The Effectiveness of Substance Abuse Treatment with Young Offenders

3. Substance Abuse Treatment for Youthful Offenders: What Works!

3. Substance Abuse Treatment for Youthful Offenders: What Works!

Unfortunately, there has been a paucity of research examining the provision of substance abuse treatment services to young offenders (Dobkin, Chabot, Maliantovich, & Craig, 1998; Gilvarry, 2000; Pickrel & Henggeler, 1996). Consequently, it was necessary to review other sources to extract relevant treatment information that may be pertinent to adolescent offender populations. More specifically, articles were reviewed within the general offender treatment literature (including both youth and adult samples) as well as, to a lesser extent, the adolescent substance abuse treatment literature. The decision was made to reduce the amount of exposure to the adolescent substance abuse treatment literature given the potential incompatibility of both samples. Support for the latter decision is based on the work of Cottle et al. (2001) who argued that first-time offenders can be reasonably assumed to be distinct from repeat offenders. This line of reasoning can be extrapolated quite nicely to argue for the potential incompatibility of delinquent and non-delinquent populations as well.

Wilson, Gottfredson, and Najaka (2001) recently argued that meta-analysis should be the method of choice for evaluating program effectiveness because it tests quantifiable post-treatment differences between a treatment and comparison group extracted from a series of studies in the extant literature. Therefore, several offender and non-offender based meta-analyses will be reviewed in this section. Single studies also will be inserted where appropriate to augment the discussion and provide concrete examples of either the points of interest or to present avenues for future research.

Catalano, Hawkins, Wells, and Miller (1990) suggested that researchers should examine pre-treatment, in-treatment, and post-treatment variables when exploring the effectiveness of substance abuse treatment. This is an elegant way of representing the multiple influences that may be operating on a correctional program at any given time and so this format has been adopted for this report.

3.1 Pre-treatment variables

Catalano et al. (1990) identified several characteristics of program participants that may play an important role in the determination of client outcome. Some examples of the more commonly discussed aspects in the research literature include age, gender, race/ethnicity, age of onset, history of substance use, and psychopathology.

3.1.1 Age

The findings related to age of participant are quite contradictory. For example, certain studies have reported that younger participants do better, while others have documented that older individuals experience enhanced programmatic effects (Feigelman, 1987; Hubbard, Cavanaugh, Craddock, & Rachal, 1985). Still others have concluded that age does not have any impact on treatment outcome (Sells & Simpson, 1979). It should be noted that these studies were conducted on non-offender populations and it may be valuable to examine what relationships, if any, occur between age of participant and programmatic effects in young offender research. The findings of two meta-analyses that have been conducted on the broader correctional treatment literature may provide some preliminary insight into this question.

Latimer (2001) conducted a meta-analytic review of the family intervention literature for young offenders and reported that programs that engaged participants who were younger than 15 years of age yielded significantly higher mean reductions in re-offending than programs that engaged older clients. More related to the present review, Dowden (1998) conducted a meta-analysis of the entire correctional treatment literature for both adult and juvenile offenders. In a preliminary analysis of programs that addressed substance abuse as at least one of its program targets, the findings were quite similar to those reported by Latimer, in that age had an inverse correlation with program success. These findings complement quite nicely the fervent arguments made by other researchers regarding the importance of early identification and treatment of substance abuse problems in adolescent populations (Webster-Stratton & Taylor, 1998).

Thus, the preliminary correctional-based evidence that is available suggests that, unlike the broader adolescent substance abuse treatment literature, age of client may have an important impact on treatment success.

3.1.2 Gender

It may be assumed, based on past evidence, that since substance abuse and juvenile delinquency co-occur in both male and female young offenders, targeting this criminogenic need for intervention would yield positive programmatic effects for both genders. Once again, the meta-analytic database of Dowden (1998) was consulted to explore this question. Although there were too few cases involved in substance abuse treatment to examine the differential effectiveness of treatment for both male and female adolescent offenders, when both adult and juvenile offenders in substance abuse treatment were examined, the same positive programmatic impacts were found for both male and female offenders.

