Program logic of a mental health outreach service for socially anxious youth.

Youth mental health is a major concern both in Australia and internationally. Identifying at-risk young people is a core service priority. This report describes a "Youth Social Inclusion Program" developed in Australia to assist young people identified as having social difficulties that prevent them linking into formal mental health services. This report describes the Australian service context and uses the "program explication" methodology to document: the program activities, therapeutic processes, and underpinning evidence for the benefits of this service. Program explication is a consulting method designed to assist services to identify and review implicit program logic assumptions against evaluation literature. We demonstrate the application of this method in exploring the assumptions and evidence for the Youth Social Inclusion Program, targeting outreach to socially anxious young people in Australia. The Youth Social Inclusion Program involved seven logically consistent activities. Our literature search revealed moderate evidence for the expected benefits of the activities. This practice-based design drew on several theoretical perspectives. Given that the program logic has been established, we outline a proposal for further evaluation.

Child, parent, and family mental health and functioning in Australia during COVID-19: comparison to pre-pandemic data.

The COVID-19 pandemic presents significant risks to population mental health. Despite evidence of detrimental effects for adults, there has been limited examination of the impact of COVID-19 on parents and children specifically. We aim to examine patterns of parent and child (0-18 years) mental health, parent substance use, couple conflict, parenting practices, and family functioning during COVID-19, compared to pre-pandemic data, and to identify families most at risk of poor outcomes according to pre-existing demographic and individual factors, and COVID-19 stressors. Participants were Australian mothers (81%) and fathers aged 18 years and over who were parents of a child 0-18 years (N = 2365). Parents completed an online self-report survey during 'stage three' COVID-19 restrictions in April 2020. Data were compared to pre-pandemic data from four Australian population-based cohorts. Compared to pre-pandemic estimates, during the pandemic period parents reported higher rates of parent depression, anxiety, and stress (Cohen's d = 0.26-0.81, all p < 0.001), higher parenting irritability (d = 0.17-0.46, all p < 0.001), lower family positive expressiveness (d = - 0.18, p < 0.001), and higher alcohol consumption (22% vs 12% drinking four or more days per week, p < 0.001). In multivariable analyses, we consistently found that younger parent age, increased financial deprivation, pre-existing parent and child physical and mental health conditions, COVID-19 psychological and environmental stressors, and housing dissatisfaction were associated with worse parent and child functioning and more strained family relationships. Our data suggest wide-ranging, detrimental family impacts associated with the COVID-19 pandemic; and support policy actions to assist families with financial supports, leave entitlements, and social housing.

Psychosocial predictors of binge-drinking residual harm in adolescence and young adulthood: Findings from the Australian Temperament Project.

AIMS: We examine the extent to which adolescent and young adult psychosocial factors are associated with variation in the experience of common types of harm (e.g., injuries, violence, sexual regrets) with respect to binge-drinking frequency - termed residual harm. METHODS: Data were from the Australian Temperament Project, a population-based cohort study that has followed a sample of young Australians from infancy to adulthood since 1983. The current sample comprised 1,081 (565 women). Residual harm was operationalised by saving residuals from models regressing number of alcohol harms onto binge-drinking frequency at each of 5 waves, two in adolescence (15-16 and 17-18 years) and three in young adulthood (19-20, 23-24, and 27-28 years). Psychosocial factors (mental health, social skills, quality of parent and peer relationships) were assessed prior to binge drinking in early adolescence (13-14 years) and then again in young adulthood (19-20 years). RESULTS: Adolescent predictors of decreased residual harm were lower depressive symptoms, and higher cooperation, self-control, and peer and parent attachment. Young adult predictors of decreased residual harm were lower depressive, anxiety, and stress symptoms and peer and parent negative appraisal, and higher responsibility, and peer and parent emotional support. Associations were evident in males and females, although the strength of some associations diminished with age. CONCLUSIONS: Adolescents and young adults with better mental health, social skills, and relationship quality experienced less harm with respect to their binge-drinking frequency. Future research should examine the potential of investment in strength-based interventions for young people.

A Cross-National Comparison of the Development of Adolescent Problem Behavior: a 1-Year Longitudinal Study in India, the Netherlands, the USA, and Australia.

