PEER RISK AND PROTECTIVE FACTORS IN ADOLESCENCE: IMPLICATIONS FOR DRUG USE PREVENTION

Veonna Goliath, Blanche Pretorius Evidence-based practice guidelines for drug use prevention with adolescents propose focusing on increasing protective factors and reducing risk factors associated with drug use. The present article reports on a qualitative study undertaken with 10 adolescent drug users and 29 non-users from a historically marginalised community in Port Elizabeth, South Africa. The findings reveal peer risk and protective factors associated with drug use, and offer recommendations for how protective factors can be increased and peer risk factors mobilised as protective mechanisms and sources for promoting resilience in drug use prevention among adolescents.


INTRODUCTION
There are grave concerns, at both national and international levels, about adolescent drug abuse and its related effects, which can continue to impact on functional domains into adulthood.For instance, the impact of drug use on relationships, marital and employment stability, physical and mental health, morbidity and mortality have been well documented in both local and international research (National Institute on Drug Abuse, 2003;World Health Organisation, 2011) and in the national literature (Parry, Plüddemann & Bhana, 2009).
Despite the acknowledged impact of drug abuse, there has been a reported decline in adolescent drug abuse in the United States of America (USA).For instance, the Healthy People Report (United States, 2010) cites statistics on high school learners who never consumed alcohol as having increased from 19% to 28% between 1998 and 2009.Similarly, the percentage of high school learners who resisted the onset of illicit drug use increased from 46% to 53%.These statistics provide a promising picture and may suggest that drug prevention efforts, particularly in the USA, are yielding positive results.South Africa, on the other hand, has experienced a notable increase in drug use since the country's transition from apartheid to democracy (Harker, Myers & Parry, 2008;Peltzer, Ramlagan, Johnson & Phaswana-Mafuya, 2010).This scourge, especially the earlier onset of drug use, has serious implications for adolescents and the country as a whole (South Africa, 2008).
It is indisputable that prevention is more affordable than treatment, and that it also has the potential to prevent a myriad of drug-related problems.This is evident from reflecting on the drug use statistics cited in the second South African Youth Risk Behaviour Survey [SAYRBS] (Reddy, James, Sewpaul, Koopman, Funani, Sifunda, Josie, Masuka, Kambaran & Omardien, 2010).Of the 10 270 adolescent learners who participated in this survey, 29% reported life-time prevalence rates of cigarette smoking, 12.7% for dagga use, 49.6% for alcohol use, 6.2% for heroin use and 6.6% for methamphetamine use.The biannual surveillance reports by the South African Community Epidemiology Network on Drug Use (SACENDU) reflect equally high rates of drug use amongst youths under the age of 20, "with 20% in the Eastern Cape and 28% in KwaZulu-Natal receiving treatment" (Dada, Plüddeman, Parry, Bhana, Vawda & Fourie, 2012:2).These alarming trends underscore the importance of prevention efforts that would reduce both commencement of drug abuse in new users and harm on existing users.These findings, and the high relapse rate of adolescents in treatment, raise questions about the effectiveness of drug prevention services in South Africa (Burnhams, Myers & Parry, 2009), especially for people from historically marginalised low-income communities.One such geographical community is the Northern Areas in Port Elizabeth, which is where the present study was located.
Drug use prevention strategies (dating back to the early 1930s in the USA) have evolved from strongly regulatory approaches involving the banning of drugs, an increase in taxes (McNeece & DiNitto, 2013) to the scare tactics employed during drug awareness campaigns in the 1960s.They have also included affective education and behavioural interventions (involving life skills training), community involvement and eventually harm reduction approaches (Van Wormer & Davis, 2008).Harker et al. (2008:3) state that prevention "in its narrowest sense, targets individuals and their peers, and at the broadest level it takes the form of international treaties, conventions and other structural interventions".