It should be noted that despite the positive findings for both genders, this analysis does not address issues related to program delivery. More specifically, it does not indicate whether delivering the program in a particular way or focusing on specific material may make it more relevant or therapeutically meaningful for one gender over another. For example, several advocates for gender-specific treatment have argued for the importance of attending to relationship-oriented issues within treatment for female offenders (Bloom, 1999; Covington, 1998). Unfortunately, this recommendation has not yet been tested. Therefore, future evaluations should explore gender-specific responses to treatment and identify positive program delivery aspects.

3.1.3 Race/Ethnicity

Unfortunately, although issues relating to gender may be seriously underrepresented in correctional treatment literature, this state of affairs is similar, if not worsened, when issues surrounding race/ethnicity are explored. Several researchers have lamented the lack of attention given to race/ethnicity issues within substance abuse treatment programs, despite their clear importance to delivering effective and appropriate treatment to this population (McNeece, Springer, & Arnold, 2001; Rhodes & Jason, 1990).

Within Canada in particular, Aboriginal offenders represent an increasingly important correctional population since they make up only 3% of the general population, but account for 17% of Canadian federal inmates (Andrews, Dowden, & Rettinger, 2001; Motiuk & Nafekh, 2000; Nafekh, 2002). This trend is expected to continue unless significant resources are committed to this problem (Boe, 2000). Unfortunately, none of the studies reviewed for this report focused on youthful Aboriginal offenders so studies focusing on adult Aboriginal offender populations were reviewed to identify potentially relevant treatment issues.

Motiuk and Nafekh (2000) reported that substance abuse was one of the core criminogenic needs for both male and female Aboriginal offenders in the Canadian criminal justice system, a finding echoed in a more recent study by Nafekh (2002). This latter study provided overwhelming evidence for the critical role of substance abuse in this population as 95% of the Aboriginals in this study exhibited problems in this area. Given the widespread substance abuse problems encountered within this population, treatment protocols for Aboriginal offenders should address this treatment need.

Finally, the design and implementation of correctional programming for Aboriginals must ensure that it is both culturally sensitive and relevant to enhance its clinical utility. Researchers have argued that key Aboriginal beliefs and cultural practices such as healing, reconciliation, spirituality, respect, accountability, balance, and restoration must be incorporated into the treatment program in order for it to be effective (Wilson, 2000).

The literature, although sparse, does provide some preliminary evidence for the provision of culturally specific treatment services to Aboriginal offenders. For example, a recent follow-up study conducted by Correctional Services Canada of federal Aboriginal offenders admitted to Healing Lodges found that almost 70% of the clients completed the programs. In addition, and arguably more important, only 6% of program clients returned to custody, a statistic which is even more impressive when one considers that the national average is 12% (Wilson, 2000). Further evidence regarding the importance of culturally-specific treatment is obtained from the work of Waldram (1994) who found promising evidence for Aboriginal spirituality using a case study approach. Finally, research conducted in New Zealand has found lower rates of re-offending among Maori offenders who reported cultural pride and cultural knowledge (Maxwell, 1999, cf. Andrews, Dowden, & Rettinger, 2001).

These findings have strong applicability to Aboriginal young offender populations and have provided several important areas of scientific inquiry. However, much more research needs to be conducted as a significant knowledge gap currently exists.

3.1.4 Age of onset

Past research has found that the age of onset of substance use is not only related to involvement in delinquent activities but also to success in treatment. More specifically, several studies have reported that the earlier the age of substance abuse onset reported by adolescents, the poorer the treatment outcome (De Angelis, Koon, & Golstein, 1990; Feigelman, 1987; Friedman, Glickman, & Morrissey, 1986). Unfortunately, these studies did not examine young offender populations and so caution should be exercised when considering the external generalizability of the results.

3.1.5 Psychopathology

Psychopathology has been noted as one of the most persistent indicators of negative program outcome and thus several researchers have stressed the need to address this issue within treatment. As discussed by Randall, Henggeler, Pickrel, & Brondino (1999), symptoms can be classified as either externalizing or internalizing, with the former being strongly related to unsuccessful program completion in both correctional (Hiller, Knight, & Simpson, 1999a) as well as non-correctional samples (Dobkin et al., 1998; Kaminer, Tarter, Bukstein, & Kabene, 1992).

