Youth in transition

Substance use disorders (SUD) are prevalent in the population, tend to follow a chronic course, are linked to many problems,
and often have their onset in adolescence. Hence, substance use and SUD among adolescents and young adults should be a
primary target of mental health research. However, the literature regarding the development, persistence or desistance, and
treatment of SUD in youth is sparse, and research in this area lags considerably behind research in adults. We will fill this gap
by investigating the course of SUD and related problems in two – interrelated – studies.
First, in Study 1, we will use existing data from the prospective Tracking Adolescents’ Individual Lives Survey (TRAILS) study
(N=2230), in which the development of mental health was studied from preadolescence (age 11 years) to adulthood (up to age
26 years) in the Dutch general population. We will investigate the rate of persistent substance dependence (DSM-IV) from
adolescence to young adulthood (up to age 26 years) in the general population, and examine which adolescent characteristics
are most predictive for persistent substance dependence in the general population.
Second, in Study 2, we will conduct a multi-center prospective cohort study in a representative sample of 420 youth who apply
for treatment at eight addiction treatment organizations in the Netherlands. We will determine the rate of persistent DSM-5
moderate to severe SUD among these youth in addiction treatment from treatment-entry to 2 year and 4 year follow-up (i.e.,
from adolescence to young adulthood).
Next, we will use the same set of predictors found among adolescents in the general population in Study 1, to predict
persistent DSM-5 moderate to severe SUD among youth in addiction treatment in Study 2. We will subsequently optimize the
predictive accuracy of this population-based set of predictors by adding patient-related predictors, collected at treatment-entry
in our treatment sample, to the prediction model. These baseline patient-related predictors include phenotype, endophenotype
(level of impulsive choice) as well as biological (hair cortisol) indicators.
In addition, in Study 2 we will examine which distinct longitudinal outcome trajectories can be identified from adolescence to
young adulthood, pertaining not only to SUD, but also to comorbid mental health problems and social functioning, using latent
growth curve modeling.
Lastly, we will examine which addiction treatment interventions (type, intensity, duration), for which youth, are associated
with favorable or unfavorable long-term treatment outcomes in the areas of SUD, comorbid mental health problems and social
functioning, to provide a more solid basis for the development of personalized treatment.
Treatment will not be modeled or otherwise adapted for the purpose of this study, but will consist of the usual, ‘real-world’,
treatment-offer provided by the addiction care organizations in the Netherlands to maximize ecological validity of the study
outcomes.
During the development of this study proposal, we organized several focus group meetings with youth in addiction treatment, to
obtain their feedback about which outcome measures they considered to be relevant for this study, and to obtain their advice
about the best ways to stay in touch with the participating youth and keep them interested in (continued) participation in the
study. We used the outcomes of these focus group meetings to adapt and refine our study assessments, outcome measures,
and procedures for re-contacting youth at the 2 year and 4 year follow-up.
The studies proposed here will help us to better understand the course of SUD, its relation with comorbid health problems and
social functioning and their determinants and consequences among youth going from adolescence to young adulthood. The
study results will provide vital, currently lacking, information for both policy makers and health professionals involved in
prevention and treatment planning, with respect to the long-term outcomes of addiction treatment, early detection of youth at
high risk for chronicity, and for the development of personalized treatment.

The overall aim is to improve early detection of youth at high risk for chronicity, and to provide a more solid basis for the development of personalized treatment.

Primary research questions:
(1) What is the rate of persistent DSM-5 moderate to severe SUD from adolescence to young adulthood among youth in addiction treatment?

(2) What is the prognostic value of a general population-based set of predictors of persistent SUD from adolescence to young adulthood, derived from a secondary analysis of data from the Tracking Adolescents' Individual Lives Survey (TRAILS) study, for predicting persistent DSM-5 moderate to severe SUD among youth in addiction treatment?

(3) Can we optimize the accuracy of predicting persistent DSM-5 moderate to severe SUD among youth in addiction treatment by extending or modifying the TRAILS general population-based set of predictors with baseline indicators from our treatment sample?

(4) Which distinct longitudinal treatment outcome trajectories can be identified from adolescence to young adulthood pertaining to SUD, comorbid mental health problems, and social functioning among youth in addiction treatment?

(5) Which treatment interventions - in terms of type, intensity and duration - are associated with favorable or unfavorable long-term outcomes, in terms of SUD, mental health problems and social functioning, for which youth in addiction treatment?

The assessments will be done at:
- Baseline (start of index treatment)
- End of treatment (i.e. the index treatment)
- 2 years after baseline
- 4 years after baseline