Evidence-based practice guidelines for drug use prevention propose increasing the protective factors and decreasing the risk factors for adolescent drug use (National Institute on Drug Abuse, 2003;South Africa, 2012).Whilst multiple common risk and protective factors are associated with drug use and non-use among adolescents, operating not only in mutually exclusive and reciprocal but also autonomous ways, there are also unique factors related to social, economic and cultural contexts (Myers, Harker, Fakier, Kader & Mazok, 2008).Several studies illustrated the value in targeting more proximal as opposed to distal risk factors to bring about change in adolescent drug use (Brook, Morojele, Pahl & Brook, 2006;Potgieter, Goliath & Pretorius, 2010); hence the focus in the present research on the adolescent peer group as the dominant source of socialisation in adolescence.Unfortunately, adolescence often has negative connotations as a time of turbulence and emotional and identity confusion, where parental authority and guidance are challenged, and the influence of adolescent peers is paramount (Kim, Zane & Hong, 2002).Supporting the opposing view of Kerr, Stattin and Burk (2010), who maintain that competence in adolescence is characterised by a decrease in parental monitoring and an increase in adolescents' autonomy, the authors of this article argue that the narrow construction of adolescence as a risk group restricts social service practitioners and adolescents alike from acknowledging the value of the adolescent peer group and its potential for increasing peer protective factors against adolescent drug use.The primary objective of the present article is thus to illustrate the peer risk and protective factors associated with drug use and its implications for drug use prevention.

LITERATURE REVIEW
In order to contextualise the theoretical underpinning and empirical findings of the study, the ensuing literature review examines key aspects in adolescence related to peer risk and protective factors, and to drug use prevention.
Adolescence connotes the complex transition between the states of childhood dependence and adult independence (Wood & Hine, 2009).Unfortunately, the negative discourse around adolescence as a vulnerable developmental period, characterised by turbulence and emotional and identity confusion, has received prominence in the literature (Gosin, Marsiglia, & Hecht, 2003;Louw & Louw, 2007).Characteristically adolescents spend more time with peers, who become their primary source of socialisation especially with regard to the learning of social norms (Kliewer & Murrelle, 2007;Kim et al., 2002).Peer influence frequently results in the challenging of parental authority and guidance (Bezuidenhout & Joubert, 2003).Compounding the tension between adolescent and parent, according to Bester (2011), is adolescents' overestimation of their abilities and responsibilities, and underestimation of their susceptibility to risk.It is these constructions that give rise to adolescents frequently being described as an "endangered and dangerous group at risk from others, to themselves, and to the fabric of communities" (Kim et al., 2002:566).
In stark contrast to these views, constructionist theorists have called for a contextual appreciation of adolescence.Fatusi and Hindin (2010:1), for instance, point out that today's generation of young people has to navigate the transition to adulthood in a world that is "vastly different from previous generationsa world where AIDS, globalisation, increasing urbanization, electronic communication, migration, economic challenges, among other external forces, have radically transformed what it means to be young".Resonating with the sample selected for the present study, Sathiparsad's (2008) reminder that the effects of apartheid and the political transition in South Africa have resulted in adolescents being exposed to vastly different life circumstances and experiences is important.
Adopting a strengths-based perspective on adolescence, several post-modernist thinkers acknowledge the opportunity for growth and maturation during the adolescent life stage (Jones, 2009).Ungar (2006) described this process as adolescents' search for health, implying that they need to generate substitutes for harm-producing behaviour rather than focusing on suppressing it.Accordingly, Kerr et al. (2010) suggest that an increase in autonomy and competence in adolescence is directly related to a decrease in parental monitoring.Pilkington (2007) maintains that the adolescent peer group becomes the context in which their autonomy and competence are shaped.She argues that the adolescent peer group is a source of reciprocal positive influence for each other in which adolescents can learn to assert their agency, thus enabling them to experience a sense of power, recreation, acceptance, protection (Pilkington, 2007) and a sense of purpose in their community.Furthermore, Karcher, Brown and Elliott (2004) propose that constructive peer influence is subject to two reciprocal processes, i.e. the formation and preservation of good relationships with prosocial peers, as well as the mastery of effective relational skills.
The literature on peer influence suggests that it can be direct and indirect, and occurs through four mechanisms: i) informed by group norms; ii) direct peer pressure; iii) peer influence through modelling; and iv) creating structured opportunities (Karcher et al., 2004;McWhirter, McWhirter, McWhirter & McWhirter, 2013).These authors argue that each of these mechanisms should be considered when designing effective interventions for adolescents.Modes of peer influence may operate simultaneously or independently, but they occur mostly in peer clusters (which is a designated section of the peer group that has become the primary source of influence on the values, attitudes and beliefs of its adolescent members) (McWhirter et al., 2013).Furthermore, the same complex set of dynamic influences is apparent in formal groups and structured activities; therefore, effective interventions must be designed with an awareness of all four modes of influence, taking into consideration both direct and indirect peer influence.Other post-modernist thinkers suggest that adolescents purposefully and voluntarily associate with the peer group whose group norms they aspire towards and whose behaviour they wish to emulate (Brook et al., 2006;Pilkington, 2007), suggesting that practitioners need to explore the value that adolescents derive from particular peer associations.