The world youth population is the largest ever and the levels of problem behavior will influence future global health. Recognition of these issues raises questions as to whether adolescent development knowledge can be internationally applied. To date, most research examining adolescent problem behavior has been completed in the United States (USA) and there has been neglected analysis of health inequalities. The aim of the present study was to compare the structure and predictors of problem behavior in representative samples from the USA, Australia, India, and the Netherlands. Two timepoints of longitudinal data were analyzed from the International Youth Development Study that originally recruited state-representative student cohorts in 2002 in Washington State, USA (analytic sample N = 1942) and Victoria, Australia (N = 1957). Similar aged samples were recruited in Mumbai, India, in 2010 (N = 3.923) and the Netherlands in 2008 (N = 682). Surveys were matched and follow-up occurred over 1-year (average baseline ages 12 to 13). CFA identified a latent problem behavior construct comprised of substance use and antisocial behavior indicators. There were cross-national differences in the indicators for this construct. Factor loadings and items were similar between Australia and the USA; however, different items loaded on the construct for the Indian and Netherlands sample. SEM identified that problem behavior at time 2 was predicted by time 1 behavior, with cross-national differences evident. Low parent education was predictive in the USA and India. The number of risk factors present was predictive of problem behavior in all four nations. The findings suggest that evaluated preventative strategies to reduce adolescent problem behavior may have international applications. The analysis of cross-nationally matched longitudinal data appears feasible for identifying prevalence and predictor differences that may signify policy and cultural contexts, to be considered in adapting prevention programs.

Exploring a causal model in observational cohort data: The role of parents and peers in shaping substance use trajectories.

AIMS: To explore the process of applying counterfactual thinking in examining causal determinants of substance use trajectories in observational cohort data. Specifically, we examine the extent to which quality of the parent-adolescent relationship and affiliations with deviant peers are causally related to trajectories of alcohol, tobacco, and cannabis use across adolescence and into young adulthood. METHODS: Data were drawn from the Australian Temperament Project, a population-based cohort study that has followed a sample of young Australians from infancy to adulthood since 1983. Parent-adolescent relationship quality and deviant peer affiliations were assessed at age 13-14 years. Latent curve models were fitted for past month alcohol, tobacco, and cannabis use (n = 1590) from age 15-16 to 27-28 years (5 waves). Confounding factors were selected in line with the counterfactual framework. RESULTS: Following confounder adjustment, higher quality parent-adolescent relationships were associated with lower baseline cannabis use, but not alcohol or tobacco use trajectories. In contrast, affiliations with deviant peers were associated with higher baseline binge drinking, tobacco, and cannabis use, and an earlier peak in the cannabis use trajectory. CONCLUSIONS: Despite careful application of the counterfactual framework, interpretation of associations as causal is not without limitations. Nevertheless, findings suggested causal effects of both parent-adolescent relationships and deviant peer affiliations on the trajectory of substance use. Causal effects were more pervasive (i.e., more substance types) and protracted for deviant peer affiliations. The exploration of causal relationships in observational cohort data is encouraged, when relevant limitations are transparently acknowledged.

A comparison of longitudinal modelling approaches: Alcohol and cannabis use from adolescence to young adulthood.

BACKGROUND: Modelling trajectories of substance use over time is complex and requires judicious choices from a number of modelling approaches. In this study we examine the relative strengths and weakness of latent curve models (LCM), growth mixture modelling (GMM), and latent class growth analysis (LCGA). DESIGN: Data were drawn from the Australian Temperament Project, a 36-year-old community-based longitudinal study that has followed a sample of young Australians from infancy to adulthood across 16 waves of follow-up since 1983. Models were fitted on past month alcohol use (n = 1468) and cannabis use (n = 549) across six waves of data collected from age 13-14 to 27-28 years. FINDINGS: Of the three model types, GMMs were the best fit. However, these models were limited given the variance of numerous growth parameters had to be constrained to zero. Additionally, both the GMM and LCGA solutions had low entropy. The negative binomial LCMs provided a relatively well-fitting solution with fewer drawbacks in terms of growth parameter estimation and entropy issues. In all cases, model fit was enhanced when using a negative binomial distribution. CONCLUSIONS: Substance use researchers would benefit from adopting a complimentary framework by exploring both LCMs and mixture approaches, in light of the relative strengths and weaknesses as identified. Additionally, the distribution of data should inform modelling decisions.

Community variation in adolescent cannabis use in Australia and the Netherlands.

