Since the review will be restricted to studies on the effects of drug prescription and substitution on re-offending (excluding use of illegal drugs as a measure of re-offending), studies using urine or blood testing will not be considered.
The staff of the Institute for Criminology and Criminal Law being multi-lingual, studies published in any of the following languages can be included: English, French, German, Dutch, Italian, Spanish, Portuguese, Romanian, Polish, Ukrainian and Russian. Through international channels, such as the European Sourcebook Group (with its network of correspondents in more than 40 countries), the European Society of Criminology and the International Society of Criminology, contacts will be established with countries not routinely covered in international reviews of research.
Studies conducted or published after 1960 will be considered.
4.3- Description of methods used in the component studies
The methods used by the studies covered by this review can be the following ones: Randomized studies, quasi-experimental studies and before-after comparisons.
All studies included in this review will have a measure of the effects of drug substitution treatments on re-offending such as arrest, conviction, incarceration or self-reports.
The characteristics of a few already located studies (see par. 10, pp. 11-13) may serve to illustrate the criteria of eligibility.
4.4- Criteria for determination of independent findings
There are three potential sources of non-independence of findings. We shall use the same criteria as in similar Campbell Review Protocols (e.g. Lipsey & Landenberger (2006); Wilson, Mitchel & Mackenzie (2007).
The first potential source of nonindependence of findings is multiple indicators of re-offending reported from a single study (e.g. arrest, conviction). When more than one such outcome is reported, only one will be selected for the analysis. To maintain as much comparability as possible across studies, coders will select the outcome measure that is most frequently represented in other studies in the collection (Lipsey & Landenberger, 2006). The second occurs when the same outcome is measured at multiple points in time, e.g., 6-months, 12-months, 18-months and two years post-treatment. In those cases, the measure with the timing closest to that most commonly used across all the studies will be chosen to maximize comparability between studies (Lipsey & Landenberger, 2006).
Finally, the third source of nonindependent findings is the same data being reported across multiple documents. We will use author's names, court location and study time frames to identify multiple publications of the same evaluation. When such multiple publications are identified, the most complete and detailed manuscript will be designated as the primary coding source. Additional manuscript will be consulted to flush-out coding if necessary (Wilson, Mitchel & Mackenzie, 2007).
4.5- Details of study coding categories
A coding protocol has been developed for this synthesis that provides for a systematic method of extracting information regarding each study's research design, program, nature of the outcome measure, and outcome data (see appendix-Coding Protocol).
4.6- Statistical procedures and conventions
Our review will comply with the current standards of meta-analysis (e.g. as specified in Practical Meta-Analysis by Lipsey & Wilson).
The three types of studies mentioned above (pre/post studies, randomized controlled trials and quasiexperimental designs, and macro-level studies) will be synthesized and meta-analyzed separately.
The effects of drug prescription or substitution programs on the criminal behavior will be encoded using the odds-ratio. The odds-ratio is well suited to dichotomous outcomes, such as those commonly used in drug treatment. When the measure of re-offending is measured continuously, we will compute a standardized mean difference type effect size and transform it into an odds-ratio (see Lipsey & Wilson, 2001).
Effect size outliers (›± 3.0 standard deviations) will be winsorized to less extreme values (next highest not judged an outlier). Small proportions of missing data on variables other than effect sizes will be imputed based on the mean for the most similar studies. When larger amounts are missing, the variable will not be used in the analysis. In all cases, attempts will first be made to contact the original researchers to determine if they can supply the missing information.
4.7- Treatment of qualitative research
At this time we have no plans to include qualitative research in this systematic review.
5. Timeframe
The review will be completed within one year after approval of this protocol, in line with the guidelines of the funding agency (Federal Office of Public Health).
6. Plans for Updating the Review
This review will be updated every five years to include new treatment studies published in any language. The primary authors will take the lead in this update.
Footnotes
7. Acknowledgments
We would like to thank the Swiss Federal Office of Public Health (SFOP) for the support given to this project.
8. Statement Concerning Conflicts of Interest
The reviewers have no personal, material, or academic interest whatsoever in the outcome of the review. Financial support by the Swiss Federal Office of Public Health does not affect the independence of the reviewers.