A literature search on the risk and protective factors for adolescent drug use confirm that risk factors have been defined as those factors that enhance the likelihood that a person will engage in drug abuse or become dependent.These risk factors are associated with harmful or otherwise negative outcomes for the person (McWhirter et al., 2013).Protective factors, on the other hand, refer to factors associated with reduced potential for drug abuse, or variables that mitigate against or buffer the effects of risk factors.These factors may be autonomous with no corresponding risk factor, or may be the direct opposite of a risk factor.Furthermore, their presence can enhance, interact with or moderate other protective factors, leading to greater or lesser drug use (Kim et al., 2002;Liddle & Rowe, 2006).The relationship between the number and type of risk and protective factors can therefore determine the likelihood of an adolescent's vulnerability to drug use.Both risk and protective factors can be categorised in five domains or settings, namely individual, family, peer, school or community domains (McWhirter et al., 2013;National Institute on Drug Abuse, 2003).As indicated earlier, the focus of this article is on the third domain, that of adolescent peer risk and protective factors.
A literature review identified the following risk factors for adolescent peer drug use: peers with a favourable attitude towards drug use; drug using peers; peers who are generally risk-prone; peers with an affiliation to a gang; peers who subject others to peer pressure (Brook et al., 2006;Karcher et al., 2004;Loxley, Toumbourou & Stockwell, 2003;McNeece & DiNitto, 2013); and alienation by prosocial peers (National Institute on Drug Abuse, 2003).Falkowski (2003) and National Institute on Drug Abuse (2003), however, argue that negative peer association, which is one of the most immediate risks to drug involvement and subsequent expanded antisocial actions, occurs mostly when supervising adult figures or nurturing parent-child relationships are absent.In contrast, Kerr et al. (2010) found that the monitoring of adolescents is enhanced by youth-rather than parent-driven initiatives.
Apart from peer monitoring, the peer protective factors that emanated from the literature review include prosocial peer association (Potgieter et al., 2010); avoidance of peers who are prone to drug use (Loxley et al., 2003); and friends who uphold non-drug use norms (Substance Abuse and Mental Health Services Administration, 2011).A study by Buckley, Sheehan and Shochetet (2010) revealed that adolescents offer protective behaviours to those peers with whom they share close friendships.Contrary to earlier findings that negative peer association enhances susceptibility to drug use, Smokowski, Reynolds and Bezruczko (2000) have found that resilient adolescents refuse to be enticed by the apparent exhilaration of the risk-taking behaviour of their peers, but instead learn from the consequences of this behaviour in others.Hill (2008), however, found that adolescents benefit from observational learning only if they consider themselves susceptible to the harmful consequences of drug use.Giving credence to this contention, Karcher et al. (2004:193) postulate that peer interaction has many "positive, growth-promoting qualities" which can be used to promote positive youth development.

THE RESEARCH METHODOLOGY
The primary goal of this article is to highlight adolescent peer risk and protective factors associated with drug use from the perspective of the adolescent research participants.A narrative tradition of inquiry research design, embedded in a qualitative research approach, was employed for the purposes of achieving this goal (Clandinin & Connolly, 2007).Consonant with this goal, it was essential to employ a narrative research design that allowed for a conversation where research participants could order their worlds through narratives and make connections and meaning by "linking past, present, self and society" (Riessman, 2008:114).The emancipatory potential of the research design furthermore allowed for a more complete story of the studied phenomenon, contextualised in the cultural and social context of the research participants.
A non-probability purposive sampling strategy was employed to recruit participants from the two sample groups (i.e.drug users and non-users), stipulating clear inclusion criteria.Generic inclusion criteria revolved around demographic variables (i.e.ages 16-18 years, with all participants residing in the specified geographical community).According to the categorisation by Louw and Louw (2007), the adolescent sample group was in the late adolescence stage.Furthermore, the drug users could be in any stage of the drug use cycle (i.e.use, misuse, abuse or dependence) (McNeece & DiNitto, 2013), and were recruited with the assistance of teachers, addiction counsellors and supportgroup facilitators.Three participants were in the misuse stage of drug use and seven in the dependence stage.