BACKGROUND: While controversial, cannabis use is common, and often begins in adolescence. In Australia and the Netherlands, cannabis use is understood and legislated within a harm-minimisation framework. In Australia it is illegal to purchase or consume cannabis. Cannabis use in the Netherlands is illegal, except in licensed coffee shops where the sale and consumption of small amounts (5 g) are permitted. METHODS: Secondary school aged children in Victoria, Australia and the Netherlands were surveyed using the same culturally adapted questionnaire, designed to measure community rates of substance use and risk factors. Youth cannabis use and predictors of use were compared between the two countries. RESULTS: Similar levels of cannabis use in the last month were reported in Australia and the Netherlands (4.5%), with slightly lower use amongst girls, compared to boys. Common predictors of use were found in both the Australian and Netherlands sample. CONCLUSIONS: The findings suggest that interventions that effectively target similar risk factors to reduce or prevent cannabis use may have applications cross-nationally.

Trial protocol: a clustered, randomised, longitudinal, type 2 translational trial of alcohol consumption and alcohol-related harm among adolescents in Australia.

BACKGROUND: This cluster randomised control trial is designed to evaluate whether the Communities That Care intervention (CTC) is effective in reducing the proportion of secondary school age adolescents who use alcohol before the Australian legal purchasing age of 18 years. Secondary outcomes are other substance use and antisocial behaviours. Long term economic benefits of reduced alcohol use by adolescents for the community will also be assessed. METHODS: Fourteen communities and 14 other non-contiguous communities will be matched on socioeconomic status (SES), location, and size. One of each pair will be randomly allocated to the intervention in three Australian states (Victoria, Queensland and Western Australia). A longitudinal survey will recruit grade 8 and 10 students (M = 15 years old, N = 3500) in 2017 and conduct follow-up surveys in 2019 and 2021 (M = 19 years old). Municipal youth populations will also be monitored for trends in alcohol-harms using hospital and police administrative data. DISCUSSION: Community-led interventions that systematically and strategically implement evidence-based programs have been shown to be effective in producing population-level behaviour change, including reduced alcohol and drug use. We expect that the study will be associated with significant effects on alcohol use amongst adolescents because interventions adopted within communities will be based on evidence-based practices and target specific problems identified from surveys conducted within each community. TRIAL REGISTRATION: The trial was retrospectively registered in September, 2017 ( ACTRN12616001276448 ), as communities were selected prior to trial registration; however, participants were recruited after registration. Findings will be disseminated in peer-review journals and community fora.

Early risk and protective factors for problem gambling: A systematic review and meta-analysis of longitudinal studies.

This systematic review aimed to identify early risk and protective factors (in childhood, adolescence or young adulthood) longitudinally associated with the subsequent development of gambling problems. A systematic search of peer-reviewed and grey literature from 1990 to 2015 identified 15 studies published in 23 articles. Meta-analyses quantified the effect size of 13 individual risk factors (alcohol use frequency, antisocial behaviours, depression, male gender, cannabis use, illicit drug use, impulsivity, number of gambling activities, problem gambling severity, sensation seeking, tobacco use, violence, undercontrolled temperament), one relationship risk factor (peer antisocial behaviours), one community risk factor (poor academic performance), one individual protective factor (socio-economic status) and two relationship protective factors (parent supervision, social problems). Effect sizes were on average small to medium and sensitivity analyses revealed that the results were generally robust to the quality of methodological approaches of the included articles. These findings highlight the need for global prevention efforts that reduce risk factors and screen young people with high-risk profiles. There is insufficient investigation of protective factors to adequately guide prevention initiatives. Future longitudinal research is required to identify additional risk and protective factors associated with problem gambling, particularly within the relationship, community, and societal levels of the socio-ecological model.

The density of alcohol outlets and adolescent alcohol consumption: An Australian longitudinal analysis.

Higher density of alcohol outlets has been linked to increased levels of adolescent alcohol-related behaviour. Research to date has been cross-sectional. A longitudinal design using two waves of annual survey data from the Australian arm of the International Youth Development Study was used. The sample comprised 2835 individuals with average age at wave 2 of 14 years (SD=1.67; range=11-17 years). GSEM was used to examine how absolute levels of alcohol outlet density was associated with student-reported alcohol use one year later, while controlling for prior alcohol use, risk factors at wave one and changes in density over the 2 years. Adolescents' perception of alcohol availability and friends' alcohol use were tested as potential mediators of the association between alcohol outlet density and adolescent alcohol use. Elasticity modelling identified a 10% increase in overall density at wave one was associated with an approximately 17% increase in odds of adolescent alcohol consumption at wave two. Living in areas with a higher density of outlets was associated with a statistically significant increase in the likelihood of adolescents developing early age alcohol consumption.

The relationship between the density of alcohol outlets and parental supply of alcohol to adolescents.

This study investigated whether the number of alcohol outlets per 10,000 population in a given area (density) influenced parental supply of alcohol to adolescents; differences in Australian born and acculturating parents were also examined. A state-representative student survey in Victoria identified that the majority of adolescents (55%) reported that they had used alcohol in the past 12months; 34 % of those who had consumed alcohol reported that it had been supplied by their parents. Multilevel modelling identified that there were no overall effects of density, however there were different effects based on parent country of birth and type of license. Specifically, each unit increase in the density of takeaway liquor stores increased the likelihood by 2.03 that children with both Australian-born parents would be supplied alcohol. Adolescents with both migrant parents on the other hand, had a 1.36 increased risk of being supplied alcohol as the density of outlets requiring at-venue consumption increased. The findings of this study suggest that in Australia, alcohol outlet density is associated with parental supply of alcohol to children, with this effect moderated by the cultural background of the parent and type of outlet density. Future research should investigate the association between the density of alcohol outlets and public approval of parents supplying alcohol to adolescents.

Associations between alcohol outlet densities and adolescent alcohol consumption: a study in Australian students.

OBJECTIVE: To assess whether the density of alcohol sales outlets in specific geographic communities is associated with adolescent alcohol consumption. METHOD: A cross-sectional representative sample of secondary school students from Victoria, Australia (N=10,143), aged between 12 and 17 years, self-reported on alcohol use in the last 30 days in 2009. The density of alcohol outlets per local community area was merged with this information. RESULTS: After controlling for risk factors, multilevel modelling (MLM) revealed a statistical interaction between age and density on alcohol consumption. While older adolescents had higher alcohol consumption, increases in the density of alcohol outlets were only significantly associated with increased risk of alcohol consumption for adolescents between the ages of 12 and 14. CONCLUSION: Increased alcohol availability was associated with an increased risk of alcohol consumption specifically for early adolescents (12 and 14 years). Potential mechanisms as to how density is associated with direct and indirect alcohol availability, such as through parents or older siblings, need to be explored in future research.

Interventions to reduce harm associated with adolescent substance use.

A major proportion of the disease burden and deaths for young people in developed nations is attributable to misuse of alcohol and illicit drugs. Patterns of substance use established in adolescence are quite stable and predict chronic patterns of use, mortality, and morbidity later in life. We integrated findings of systematic reviews to summarise evidence for interventions aimed at prevention and reduction of harms related to adolescent substance use. Evidence of efficacy was available for developmental prevention interventions that aim to prevent onset of harmful patterns in settings such as vulnerable families, schools, and communities, and universal strategies to reduce attractiveness of substance use. Regulatory interventions aim to increase perceived costs and reduce availability and accessibility of substances. Increasing price, restricting settings of use, and raising legal purchase age are effective in reducing use of alcohol and tobacco and related harms. Screening and brief intervention are efficacious, but efficacy of a range of treatment approaches has not been reliably established. Harm-reduction interventions are effective in young people involved in risky and injecting substance use.

Mental health promotion and socio-economic disadvantage: lessons from substance abuse, violence and crime prevention and child health.

ISSUE ADDRESSED: Mental health promotion aimed at populations with low socio-economic status (SES) may benefit by investigating prevention strategies that effectively address related child and adolescent problems. METHODS: Evidence from a number of literature reviews and program evaluations was synthesised. First, the impact of SES on development from childhood to adulthood is considered in light of research on substance abuse, violence, crime, and child development problems. Second, evaluations of interventions are reviewed to identify those that have shown outcomes in research studies (efficacy) or in real-world settings (effectiveness) in reducing developmental problems associated with low SES. Low SES is measured in different ways including low levels of education and/or income or definitions that combine several variables into a new indicator of low SES. RESULTS: Factors associated with low SES are also associated to varying extent with the development of violence and crime, substance abuse and child health problems. Interventions that address underlying determinants of low SES show strong efficacy in decreasing adolescent crime and violence and effectiveness in improving child health outcomes. Although there is limited efficacy evidence that substance abuse prevention can be effectively addressed by targeting low SES, programs designed to improve educational pathways show some efficacy in reducing aspects of adolescent substance use. CONCLUSION: Mental health promotion strategies can draw on the approaches outlined here that are associated with the prevention of child and adolescent problems within low SES communities. Alternatively, such interventions could be supported in mental health promotion policy as they may assist in preventing related problems that undermine mental health.

Consequences of youth tobacco use: a review of prospective behavioural studies.

