Youth retention: Factors associated with treatment drop-out from youth alcohol and other drug treatment

Transcription

1 Drug and Alcohol Review (November 2009), 28, DOI: /j x Youth retention: Factors associated with treatment drop-out from youth alcohol and other drug treatment RIA SCHRODER, DOUG SELLMAN, CHRIS FRAMPTON & DARYLE DEERING National Addiction Centre, Department of Psychological Medicine, University of Otago, Christchurch, New Zealand Abstract Introduction and Aims. This study examined factors associated with treatment drop-out among young people aged years attending alcohol and other drug (AOD) treatment. Design and Methods. Data were gathered from structured interviews (n = 79) and a clinical file search of 184 randomly selected young people who had attended youth specific AOD treatment services in Aotearoa, New Zealand during 2003 or Results. The median length of stay was 2.7 months for those attending day/residential services (n = 42) and 4.0 sessions for those attending outpatient services (n = 37) 16.7% of participants from day/residential services dropped out of treatment early (within the first month) and 32.4% of participants from outpatient treatment services dropped out of treatment early (before the third session). Fixed client characteristics, such as age, sex, ethnicity, substance use and mental health diagnoses were not found to be associated with treatment retention. Dynamic client characteristics, such as motivation to attend treatment and expectations about treatment outcomes and program characteristics, such as positive experiences with treatment staff and feeling involved in the treatment process were found to be associated with treatment retention. Discussion and Conclusions. The findings of this study support previous research indicating that fixed client characteristics are not sufficient to explain youth retention in AOD treatment. Of more use are dynamic client characteristics and program variables. These findings stress the potential for improving treatment retention by creating more youth appropriate services. [Schroder R, Sellman D, Frampton C, Deering D. Youth retention: Factors associated with treatment drop-out from youth alcohol and other drug treatment. Drug Alcohol Rev 2009;28: ] Key words: youth, substance abuse treatment, retention, drop-out. Introduction Youth are notorious for dropping out of alcohol and other drug (AOD) treatment [1 4].This fact combined with an increased understanding of the importance of treatment retention in improving AOD treatment outcomes [5,6] stresses an urgent need to better understand factors associated with young people staying in or leaving AOD treatment. To date, the few existing studies examining youth retention in AOD treatment programs have produced inconclusive findings. Variables that have been examined can broadly be described as fixed client characteristics, dynamic client characteristics and program characteristics. Fixed client characteristics are described as unchangeable characteristics, such as demographic and background variables [7]. Age [3,4,8 12], ethnicity [4,11,12], sex [3,4,10,11], substance use [1,3,8,10,12], mental health status [1,3,10,11,13 16], criminal history [1,8,10,11], family relations and parental substance use [1,3,10,11], peer relations and substance use [1,3,10], educational and employment history [8,10,11] and source of referral [6,12] are the main fixed client characteristics that have been examined in relation to adolescent AOD treatment retention. Of these, presence of conduct disorder (CD), attention deficit hyperactivity disorder (ADHD) and source of referral are the only variables to have shown any consistent association with treatment retention among youth in AOD treatment. Conduct disorder and ADHD have been found to be associated with decreased treatment retention [13 15] and court referral has been associated with increased treatment retention [6,12]. Ria Schroder PhD, Research Fellow, Doug Sellman MB ChB, PhD (Otago), FRANZCP, FAChAM, Professor, Chris Frampton PhD, Associate Professor, Daryle Deering PhD (Otago), RCN, Lecturer. Correspondence to Dr Ria Schroder, National Addiction Centre, University of Otago, Christchurch, PO Box 4345, Christchurch 8140, New Zealand. Tel: ; Fax: ; ria.schroder@otago.ac.nz Received 29 July 2008; accepted for publication 10 January 2009.

