Date of Degree


Document Type

PhD diss.

Degree Name

PhD (Doctor of Philosophy)


Social Work

First Advisor

Jeanne Saunders


Substance use disorders (SUD) and mental health disorders are often thought of as completely separate problems even though these disorders commonly co-occur. Among adolescents who seek treatment for substance use problems, co-occurring mental health problems (MHP) are common. This is concerning because co-occurring disorders among adults have been associated with more severe MHPs, relapsing to substance use sooner, being less likely to maintain abstinence, and other problems. Despite the awareness that co-occurring disorders are problematic for adolescents, few studies have been conducted to understand these problems with an adolescent sample.

The purpose of this study was to understand if factors commonly related to co-occurring disorders among adults were the same for adolescents and to examine two measurement models for the dependent variable, substance use. This secondary analysis of data first examined characteristics that are related to MHPs among a sample of adolescents (N=801) who use substances. Factors in three domains were examined: demographics, substance use, social factors. The results indicated that among these adolescents, MHPs were common. Two key factors related to having a MHP were gender and the severity of the SUD. Females and adolescents with more severe SUDs, like dependence, were more likely to have MHPs. Furthermore, severity of the SUD partially mediated the relationship between several of the other factors and MHPs. The type of substance an adolescent reported using was also important. In addition, adolescents who had more peers and more family members who participated in deviant activities had more severe substance use problems and were more likely to have a MHP.

In addition to examining the factors related to mental health problems among a substance using population, this study modeled the dependent variable in two different ways which had not been done before and allowed for variance in the measure to be accounted for in the model. MHPs were measured both continuously and categorically. The results of the comparison indicated that there were not major differences between the two models. Implications for social work practice, policy and research are discussed.


2, v, 147




Copyright 2010 Amanda Rose Reedy