Introduction
Adolescence represents a developmental period which appears to be essential regarding substance use initiation and the development of mental and behavioral disorders due to psychoactive substance use [
1,
2]. There is strong evidence that adolescent substance use can be connected to a number of negative cognitive, psychosocial or mental health consequences [
2‐
5] and that psychiatric disorders are risk factors for the involvement in adolescent substance use [
5‐
9].
In epidemiological studies on alcohol, nicotine and illegal drug use among European youth, prevalence of psychoactive substance use was found to be high [
10,
11]. Adolescents aged 15 to 16 reported using alcohol at a rate of 80%, nicotine at 46% and illicit drugs at 18% at least once [
10]. Prevalence within adolescents in Austria was found to be even higher [
12]. Regarding alcohol, abuse or even dependence could be found among adolescents [
2]. Furthermore, heavy episodic drinking (“binge drinking”), described as the consumption of 5 or more drinks within one drinking event, is the dominant type of alcohol misuse in youngsters [
2]. Nicotine consumption shows the highest rates in terms of regular use. One in five adolescents in Austria smokes daily [
12]. Cannabis was the most frequently used illicit drug [
12‐
14], with prevalence of lifetime use being lower in Austria (16%) than in other European countries (20%) [
10,
12]. Use of other illicit drugs was less common among adolescents with prevalence of 1–2% [
10,
13]. The mean age of initiating alcohol use was 13 years, [
11,
12,
15], that of smoking the first cigarette 14 years [
13]. Illicit drug use usually started at an older age, very early consumption was found among 3% of users [
10].
In an attempt to explain the development of substance-related behavior, previous studies described several psychosocial risk factors predisposing alcohol, nicotine and illegal drug use among the adolescent population. Firstly, chance of substance use and misuse increased with age [
11,
16]. Secondly, physical, sexual and mental abuse as well as dysfunctional families have been identified as being related to substance use in adolescents [
17]. Thirdly, peer influence as another strong determinant was described [
18]. Further research has demonstrated that adolescent substance use increased the risk of school failure [
2] and suicide attempts [
2,
8]. Substance using adolescents were more likely to be involved in serious risk behavior, like being involved in criminal activities, fights or drunk driving [
1,
2].
Several studies have evaluated a noteworthy relationship between substance use and psychopathology. On the one hand, adolescents suffering from mental and behavioral disorders due to psychoactive substance use were more likely to show symptoms of mental disorders [
8]; on the other hand, psychopathology predicted substance use [
19]. Elevated rates of alcohol, nicotine and illicit drug use and misuse were found among adolescents entering psychiatric care compared to their counterparts in the general population. Lifetime use rates of alcohol in adolescent psychiatric population had been reported to reach levels between 45% [
8], 75% [
20] and almost 100% [
21]. Adolescents suffering from mental disorders also had significantly higher lifetime use rates of nicotine with values ranging from 60 to 90% [
20,
22,
23]. Likewise, mentally ill adolescents were significantly more likely to use illicit drugs [
8]. Furthermore, numerous investigations concluded that within samples of mentally ill youth significant differences in substance use rates could be found. Attention deficit/hyperactivity disorder (ADHD), conduct and personality disorder as well as affective disorders have been linked to elevated substance use and misuse [
8,
23,
24]. In contrast, individuals suffering from restrictive eating disorders were significantly less likely to report lifetime prevalence of alcohol, nicotine and drug use compared to other diagnostic groups [
25,
26].
The aim of the current study was to investigate prevalence, pattern and frequency of alcohol, nicotine and illicit drug use in a clinical population of adolescents suffering from mental disorders in Vienna, Austria. Psychosocial correlates of substance use behavior should be targeted with regard to age, history of trauma, family dysfunction or suicidal ideation as well as peer influence. Another purpose of this study was to examine psychosocial consequences of substance use and to determine differences between different diagnostic groups. Based upon previous research, we expected high prevalence of substance use and misuse as well as risky pattern of use among our sample of adolescent psychiatric inpatients exceeding findings from the adolescent general population. In addition, we assumed to find subpopulations with elevated risk, i. e., subjects with history of trauma, dysfunctional families or suicidal ideation, subject reporting peer substance use, as well as older adolescents and adolescents suffering from externalizing or affective disorders. Finally, we speculated to find significant differences between using and non-using respondents regarding psychosocial consequences like school failure or problem behavior.
Methods and statistics
We included patients aged between 11 and 19 years admitted to a child and adolescent inpatient psychiatric unit at the Medical University of Vienna. Data collection was performed between May 2015 and April 2016. Exclusion criteria were inadequate language skills, poor cognitive function, short-term admission due to crisis intervention and lack of consent. For patients who had more than one inpatient admission during the 12-month period of data collection only the first consultation was considered.
Lifetime use rates and initiation, frequency and quantity of alcohol, nicotine and illicit drug use were collected by self-report measures. In addition, sociodemographic, family and school-related data, including peer substance use, family dysfunction, academic achievement and substance use related behavioral problems were queried. Alcohol-related behavior was detected using the CAGE questionnaire (Cutting down, Annoyance by criticism, Guilty feeling, Eye-openers), a 4-item scale proved to have good sensitivity and specificity [
27]. Nicotine consumption was investigated using Fagerström Test of Nicotine Dependence (FTND), a 10-item questionnaire highly reliable in rating physical dependence to nicotine [
23]. Patients’ answers were objectified using alcohol and carbon monoxide breath analyzer tests [
28], %CDT (Carbohydrate-Deficient-Transferrin-Asialotransferrin) as a serological marker to determine recent alcohol consumption [
29] and urine screening for psychoactive substances [
19]. Clinical diagnosis according to ICD10 and several factors known as risk factors including history of trauma (physical/sexual abuse, mobbing, loss of a first degree relative, refugee with traumatic experience) and suicidal ideation were extracted from medical paper files. Written informed consent was obtained by participants and their legal guardians. The study was approved by the Ethics Committee of the Medical University of Vienna (EK 1080/2014).
IBM SPSS Statistics 23.0 software was used for analyzing data; 2 × 2 contingency tables were calculated and, due to a small sample size, Fisher’s exact test was used to examine differences in sociodemographic factors (i. e., history of trauma, family dysfunction, suicidal ideation, peer substance use, school absenteeism, problem behavior) between substance using and non-using youth as well as to examine any effects of different psychiatric disorders. T‑test and Mann–Whitney U test for not normally distributed variables and small sample size were used to compare arithmetic means (mean age, school grade). Regression analyses were performed to investigate the effect of risk factors on substance use including age, family dysfunction, peer substance use or trauma. P values of less than 0.05 were considered significant.
Discussion
This study investigated substance use in 25 minors admitted to an inpatient psychiatric facility. Three quarters of our sample had reported lifetime consumption of alcohol, which is comparable to findings within German psychiatric inpatient populations [
20,
21], but higher than French and Norwegian investigations describing lower prevalence of 45 to 60% [
8,
23]. Also in agreement with previous findings, 20% of our sample used alcohol regularly, whereas our sample was more likely to experience alcohol intoxication (40% vs. 25%) [
20].
Lifetime nicotine consumption was high with 50%. Most other investigations however found even higher rates of up to 90% in mentally ill adolescents [
21‐
23].
Previous studies have indicated that the use of illicit drugs (50–60%) [
20,
21] and especially cannabis use (50%) [
9,
21,
30] are common among mentally ill adolescents. In our sample two thirds had already consumed an illicit drug, with cannabis being most frequently used. Regarding prevalence of the use of other illicit drugs than cannabis, our data correspond with the rate of 10% determined by other investigations [
23,
30]. Moreover, the study sample showed an earlier age of initiation of consumption of any substance investigated, once again indicating high vulnerability to substance use and elevated risk of SUD in this clinical sample [
1,
8].
As previously described [
17,
23], subjects having experienced trauma or family dysfunction were significantly more likely to use alcohol, nicotine or drugs or to exhibit risky patterns of consumption. Also in agreement with findings in previous studies [
18,
31], we identified peer groups as significant determinants. While each non-user denied peer substance use, adolescents affirming risky patterns of use as well as nicotine or drug consumption all had drug-using peers. High peer influence regarding substance use most likely originates from greater availability of legal and illicit drugs and social pressure.
Concerning psychosocial consequences, subjects reporting multiple consumption or meeting SUD criteria had significantly lower school performance and were more often involved in problem behavior like school absenteeism or legal conflicts [
2,
23,
32]. These results are once again in line with previous studies, reporting that mentally ill adolescents consuming in risky pattern were at high risk to develop substance-related problem behavior [
2,
23] and had significantly lower school performance [
2,
32] than nonconsumers.
A history of suicidal behavior was also significantly linked to regular substance use as well as to alcohol and illicit drug use; all patients reporting a history of suicide ideation were alcohol users. Similar results have been described previously [
23,
33].
In investigating the impact of externalizing and affective disorders on consumption patterns, we found that subjects suffering from externalizing disorders showed higher prevalence of alcohol (85%) and drug use (57%) compared to subjects with other diagnosis. Surprisingly, rates of nicotine use did not differ between diagnostic groups which is in contrast with earlier findings [
8,
20,
22]. As previously reported [
34,
35], we also found a significant association of cigarette smoking and affective disorders. More than two thirds of respondents with affective disorders had smoked at least once, this prevalence rate being twice as high as recently described [
8].
While earlier reports indicate that adolescents with AN had a lifetime history of alcohol use of about 20% [
25], alcohol use among our sample was stunningly high with 60%. Subjects suffering from a restrictive eating disorder were however significantly less likely to have tried nicotine or illicit drugs compared to other diagnostic groups. Consistent with previous findings [
26], this diagnostic group was significantly less likely to meet SUD criteria.