The non-users were recruited with the assistance of school teachers acting as gatekeepers.Adolescents who were self-proclaimed non-users of both licit and illicit drugs were invited to notify the teacher of their interest to participate in the study.The small sample size of 10 drug users and 29 non-users was determined on the basis of data saturation (Patton, 2002).The process of recruiting and retaining the research participants was a time-invested process, since it entailed obtaining parental consent and adolescent assent, building rapport with the prospective participants, engendering trust through demonstrating a genuine interest and reassuring them of confidentiality (De Laine, 2000).It was also crucial to explain the researchers' role and to refer prospective participants to social service practitioners when this was requested.
Data were triangulated in this qualitative study through two different methods of data generation with the two different sample groups (Denzin & Lincoln, 2003).Participants in the first sample of adolescent drug users were interviewed individually using a lifegrid interview (Wilson, Cunningham-Burley, Bancroft, Backett-Milburn & Masters, 2007) as an interview guide.The life-grid is a tool that prompts research participants to narrate their life experiences from birth to the present.Secondly, the adolescent nonusers' views were gleaned from individually written narratives in response to five semistructured questions.The stimulus questions for this group prompted them to share their views on the reasons for adolescent drug use and non-drug use, as well as their recommendations for drug prevention amongst adolescents.Particular attention was given to Riessman's (2008) cautionary note that the researcher would be listening to the participants' interpretation of their experiences rather than having direct entry to their experiences.Furthermore, participants' narratives represented interpretative repertoires, i.e. a coherent system of meanings that have developed over time and are used to evaluate actions or events from a cultural context rather than an individual perspective.In turn, the researchers' reflections, prompts and questions to their interpretations brought about a co-construction, implying that the researchers became an intricate, subjective part of the narration (Glover, 2004).
Narrative thematic analysis was employed for the purpose of analysing the varying meanings of constructions represented in the stories of the different sample groups (Denzin & Lincoln, 2003).The use of multiple-data generation methods in different contexts (i.e. the homes of drug users and the school for non-users), remaining sensitive to the specific socio-cultural context, and the researchers' protracted commitment to the field of study all served to enhance the trustworthiness of the study (Yardley, 2000).

DISCUSSION OF FINDINGS
The findings are presented in this section of the article and the terms "users" and "nonusers" are used to differentiate between the narratives from the two sample groups.The findings concur largely with existing literature on adolescent peer risk and protective factors associated with drug use (Myers et al., 2008;National Institute on Drug Abuse, 2003).However, the ensuing discussion highlights that the negative description of the adolescent peer group in the literature, which is largely produced from an adult perspective, is also prevalent in the adolescent participants' perspectives.These findings are juxtaposed against the participants' relative silence on and limited description of the adolescent peer group as a protective factor.Most (25) of the 29 non-users made reference to adolescent peer influences as risk factors for drug use, compared to six of the 29 non-users identifying peer influences as protective factors.These findings confirm the tendency in the literature to pathologise adolescent peer influence and further reinforce the categorisation of peer influence into mutually exclusive categories as either positive or negative (Hanson, Miller & Diamond, 2011).The non-users' suggestions to actively avoid negative peers, in order to reduce susceptibility to drug use, and to increase association with positive peers for the inverse effect, resonate with the tendency to categorise adolescent peers as either negative or positive (Kim et al., 2002).The relative silence in both sample groups on adolescent protective factors suggests a gap in the identification of peer factors and processes that could be of benefit in drug use prevention.The respective peer risk and protective factors identified by the research participants are discussed in the ensuing section.

Theme 1: Adolescent peer risk factors associated with drug use
There was agreement between users and non-users that negative peer association, the nature of peer influence, and the factors impelling submission to peer influence interact to compound adolescents' susceptibility to the onset of drug use.These three risk factors are discussed in an integrated manner given their interrelatedness.