BACKGROUND: Cigarette smoking represents a significant health problem and tobacco has been identified as causing more preventable diseases and premature deaths than any other drug. Although health consequences from smoking have been documented, there has been a surprising lack of research into behavioural consequences. AIMS: To review what is known of the long-term relationship between patterns of tobacco use prior to age 18 years and behavioural consequences in adulthood. METHOD: A literature search of electronic abstraction services from 1980 to September 2005 was conducted. To be included in the review, studies had to have large, representative samples, be longitudinal studies with baseline age under 18 years and follow-up age 18 years or older and clarify effects due to attrition, leaving 16 articles that met the inclusion criteria. Two reviewers evaluated each paper. FINDINGS: Adolescent tobacco use predicts a range of early adult social and health problems. Surprisingly few studies met the inclusion criteria. The limited evidence available suggests that adolescent tobacco smoking increases the likelihood of early adult tobacco use and the initiation of alcohol use or the development of alcohol-related problems. The link between adolescent tobacco use and subsequent cannabis use was not resolved convincingly from the studies summarized. The effects of tobacco use on later illicit drug use tended to fall away when adjusting for underlying risk factors. Existing studies of the effects of tobacco use on later mental health have many limitations. Nevertheless, a finding that youth tobacco use may predict subsequent mental health problems deserves further investigation. The possible effects of tobacco use on academic/social problems and sleep problems also warrant further investigation. CONCLUSION: This review highlights links between youth tobacco use and subsequent behavioural and mental health problems. It provides health care professionals with evidence of the possible harmful effects of youth tobacco smoking on later social, emotional, and behavioural well-being.

Youth alcohol and other drug use in the United States and Australia: a cross-national comparison of three state-wide samples.

Although youth drug and alcohol harm minimization policies in Australia are often contrasted with the abstinence and zero tolerance policies adopted in the United States, there has been little research directly comparing youth substance use behaviour in the two countries. Three state representative samples in Victoria, Australia (n = 7898) and in the US states of Oregon (n = 15,224) and Maine (n = 16,245) completed a common cross-sectional student survey. Rates of alcohol use (lifetime alcohol use, recent use in the past 30 days), alcohol use exceeding recommended consumption limits (binge drinking: five or more drinks in a session), other licit drug use (tobacco use), and norm-violating substance use (substance use at school, use in the past 30 days of marijuana or other illicit drug use) were compared for males and females at ages 12-17. Rates were lower (odds ratios 0.5-0.8) for youth in Maine and Oregon compared to Victoria for lifetime and recent alcohol use, binge drinking and daily cigarette smoking. However, rates of recent marijuana use and recent use of other illicit drugs were higher in Maine and Oregon, as were reports of being drunk or high at school. In contradiction of harm minimization objectives, Victoria, relative to the US states of Oregon and Maine, demonstrated higher rates of alcohol use exceeding recommended consumption limits and daily tobacco use. However, findings suggested that aspects of norm-violating substance use (substance use at school, marijuana use and other illicit drug use) were higher in the US states compared to Victoria.

Health-risk behaviours among adolescent survivors of childhood cancer.

BACKGROUND: The purpose of the study was to investigate smoking, alcohol use, illicit drug use and sexual risk taking among adolescent survivors of childhood cancer treated in Australia. PROCEDURE: A comparison study selecting on exposure (cancer vs. healthy) and administering a branched computerised questionnaire assessing health-risk behaviour, predominately by telephone interview. One hundred fifty-three adolescent survivors of childhood cancer were compared with age matched healthy adolescents drawn from one of two Australian population based surveys of adolescent health. Behaviours assessed were tobacco use, alcohol use, binge drinking, cannabis use, pain reliever use, other illicit drug use and unprotected sex. RESULTS: Compared to their healthy peers, younger survivors (13- to 17-years) were at an increased risk of reporting pain reliever use (OR = 2.1) for non-medical purposes, but lower risk of binge drinking (OR = 0.20), cannabis use (OR = 0.25), other illicit drug use (OR = 0.31), tobacco use (OR = 0.38) and alcohol use (OR = 0.44). Older survivors (18- to 24-years) were at an increased risk of reporting alcohol use (OR 1.5), but at lower risk of reporting cannabis use (OR = 0.27), other illicit drug use (OR = 0.44) and tobacco use (OR = 0.47). Survival analysis using the full adolescent survivor cohort (13- to 24-years) showed that the age of onset of tobacco use was later for cancer survivors (hazard ratio HR = 0.65). CONCLUSION: Adolescent cancer survivors show reduced involvement in most health-risk behaviours, with the exception of pain reliever use among younger survivors and alcohol use among the older survivors. Although risks were reduced a substantial proportion of survivors engage in these behaviours.