2 664 R. Schroder et al. Given the lack of conclusive findings on fixed client characteristics many researchers now agree that fixed client characteristics are not sufficient to fully explain treatment retention [7]. As such, other important variables referred to as dynamic client characteristics [7] must also be considered. Dynamic client characteristics [7] are those characteristics which describe the changing or dynamic characteristics of the individual (e.g. client perception variables, such as motivation and readiness) (p. 169). The adolescent treatment retention literature examining dynamic client characteristics is sparse. However, the two dynamic client characteristics that have been investigated are motivation and readiness for treatment, with internal motivation shown to be associated with increased treatment retention [3,12]. The importance of understanding the impact of program characteristics is particularly necessary for understanding adolescent treatment retention given that most adolescent AOD treatment programs are not designed specifically for adolescents but are based on adult models of treatment. Studies that have examined staffing factors have shown that both higher levels of experience and higher staff to client ratios have a positive impact on treatment outcome [16,17] and treatment retention [9]. Additionally, receiving help for issues in addition to AOD problems and perceiving that the treatment program was safe has also been associated with increased treatment retention [5]. The aim of the current study was to measure a variety of client and program factors associated with the early treatment drop-out of youth attending AOD treatment services in Aotearoa, New Zealand. A range of instruments were used to obtain objective measures of fixed, program and dynamic characteristics and to gain young people s perspectives on a range of factors found to be associated with treatment retention. Methods Participants Participants were eligible for inclusion in this study if they had attended one of eight youth specific AOD services selected to take part in this study during 2003 or These eight services were selected from a possible 11 services in New Zealand that met the criteria of being either a youth specific AOD treatment service, an AOD treatment service with a youth specific stream or a general youth mental health service with an AOD treatment stream. Each type of service was also required to have at least three staff members. These eight services were chosen because they provided sufficient geographical spread and adequate coverage of the ethnic diversity in New Zealand youth services. Participants were randomly selected from a total pool of 660 possible participants. It was envisaged that approximately 20% of this total pool (n = 140) would be recruited. It was calculated that with a total sample size of 140 participants, there would be 80% power to detect effect sizes >0.5 between split subsamples. During recruitment, contact attempts were made with 433 of the 660 possible participants in random order. Of these 433 participants, 79 (18.2%) were recruited to take part in the study. Of the remaining, 313 (72.3%) were not able to be contacted despite extensive efforts and 41 (9.5%) declined to be interviewed. Given the low rate of successful recruitment further ethical permission was sought to conduct confidential file searches for a selection of the non-participant group in order to test for differences between those young people who attended treatment and participated in the study and those young people who attended treatment but did not participate in this study. For reasons of privacy, recruitment and nonparticipant file searches were conducted by a trained staff member (either administrative or clinical) within each treatment service. Participant interviews and participant file searches were conducted by researchers from the National Addiction Centre or trained interviewers from each treatment service. The data presented in this paper were collected from two main sources: participant interviews and participant file searches. Participant interviews. Interviews were completed in person at a location of the participant s choice. Where this was not possible interviews were conducted by telephone (n = 19). Most participants completed the interview in one session but nine participants (11.4%) chose to complete the interview over two to three sessions. Participant file searches. File searches provided information on exact admission and discharge dates, number of sessions attended, types of substances used before treatment entry, substance use and psychiatric diagnoses at treatment entry and reasons for discharge. Measures Sociodemographic details were collected from all participants. The Mini International Neuropsychiatric Interview Version [18] was used to assess the presence of a range of common psychiatric disorders associated with youth or substance use disorders (SUD). The antisocial personality disorder section was modified to provide a measure of CD before the age of 15 years (CD past) and in the 6 months before interview (CD current).

3 Early drop-out in youth AOD treatment 665 The Alcohol Use Disorders Identification Test (AUDIT) [19] and Cannabis Use Disorders Identification Test (CUDIT) [20] were completed twice; covering the last 6 months and then the 6 months before entering treatment in order to gain a proxy AUDIT and CUDIT score at treatment entry. The Addiction Treatment Retention Questionnaire is a 68 item questionnaire specifically designed for this study by the project team. Items were based on factors that had previously been investigated in the treatment retention literature and on factors that emerged through the authors own clinical experiences. The questions covered a range of variables and examined key areas, such as contexts of substance use, experiences at treatment, expectations of the program before entry, perceptions of substance use issues before treatment, goals for substance use at treatment and motivation for treatment. Each item was framed as a statement and participants were asked to indicate on a 5 point Likert scale the extent to which they agreed or disagreed with each statement. Higher scores indicated greater agreement with each of the items. Nineteen subscales were created from the 68 items, shown to have face validity on piloting, by calculating the mean scores of the appropriate items that represented each subscale. Internal consistency of these subscales was measured using Cronbach s alpha (range , mean 0.75). These subscales depicted three fixed client characteristics, nine dynamic client characteristics and seven program characteristics. For day and residential services treatment length of stay (LOS) was measured in months and days. For outpatient services treatment LOS was measured as number of sessions attended. In order to allow comparison across the whole participant sample retention was measured as a dichotomous variable depicting early treatment drop-out. Early drop-out was defined as leaving outpatient treatment before the third session and leaving day/residential treatment within the first month of treatment. These criteria were based on those commonly used in clinical practice. These cut-offs fall well below the median stay of 3.4 months in day/residential treatment and 6.4 sessions in outpatient treatment and as such indicate a group of young people who drop-out of treatment well before the average client. Analysis Data were entered into spss (Statistical Package for the Social Sciences,Version 13.0) [21] from which descriptive and comparative summaries were generated. Chisquared tests for independence, Fisher s exact tests, Spearman s Rank Order Correlation and Kruskal Wallis non-parametric ANOVA s were carried out to explore relationships between independent variables and treatment modality, study inclusion and retention measures. Results Participants were 79 young people who attended one of eight youth AOD treatment services in New Zealand in 2003 or A total of 55.7% were male, 53.2% had attended a day or residential treatment service and 46.8% had attended an outpatient treatment service. The majority of participants identified as being Maori (45.6%) or European (41.8%) and a small number identified as being of Pacific (10.1%) ethnicity. The only differences noted between young people who took part in the study (participants) and those who did not take part in the study (non-participants n = 105) were related to diagnoses of CD and SUD ascertained from file search data. More nonparticipants than participants were diagnosed as having CD when they entered treatment (P < 0.01) and significantly more participants than non-participants presented at treatment with a SUD only (P = 0.007). At the time of entering treatment participants were aged between 13.8 and 19.6 years (median 16.1 years). A total of 58.2% had a criminal conviction, 53.2% met criteria for cannabis dependence, 50.6% for nicotine dependence and 32.9% for alcohol dependence. Proxy AUDIT and CUDIT scores were 23.0 and 24.5, respectively, at the time of treatment entry. A total of 48.1% of participants were found to have a co-existing SUD and mental health disorder (MHD) at treatment entry namely major depression (26.6%), CD (22.8%) and ADHD (11.4%). Day or residential treatment services were attended by 42 participants and the median LOS was 2.7 months. The majority of young people attending these services stayed between 1 and 4 months (52.4%), whereas 16.7% dropped out of treatment within the first month. The remainder of the participants (n = 37) attended outpatient treatment and the median LOS was 4.0 sessions. A total of 32.4% of clients dropped out of treatment after one or two sessions whereas 40.5% stayed for six or more sessions. Across both treatment modalities 24.1% of participants dropped out of treatment early. Factors associated with early treatment drop-out As seen in Table 1 none of the measured fixed client characteristics were associated with early drop-out.this included factors measured by client self-report and client file searches. The only variable in Table 1 found to be associated with early treatment drop-out was the