The dialogue quotes are presented in the language of the research participants and have thus not been corrected grammatically.However, where participants used slang or colloquial terms, an English alternative has been offered in brackets.The dialogue quotes below illustrate the participants' views that association with drug using peers increases adolescents' exposure and susceptibility to peer pressure, and their subsequent alienation by prosocial peerssentiments that have been endorsed in previous research (e.g.National Institute on Drug Abuse, 2003).An example is Non-user 1's sentiment that: "Friends play a big role.The behaviour of ones' friends can influence you, mostly because we want to impress our friends.People who are friends with people who use drugs or alcohol will try to impress their friends by also using these drugs."On the other hand, User 9 stated: "Hmm, one of my friends just saidhere take a puff and so I told him -no, no my bru[brother] ….and so afterwards I took a puff, I blew out, so I did think, jooh, this is lekka [great] this!So I smoked on, so afterwards, I just smoke the Hooka-pipe." The narratives by both non-users and users illustrate that peers, being the primary socialising agents in adolescence, are influential mediators of the choices that adolescents make about risks.Sharland (2006) argues that adolescents' normalisation of risk behaviour (such as drug use) is motivated more by their need to blend in with their peers rather than by a desire to defy adult authority.Mazzardis, Vieno, Kuntsche and Santinello (2010) term this external motive to blend in with peers the conformity motive, which is closely associated with adolescents' internal enhancement motive to have fun and experience excitement (Kuntsche et al., 2005).Louw and Louw (2007) reiterate that the enhancement motive is a normal need in adolescence, since this developmental stage is characterised by curiosity and excitement seeking, thereby suggesting that experimenting with drugs could be regarded as normative behaviour for adolescents (McWhirter et al., 2013).
It is evident from the narratives of the users that the peer group fulfilled several functions during adolescence, ranging from encouraging a movement away from parental influence, identity development, serving as a social outlet and a source of feedback, learning the rules that govern social behaviour, and serving as an active support and informational/consultancy network (Smetana, 2011;Smokowski et al., 2000).The narratives of the users reveal that they were attracted to non-conforming peers by a number of factors.These are listed and illustrated below:  The need to have fun with their friends: User 6: "and I didn't want to listen, I just wanted to do my own thing, because they were fun also for me."  The need to be associated with popular peers at school: User 3: "I was hanging out with them and stuff, because they are the cool guys on the school."  The need for assimilation with their friends: User 4: "All my friends was there and then they also tried it and then I didn't want to be left out."  The need for a sense of belonging derived from older peers, in the absence of parental warmth and support: User 6: "There was a feeling of being unwanted [by parents], making it easier to listen to my friends."  The need for an avoidant coping response to being confronted with adult responsibilities: Non-user 6: "Teenagers sometimes feel like the responsibilities are too much, especially when they have to take on the responsibilities of parents.They then have a need to break free and in the process turn to drugs for comfort."  The benefit of protection (against being bullied) afforded by their peers, expressed as follows by User 3: "and I eventually started to jump school with them and bunk…They are the cool gang and I actually wanted to be just left alone on school, 'cause if you with them, nobody bothers you." These findings attest to the importance of identifying the benefits that adolescents derive from their peer group and for the need to equip them with the skills to elicit these benefits in more constructive ways.The users' experiences resonate with the views expressed by Falkowski (2003) and the National Institute on Drug Abuse ( 2003) on the factors that enhance attraction to negative peer association.These negative peer associations result in the reinforcement of antisocial actions, since their friends' responses provide them with the attention and status they often desire (McWhirter et al., 2013) and the opportunity to escape from the increasing demands placed on adolescents in the 21 st century (e.g.Fatusi & Hindin, 2010).
Several South African and international studies support the findings from the present study, namely that adolescents who have drug-using peers in their friendship circle (i.e.peer social influences) are more inclined to engage in drug use themselves (Brook et al., 2006;Potgieter et al., 2010).The findings furthermore confirm the four mechanisms of peer influence alluded to in the literature review (Karcher et al., 2004).In the narrative that follows User 4 illustrates how she experienced peer influence in a specific social context, where loud mood-altering music, drinking of alcohol and the presence of other friends reduced her resistance to a drug offer: "It was my friend's birthday party, but like it all happened there….There was crowds, there was music playing and it was my first time.I felt like… uhm, I'm now gonna try it out and so, but first I didn't want to, but then I tried it… Ya, and it's how can I say, music playing and we were drinking and then, all my friends was there, and then they also tried it and then I didn't want to be left out." The narratives of User 8 allude to her involvement in gangs where the group norms prescribed involvement in crime as a way of securing access to her drug of choice: "I started hanging out with gangsters who used me to sell drugs from street corners just so that I could get something to smoke." The peer influence mechanism of modelling resonates in the narrative of User 9: "Now I always used to look up to people like [name of friend] because they always used to get the girls and things, so I also wanted to be like them", illustrating his attraction to the confident personas he observed in his friends.