Prediction of alcohol-related harm from controlled drinking strategies and alcohol consumption trajectories.

AIMS: To establish predictors of age 21 alcohol-related harm from prior drinking patterns, current levels of alcohol consumption and use of controlled drinking strategies. PARTICIPANTS: One thousand, five hundred and ninety-six students recruited from an initial sample of 3300 during their final year of high school in 1993. DESIGN: Longitudinal follow-up across five waves of data collection. SETTING: Post high school in Victoria, Australia. MEASUREMENTS: Self-administered surveys examining a range of health behaviours, including alcohol consumption patterns and related behaviour. FINDINGS: Drinking behaviours at age 21 were found to be strongly predicted by drinking trajectories established through the transition from high school. Multivariate regression analysis revealed that alcohol-related harms at age 21 were reduced where current levels of alcohol use fell within limits recommended in Australian national guidelines. After controlling for this effect it was found that the range of strategies employed by participants to control alcohol use maintained a small protective influence. Post-high-school drinking trajectories continued to demonstrate a significant effect after controlling for current behaviours. FINDINGS: revealed that over one quarter of males and females drank alcohol, but on a less-than-weekly basis. This pattern of alcohol use demonstrated considerable stability through the post-school transition and was associated with a low level of subsequent harm at age 21. CONCLUSIONS: Future research should investigate whether encouraging more Australian adolescents to drink alcohol on a less-than-weekly basis may be a practical intervention target for reducing alcohol-related harms.

Family risk factors for cannabis use: a population-based survey of Australian secondary school students.

The objective of this study was to investigate relationships between adolescent cannabis use and indices of parent - child attachment, family functioning and parent attitudes to drugs and delinquency. A total of 2848 year 9 and 2363 year 11 students participated in the Victorian Adolescent Health and Well-Being Survey (1999). The study was a school-based random sample of 535 metropolitan and rural, government and non-government secondary schools throughout Victoria, Australia. Cannabis use was defined as 'any' and 'weekly' use in the last 30 days. Multivariate logistic regression was used to identify independent associations between cannabis use and parent - child attachment, family functioning and parent attitudes to drugs and delinquency. Cannabis use in year 9 was associated with permissive parent attitudes to drugs and delinquency (any use: adjusted odds ratio (OR) = 8.1; weekly use: adjusted OR = 7.6), and was particularly sensitive to small changes in the quality of the parent - child relationship with risk increasing threefold for those describing their attachment as 'good' compared with 'very good' (any use: adjusted OR = 2.8, weekly use adjusted OR = 2.9). A similar, but more moderate pattern association was evident in year 11. After adjusting for other family and background factors, poor family functioning showed minimal association with level of cannabis use at both year levels. Results suggest that intervention efforts might sensibly target strengthening parent - children relationships and promoting less permissive parent attitudes to drug use.

Treatment level progress and time spent in treatment in the prediction of outcomes following drug-free therapeutic community treatment.

UNLABELLED: Previous research has suggested that time spent in treatment is the most important predictor of therapeutic community treatment outcomes. Although level divisions or treatment stages represent the basic structure for segmenting responsibility and authority within traditional therapeutic communities, the relationship of this treatment component to treatment outcome has not previously been investigated. AIM: To test the hypothesis that higher treatment level attainment, more time spent in treatment and additional time spent within a treatment level has a linear association with improvements at outcome. DESIGN: Retrospective quasi-experimental. PARTICIPANTS: Four hundred and twenty-seven ex-residents, stratified according to their highest level of treatment in the Melbourne Odyssey House therapeutic community between 1984 and 1988 were targeted for follow-up and 60% were successfully located and interviewed an average of 5.6 years after their first Odyssey entry. Analysis of official records of methadone treatment, convictions and incarcerations demonstrated no post-treatment differences comparing those interviewed with those not interviewed. MEASUREMENTS: Drug use, criminal involvement and employment. FINDINGS. Although both level attainment and time spent in treatment had a linear relationship to improved outcomes, level attainment was a better predictor of outcomes at treatment exit. An unexpected finding was that those who had spent the median time or longer in particular levels demonstrated worse outcomes on official conviction records and on self-reports of employment compared to those remaining for less than the median time. CONCLUSION: The results suggest that it may be level progress rather than simply time spent in treatment that best explains improved functioning following exit from the therapeutic community.