A recent study conducted in Quebec by Dobkin et al. (1998) explored the therapeutic impact of a multistage 12 month substance abuse treatment program that was divided into two months of inpatient treatment, three months of outpatient treatment, and seven months of aftercare. The goal of the study was to explore any client factors that may have been related to program outcome by looking at three separate groups of program clients: a) treatment completers who improved on program measures; b) treatment completers who did not improve; and c) program dropouts. Interestingly, there were several significant between-group differences on the various assessment tools employed within the program. More specifically, while the non-completers exhibited significantly higher social maladjustment, aggression, and school-related problems, program completers who improved in the program had significantly higher scores on repression and denial of emotions. Finally, program completers who did not improve evidenced significantly higher levels of depression and social anxiety.

Based on these findings, the authors asserted that there were two separate underlying motivations for adolescent substance abusers with externalizing disorders and internalizing disorders. More specifically, the externalizing group may be abusing substances to "self-stimulate", which makes them much less likely to complete treatment. The internalizing group, on the other hand, engages in substance abusing behaviour to self-medicate (e.g., to alleviate their level of emotional unrest) and, as such, may be more likely to complete treatment.

The clear message from the above discussion is that in order for substance abuse treatment programs to be more effective, more attention must be paid to the psychopathology of adolescent substance abusers (Catalano et al., 1990; Dobkin et al., 1998; Kaminer et al., 1992). In particular, those with externalizing problems must be appropriately identified to ensure that their motivation for program participation is addressed (Catalano et al., 1990; Gilvarry, 2000). Several recommendations in this regard have been forwarded, such as ensuring the treatment plan is maximally personalized so that the youth can see how the program will meet their goals and also by helping the adolescent overcome social and personal deficits and/or problems which may be hampering their involvement in treatment (Catalano et al., 1990; Pickrel & Henggeler, 1996). Identifying and distinguishing substance abusers who exhibit different types of psychopathology at intake is also important as each of these populations will have unique treatment needs (Dobkin et al., 1998; Kaminer et al., 1992; Randall et al., 1999). For example, Randall et al. (1999) documented that substance abusers with externalizing disorders also have poorer familial environments, thus, engaging the parents and other family members in treatment may have a positive impact on the youth. In particular, emphasis could be placed on assisting the parents in more effectively structuring, monitoring and supervising the behaviour of the adolescent. For those with internalizing disorders (such as anxiety or depression), programs could be tailored so that the youth are taught coping skills to help deal with these emotional difficulties (Randall et al., 1999). Although these preliminary suggestions admittedly need to be validated through additional empirical study, the value of these types of recommendations is clearly evident given the above reviewed findings on the negative impacts of psychopathology on program success.

3.1.6 Risk

Although this variable has not received any attention within the general adolescent substance abuse treatment literature, the risk principle of case classification has been lauded as an integral component in the delivery of effective correctional treatment for offenders (Andrews et al., 1990; Andrews & Bonta, 1998). The risk principle states that the amount of intervention an offender receives must be appropriately matched to his/her risk level to re-offend. More specifically, higher-risk offenders should receive more intensive and extensive services whereas lower-risk clients should receive minimal or no intervention.

Meta-analytic reviews of the young offender correctional treatment literature have provided strong empirical evidence for this principle as these reviews have demonstrated that programs that target high-risk cases exhibit more positive programmatic effects than those that target low-risk cases (Dowden & Andrews, 1999a; Lipsey, 1995; Lipsey & Wilson, 1998). The clinical utility of this principle has been also demonstrated in female (Dowden & Andrews, 1999b) and minority offenders (Andrews et al., 2001). Therefore, administrators of substance abuse interventions should ensure that appropriate risk assessments are conducted on program clients and that the high-risk cases are prioritized for treatment involvement and receive more intensive treatment exposure.

3.1.7 Additional client characteristics

In a review of the substance abuse treatment literature, Annis (1990) described several client characteristics that were related to positive treatment outcomes in substance abuse interventions. These factors included "being married, employed, of a high social class, financially secure, socially active and well adjusted to work and marriage, and having little history of arrest." Unfortunately, these factors are not commonly found in correctional populations (Annis, 1990), and are even less likely to be found in young offenders. Those client factors most strongly related to negative program outcome, on the other hand, were quite prevalent in offenders including aggressiveness, high rates of attempted suicide, and various problematic psychiatric sympotomatology (Annis, 1990).

Date modified: