The Relationship Between Mental Health and Substance Abuse Among Adolescents 

Previous PageTable Of ContentsList Of TablesNext Page

1.3 Literature Review

A growing body of evidence indicates that the comorbidity of mental disorders and substance use is far more extensive than was previously assumed. The National Institute of Mental Health Epidemiological Catchment Area Program (ECA) evaluation of comorbidity in a general population sample aged 18 and older found that any past history of mental disorder was associated with more than twice the risk of having an alcohol disorder, and over four times the risk of having another drug disorder (Regier et al., 1990). The National Comorbidity Survey (NCS) found that mood, anxiety, antisocial personality disorder, and substance use disorders were highly comorbid in a general population sample aged 15 to 54 (Kessler et al., 1994). The lifetime co-occurrence of mental disorders with addictive disorders was estimated to be approximately 50 percent (Kessler et al., 1996). The mental disorder preceded the addictive disorder in 83.5 percent of co-occurrences, developing most frequently during adolescence.

Research and theory concerned with substance use and with mental disorders are guided by the common assumption that an underlying predisposition interacts with psychosocial risk factors to determine the expression of disorder. Overall risk may be influenced by biological, psychological, and  sociocultural factors. While risk factors have been identified for both mental disorders and substance use among adolescents, the extent of association has not been established. Failure to consider the association may have serious consequences to the interpretation of findings concerned with adolescent psychopathology. Substance use associated with emotional or behavioral problems may evidence different developmental history, more severe functional impairment, and poorer prognosis. If the co-occurrence is not taken into consideration, the presence of an unmeasured condition may confound findings concerning etiology, course, treatment, transmission, and classification for either mental illness or substance abuse (Angold et al., 1993).

The prevalence of co-occurring emotional and behavioral problems and addictive disorders across the period of adolescence has not been clearly established. Because restricted samples for this age range were surveyed, and adolescent-specific psychiatric disorders were not considered, the conclusions of the ECA and the NCS concerning adolescence are limited. Estimates of co-occurring mental disorders and substance use problems among adolescents range from 22 to 82 percent (Boyle & Offord, 1991; Cohen et al., 1993; Brown, Mott & Stewart, 1992; Eisen, Youngman, Grob & Dill, 1992; Caton et al., 1989; Greenbaum et al., 1991; Groves, Batey & Wright, 1986; Lewinsohn, Rohde & Seeley, 1995; Roehrich & Gold, 1986; Rohde, Lewinsohn & Seeley, 1991; Stowell & Estroff, 1992). Inconsistent assessment of both substance use and mental disorders, and other methodological differences, make it difficult to generalize or compare findings, and have resulted in conflicting prevalence estimates.

A striking pattern of association among adolescents was detailed in a SAMHSA report based on the 1994 NHSDA. This report focused on psychosocial difficulty and specific substance use among non-institutionalized adolescents aged 12 to 17 ( SAMHSA, 1996c). Psychosocial problems for the prior six months were assessed with the YSR. Measures of psychosocial problems were found to be highly associated with substance use among adolescents. Past-month cigarette smoking and binge drinking (five or more drinks on the same occasion) were associated with high ratings for psychosocial problems in the past six months. For marijuana use, the relationship was more pronounced. While overall substance use is generally higher for adolescent males than for females, females with high ratings for psychosocial problems as measured by the YSR3 were as likely as males to smoke cigarettes, binge drink, or use illicit drugs.

Many studies concerned with adolescent psychopathology have found evidence of a relationship between emotional or behavioral problems and substance use. Hyperactivity in childhood (Hechtman, Weiss & Perlman, 1984) and antisocial behavior (e.g., Jones, 1968; Cadoret et al., 1986) have been associated with vulnerability to substance use. Tobacco use is most pronounced among high schoolstudents with behavioral problems (Sussman et al., 1990). In a longitudinal study that spanned early childhood through late adolescence (age 18), Shedler and Block (1990) found that at very early ages signs of emotional disturbance distinguished those who were to become heavy marijuana users. At age seven these children were characterized as not getting along with others, unconcerned with fairness, indecisive, untrustworthy, unreliable, unable to admit negative feelings, lacking confidence and self-esteem, and demonstrating physical signs of stress. At age 11, those who were to become heavy marijuana users were distinguished by characteristics such as emotional instability, inattentiveness, inability to concentrate, lack of involvement in activities, and stubbornness.

Theories concerned with the relationship of substance use and mental disorders have emphasized the interaction between the symptoms of the mental disorder and the mood-altering characteristics of specific substances. It has been proposed that substance use is an attempt to self-medicate for difficult feeling states such as depression and anxiety (Khantzian, 1985). Within this view, the mood-altering characteristics of the substance of choice would likely correspond to the symptoms associated with particular mental disorders. A related view is that substance use is the result of the disinhibiting influence of psychological factors such as impulsivity or impaired judgment. Adolescence is a difficult developmental period that may precipitate the onset of emotional problems or substance use (Erikson, 1950). Self-medication (Weiss & Mirin, 1987) and using substances to forget unpleasant experiences, or to fulfill a need state that cannot be otherwise gratified (Mainous et al., 1996), have been identified as motivations for adolescent substance use. A number of studies have found that positive expectancies on the part of the potential user (e.g., stress reduction) predict alcohol use (Rather et al., 1992; Sher et al., 1991) or more general substance use (Stacy, Newcomb & Bentler, 1991). Parents who use substances as a coping behavior may foster such expectations.

Parental influence has been found to be of critical importance in studies of risk factors for adolescent substance use. Among significant determinants of adolescent substance use are parental attitudes and behavior, role modeling, parental behavioral management, and the quality and consistency of family communication (Kandel, 1982; Kandel et al., 1978; Donovan & Jessor, 1978). Parental supervision and the perception of parental concern have been found to be associated with reduced likelihood of substance use (Fletcher et al., 1995; Richardson et al., 1989). Peers also influence risk of initiation and progression of substance use (Kandel et al., 1978; Jessor & Jessor, 1978; White, 1987). Personality factors that influence peer group resistance, such as self-efficacy, have been found to be protective against substance use (Stacy et al., 1992). It may be that weak parent-adolescent bonds augment vulnerability to peer pressure. Impaired parent-adolescent attachments may reflect difficulties that date to childhood and escalate to crisis proportions with the developmental challenges and associated turmoil of the adolescent life stage.

Family structure is also associated with differential risks of substance use. Adolescents living with both biological parents are less likely to use alcohol, cigarettes, marijuana, or other illicit substances; family structures consisting of the biological father and the absence of the biological mother are associated with greatest risk (Johnson et al., 1996d). However, the causal mechanisms underlying the risk relation between family structure and substance use are only partially understood (Flewelling & Bauman, 1990; Hoffman, 1993, 1995; Johnson, et al., 1995; Johnson & Kaplan, 1990; Needle, Su & Doherty, 1990; Newcomb & Bentler, 1986, 1987, 1988; Kandel et al., 1986; Yamaguchi & Kandel, 1985; Kandel, 1984).

Familial influence may involve various risk factors such as genetic influences, parental psychopathology, parental substance use, and the availability of substances. Family background may involve indirect adverse effects such as poor parental support, lack of care and nurturance during childhood, physical or sexual abuse, poor supervision of peer relations, weak coping resources, high stress within the home, and low SES. Determining the relative contribution of risk factors is difficult. Risk factors vary for different disorders and substances. Demographic characteristics such as age and gender influence the expression of various adolescent disorders, as well as the expression of early risk factors for emotional problems (Achenbach et al., 1989; Angold et al., 1991; Angold & Costello, 1993; Block & Gjerde, 1990; Loeber et al., 1992; Werry, 1986; Loeber, 1989; Klein & Mannuzza, 1988; Marks, 1987; Prior et al., 1993). Kandel (1982) has proposed that risk factors are of differential relevance during different phases of progression. Parental and peer influences may be particularly critical in early phases of involvement with substances; in later phases, the influence of peers may increase. Factors such as parental use, troubled family relationships, and emotional or behavioral problems may be most predictive of escalation to more serious substance use.


3 "Psychosocial problems" is the sum of the responses to all items on Achenbach's Youth Self-Report.
Previous PageTop Of PageTable Of ContentsList Of TablesNext Page

This page was last updated on August 05, 2008.