The narratives by both the non-users and users suggest that the adolescents' susceptibility to peer influence was subject to the consequences of resisting peer pressure.The potential consequences of resisting peer pressure is evident from a small selection of dialogue quotes that follows: User 9: "One of my new friends got beaten up by my old friends last year.Because they're now taking me away." User 3: "Now I never, I don't wanna join them [referring to prosocial group], because that group is called the church people, the nerds and stuff." Non-user 17: "Some are forced to use it and if you refuse, they will call you a loser and spread mean rumours about you even if you know what they say ain't true.When you a teen all you can think of is popularity and your reputation at school.You will do anything to be that girl or that guy even if it means taking drugs, you will really be surprised how low people would go to be seen.Teens like to experiment on drugs to impress their friends.Teens do drugs because they think it's cool, they sell drugs to be seen, because they think they have power over learners or people who are not using it.They do it because they want to be respected by others." The narratives of all the users suggest that the consequences of resisting peer pressure range from being ostracised, excluded, blackmailed, having one's reputation tarnished and having one's life threatened, thus corresponding with findings by Kim et al. (2002).Similarly, Smetana (2011) claims that adolescents' judgment on exclusion from their peer groups is multifaceted and informed by their moral views and social hierarchy, with those from low-status school peer groups experiencing exclusion as marginalisation.

Theme 2: Peer protective factors in adolescence
The participants' presentation of peer protective factors mitigating against drug use were very limited in comparison to the peer risk factors.Both non-users and users identified the following three distinct peer protective factors against drug use: i) associating with non-drug using peers; ii) disengaging from negative peer influences; and iii) employing specific peer resistance skills.
The majority of the 29 non-users stated that their association with non-drug using and/or prosocial peers served as a direct deterrent from becoming involved with drugs, as illustrated by Non-user 11: "Well I have friends who don't use it so that makes it easier for me to not become involved with drugs".The narrative by User 5 (who was in the recovery stage of drug dependence) echoed the importance of positive peer selection in a community ravaged by drug use amongst adults and adolescents: "Here in this neighbourhood, there are drugs wherever you go.You therefore have to choose your friends carefully.If you choose the wrong friends, well then that is a decision you make [to use drugs]." The combined narratives of most users and non-users resonate with both international and national research that has confirmed that an environment in which adolescents are surrounded by prosocial peer influences has a greater potential to facilitate the promotion of resilience, as opposed to disorganised communities characterised by the presence of negative peer influence, negative adult role modelling and the absence of positive social and recreational outlets (Eriksson, Cater, Andershed & Andershed, 2010;Van der Westhuizen, 2010).
Notwithstanding these findings from the literature, it is disconcerting to note that few non-users in the present study utilised their witnessing of drug-related harm experienced by adolescents in their neighbourhood as a protective factor against drug use.The quotes from three non-users illustrate how they employed learning derived from witnessing drug using peers as a protective mechanism: Non-user 3: "I am no better than them [referring to the drug using adolescents in her community], but I have dreams and will do everything to reach my goals."These narratives challenge the popular belief that adolescents are passive victims of peer pressure, instead suggesting that they can be active participants in their respective peer circles, asserting their agency to achieve health-promoting outcomes (Ungar, 2006).The narratives of the non-users further highlight the point that peer influences are informed by other contextual factors, which may include the strength of adolescents' relationships with parents, intrapersonal factors (such as goals for the future, high self-esteem and self-efficacy), and attachment to schoolall of which have important implications for the development of drug use prevention interventions.