4 666 R. Schroder et al. Table 1. Sociodemographic, substance use and psychiatric factors associated with early drop-out among a sample of young people attending AOD treatment in day, residential or outpatient treatment settings Variables All participants (n = 79) Early drop-outs (n = 19) Early engagers (n = 60) P a Median age at treatment entry (years) b Sex (%) 0.76 Male Female Ethnicity (%) 0.57 European Maori Pacific Other Living circumstances at treatment entry (%) 0.52 Family Independent Controlled environment Foster care Ever convicted of a crime (%) 0.97 Yes Source of referral (%) 0.20 Other referred Justice referred Self referred Reason for discharge (%) 0.02* Conducive to treatment Not Conducive to treatment Year at treatment SUD at treatment entry (%) Nicotine dependence (n = 79) Alcohol dependence (n = 50) c Cannabis dependence (n = 56) c AUDIT at treatment entry (median) b CUDIT at treatment entry(median) b Mental health diagnosis at treatment entry (%) Depression Conduct disorder c ADHD c Co-existing disorder 0.58 SUD & MHD No diagnosis SUD only MHD only ATRQ subscales (median) Fixed client characteristics Mother substance user b Father substance user b Experiencing cravings b *P < a Chi-squared analysis unless otherwise stated. b Mann Whitney U. c Corrected for low cell numbers. ADHD, attention deficit hyperactivity disorder; ATRQ, Addiction Treatment Retention Questionnaire; AUDIT, Alcohol Use Disorders Identification Test; CUDIT, Cannabis Use Disorders Identification Test; SUD, substance use disorder; MHD, mental health disorder. reason for discharge variable, where not surprisingly, people who dropped out of treatment early were more likely to have left treatment because of disciplinary reasons or through self-discharge (P = 0.02). Table 2 indicates that five of the nine dynamic client characteristics measured by the Addiction Treatment Retention Questionnaire were found to be associated with early treatment drop-out. Participants were more likely to drop-out of treatment early if they reported less internal motivation (P = 0.03) and greater external pressure (P = 0.01) to engage in treatment; were less likely to have set abstinence as a goal for their substance

5 Early drop-out in youth AOD treatment 667 Table 2. Dynamic client characteristics and program variables associated with early drop-out among a sample of young people attending AOD treatment in day, residential or outpatient treatment settings Variables All Participants (n = 79) Early Drop-Outs (n = 19) Early Engagers (n = 60) P a ATRQ subscales (median) Dynamic client characteristics Perceived Problem with Substance Use Internally Motivated to Attend Treatment * Externally motivated to attend treatment * Substance-related treatment goals * Treatment goals achieved Expected positive general life outcomes <0.01** Expected positive AOD outcomes * Important to complete treatment Treatment as expected Program characteristics Treatment goals set and involved <0.01** Good relationship with staff <0.01** Good relationship with other youth Felt connected to program Fun important in program Transportation issues Treatment format useful *P < 0.05, **P < a Mann Whitney U. AOD, alcohol and other drug; ATRQ, Addiction Treatment Retention Questionnaire. use (P = 0.04); were less likely to expect that treatment would help them to make changes in their life in general (P < 0.01) and in relation to their substance use specifically (P = 0.01). A number of program-related variables were also found to be either significantly associated with treatment drop-out or indicated a trend towards significance. A highly significant association was found between participants perceptions of being involved in goal setting and treatment retention. Participants were significantly more likely to drop-out of treatment early if they felt they had failed to set clear treatment goals and had not been included in setting treatment goals (P < 0.01); reported less positive experiences with treatment staff in terms of feeling safe, comfortable and supported by staff and being able to express themselves openly and honestly to staff (P < 0.01). Discussion The aim of this study was to examine client and program factors associated with treatment retention among young people attending AOD treatment services in New Zealand. This is the first study of its kind in New Zealand and some useful information about factors associated with treatment retention has emerged. The findings of this study support previous research indicating that dynamic client characteristics play an important role in treatment retention [7]. The current study also suggests that, for the most part, fixed client characteristics are not particularly useful in predicting who might stay or leave treatment. Similarly, Melnick [12] found that a young person s internal motivation to engage in treatment is an important factor related to longer treatment retention. To the best of our knowledge, previous studies have not examined how factors, such as expectations about treatment are related to youth treatment retention. Positive results from the current study suggest that this might be a useful area to explore in future research. The findings that dynamic client characteristics play a significant role in what makes young people stay in or leave AOD treatment stresses that treatment services, and the individuals attending them, can work together to improve treatment retention. This fits with the conceptualisation of treatment as a client journey that is influenced by a multitude of factors. The potential for service providers to influence client engagement and retention is limited when the focus is primarily on fixed client characteristics. The findings from this and other studies suggest that these client characteristics cannot be relied on to provide an accurate picture of why a young person might stay in or leave treatment. Program variables, such as feeling connected to the program, having a good relationship with staff and most importantly feeling empowered to have a say in areas, such as goal setting and what happens in treatment

6 668 R. Schroder et al. were found to be associated with longer stays in treatment. These findings support existing literature, which suggests that staff attitudes and abilities to form relationships with young people in treatment have a significant impact on treatment retention [9,3]. By being aware of dynamic client characteristics and program variables that might impact on treatment retention, service providers have the opportunity to influence treatment retention and to contribute to greater positive client outcomes. Furthermore, recognising that factors, such as internal motivation and expectations of treatment might impact on how long a young person stays in treatment indicates important areas to focus on with youth at treatment entry. The conclusions that can be drawn from this study must be considered in relation to its methodological limitations. The most obvious limitations are the small sample size, the low participation rate, the retrospective nature of the data collected and the under representation of young people with CD. The small sample size and the under representation of young people with CD suggest there is a specific population of youth who have not been accessed in this study, thus limiting the generalisation of findings to this group. In addition, the retrospective nature of the data means that many of the measures of pre-treatment characteristics and treatment experiences can only be viewed as proxy measures. As such this study should be viewed as an important first step in investigating factors associated with AOD treatment retention among young people in Aotearoa, New Zealand, but should provide fertile ground for future prospective studies in this area. References [1] Galaif E. Prospective risk factors and treatment outcomes among adolescents in DATOS-a. J Adolesc Res 2001;16: [2] Orlando M, Chan KS, Morral AR. Retention of courtreferred youths in residential treatment programs: client characteristics and treatment process effects. Am J Drug Alcohol Abuse 2003;29: [3] Hser YI, Grella CE, Hubbard RL, et al. An evaluation of drug treatments for adolescents in 4 US cities. Arch Gen Psychiatry 2001;58: [4] Pompi KF, Resnick J. Retention of court-referred adolescents and young adults in the therapeutic community. Am J Drug Alcohol Abuse 1987;13: [5] Stark MJ. Dropping out of substance abuse treatment: a clinically oriented review. Clin Psychol Rev 1992;12: [6] Williams RJ, Chang SY. Addiction Centre Adolescent Research Group. A comprehensive and comparative review of adolescent substance abuse treatment outcome. Clin Psychol: Sci Pract 2000;7: [7] De Leon G, Melnick G, Kressel D. Motivation and readiness for therapeutic community treatment among cocaine and other drug abusers. Am J Drug Alcohol Abuse 1997;23: [8] Friedman AS, Glickman NW, Morrissey MR. Prediction to successful treatment outcome by client characteristics and retention in treatment in adolescent drug treatment programs: a large-scale cross validation study. J Drug Educ 1986;16: [9] Friedman AS, Glickman NW. Residential program characteristics for completion of treatment by adolescent drug abusers. J Nerv Ment Dis 1987;175: [10] Blood L, Cornwall A. Pretreatment variables that predict completion of an adolescent substance abuse treatment program. J Nerv Ment Dis 1994;182:14 9. [11] Mossman E. What works with youth? an evaluation of the Adventure Development Counselling programme. Unpublished doctoral thesis. University of Canterbury, Christchurch, New Zealand, [12] Melnick G, De Leon G, Hawke J, Jainchill N, Kressel D. Motivation and readiness for therapeutic community treatment among adolescents and adult substance abusers. Am J Drug Alcohol Abuse 1997;23: [13] Horwood LJ, Fergusson DM. Psychiatric disorder and treatment seeking in a birth cohort of young adults. Christchurch, New Zealand: Ministry of Health, [14] Wise B, Cuffe S, Fisher J, Cole P, Wagner J. Dual diagnosis and successful participation of adolescents in substance abuse treatment. J Subst Abuse Treat 2001;21: [15] Siegal HA, Rapp RC, Fisher J, Cole P, Wagner JH. Treatment dropouts and noncompliers: two persistent problems and a programmatic remedy. In: Inciardi JA, Tims FM, et al. eds. Innovative approaches in the treatment of drug abuse: program models and strategies Contributions in criminology and penology. No 39. Westport, CT, USA: Greenwood Press/Greenwood Publishing Group, Inc., 1993: [16] Fergusson DM, Horwood LJ, Lynskey MT. The comorbidities of adolescent problem behaviors: a latent class model. J Abnorm Child Psychol 1994;22: [17] Ralph N, McMenamy C. Treatment outcomes in an adolescent chemical dependency program. Adolescence 1996;31: [18] Sheehan DV, Lecrubier Y, Sheehan KH, et al. The Mini- International Neuropsychiatric Interview (M.I.N.I.): the development and validation of a structure diagnostic psychiatric interview for DSM-IV and ICD-10. Journal of Clinical. Psychiatry 1998;59 (Suppl 20): [19] Saunders JB, Aasland OG, Babor TE, de la Reuente JB, Grant M. Development of the Alcohol Use Disorders Identification Test (AUDIT): WHO collaborative project on early detection of persons with harmful alcohol consumption, part 2. Drug Alcohol Rev 1993;88: [20] Adamson SJ, Sellman JD. A prototype screening instrument for cannabis use disorder: the Cannabis Use Disorder Identification Test (CUDIT) in an alcohol-dependent clinical sample. Drug Alcohol Rev 2003;22: [21] SPSS for Window Release13.0. Chicago: SPSS Inc., 2004.