It was apparent during the data-generation phase with the users that many of them were able to detach themselves from negative peer influences by automatically connecting to prosocial outlets instead of embarking on a goal-directed process of sourcing constructive alternatives.It was also evident that the presence of prosocial adolescents facilitated the entry to these constructive alternatives.User 8, for example, reported joining a youth structure in her church after her sister's peers prompted her to make an exploratory visit to the group.Similarly, User 10 was introduced to an adolescent drug support group by a former prosocial peer from her school.User 5 reported receiving ongoing support from a prosocial girlfriend and a network of prosocial adults, as is evident from the following dialogue: "So the beginning of the year, basically for the half of the year, I was still out of it, way out of it, but I was starting to get slowly but surely my life back on track.The users' decisions to find an alternative social outlet, instead of withdrawing into social isolation, attested to their need for social connectedness (Patrick, Patel, Caldwell, Gleeson, Smith & Wegner, 2010).Supporting the importance of a social network in adolescence, Smokowski et al. (2000) found among disadvantaged youths that positive relations occurred in clusters, and that the value of these clusters was located in positive role modelling and continuous motivational messages.This implies that adolescents who maintain good relations with positive peers also enjoy positive relationships with their parents and/or siblings, as seems to have been the case for most of the users in the present study.
The final peer protective factor that emerged in the findings related to the different mechanisms that users employed to distance themselves from their earlier drug using peer circles.User 7 disengaged systematically by initially remaining in the friendship circle, using a legal drug (alcohol), whilst his peers used illicit drugs (dagga and methamphetamine).Others actively disengaged from their peer groups by altering their daily routines, for instance, avoiding earlier meeting places, travelling a different route to school and volunteering to remain in the classroom or the principal's office during school breaks, as illustrated in the quote by User 3: "I just prevented going to them and stuff.I used to walk a specific road through, but I walk the long way rather now.I sit in the office also sometimes then I just go if the bell rings to class.I used to like stand in the corridors for a few minutes after break, I no more do that -I go straight to class." User 1 reported how his grandparents moved him to a different school, whilst Users 7 and 10 were moved to different towns in their parents' attempts to create distance between them and the availability of drugs.User 4's parent also initiated admission to a Child and Youth Care Centre, as the mother seemingly felt powerless against the peer and community risk factors to which her daughter was exposed, especially as her role as monitoring agent was diminished by her status as a single parent with long working hours.The action taken by User 1, "I actually burn my tattoos out now", demonstrates a more radical step to detach from his affiliation to the gang he belonged to.
Several users employed peer resistance skills, which could also be categorised as individual protective factors, e.g.User 3, who previously succumbed to the taunting of his peers, described how he chose to ignore the teasing by effectively applying the skills of emotional regulation (Bower, Carroll, & Ashman., 2012;Karapetian & Grados, 2005) as follows: "It's still a problem, but I, I don't take note of the people anymore", as well  Peer-focused prevention strategies must focus on facilitating a positive construction of adolescence, and stimulating ideas on how the value of the adolescent peer group can be accessed through growth-promoting avenues.
 Adolescents should be trained on the nature and mechanisms of prosocial peer influence.This approach has the potential to stimulate the co-construction of a wide range of healthy alternatives to drug use during adolescent interactions, satisfying the adolescent's "need for power, recreation, acceptance, or a sense of meaningful participation" (Ungar, 2006:7) needs that are frequently fulfilled by drug use.
 Training in the type of adolescent-centred drug resistance strategies elucidated in the study will also reinforce trust in the adolescents as active agents of change, capable of reducing risk and increasing protection against drug use.
Suggestions for future research include an investigation of how adolescent peer mentors have been effecting positive adolescent development as an alternative to drug use in growth-limiting socio-cultural contexts.Future studies should also be undertaken with a representative sample, which would allow for the generalisation of the results.
-user 18: "I witness the negative impact of drugs on others daily and it is very disappointing and sad to see.So I don't want that for myself."Non-user15:"The things I see in my environment.It is hard to see how teenagers who use drugs destroy their lives."User 2 questioned the dissonance between his drug-using friends' subjective experience of the effects of drugs and his observation of the consequences the drugs were having on them: he challenged his peers' use of drugs directly: "Yeah, I ask them, what is so nice about the drug?So they said, no, it's the feeling.They give me that answer the whole time … the feeling.I don't understand what they feel if they are so lam [inert] and lazy." Furthermore, the dialogue quotes by users 9 and 10 demonstrate how trusting relationships with peers from whom they receive validation and honest feedback strengthened their support base, making it easier to resist negative peer influence.User 9: "Ja [yes] like me and my cousin, we're very close, we will speak everything to each other, and we will give feedback to each other.No matter how big is the problem, she would help me through the problem." The person that gave me the most strength was my girlfriend… she threatened to leave me if I don't stop."