Relationship Between Service Utilization and Runaway Youths’ Alcohol and Other Drug Use

Abstract

Runaway youths represent a vulnerable, high-risk population that has received little societal attention. Studies show that youths at greatest need are the least likely to access mental health, medical, and other social services. This study evaluated service utilization, including medical visits, psychological, alcohol and drug counseling and 12-step attendance, and substance use. Runaway youths (n = 51) between the ages of 12 and 17 with a diagnosis of alcohol abuse or dependence were recruited from two southwestern runaway shelters. Results showed that psychological counseling and medical visits were associated with less alcohol use while 12-step attendance was associated with more alcohol use at pretreatment. Service utilization was not associated with illicit drug use. The findings suggest that alcohol and other drug use are affected differently by service utilization in this sample of youths. More research is required to evaluate and develop treatments which can effectively intervene with this high-risk group.

Runaway and homeless youth constitute a vulnerable population that faces a multitude of problems. Studies document high rates of alcohol consumption and illicit drug use, physical and sexual abuse, depression, criminality, family conflict, and frequent prostitution (Johnson, Aschkenasy, Herbers, & Gillenwater 1996; Zimet et al., 1995). The current level of drug involvement among runaway youths is at least double that of school youths (Forst & Crim, 1994) and is estimated to range from 70% to 95% (Booth & Zhang, 1996; Rotheram-Borus et al., 1989; Shaffer & Caton, 1984; Yates, MacKenzie, Pennbridge, & Cohen, 1988). In addition, several studies found between 30% and 40% of runaways had used intravenous (IV) drugs (Anderson, Freese, & Pennbridge, 1994; Pennbridge, Freese, & MacKenzie, 1992; Kral, Molnar, Booth, & Watters, 1997; Yates et al., 1988), underscoring the severity of these youths’ drug involvement and high-risk behaviors.

Limited evidence suggests that rates of alcohol abuse among homeless youth are similar to rates reported among homeless adults (Robertson, 1989). Smart and Adlaf (1991) studied 145 street youth and found that almost 50% reported current alcohol problems at a clinically significant level and 9% drank alcohol daily. Similarly, two additional studies found that nearly 50% of their respective sample of runaway youths met DSM-III criteria for a diagnosis of alcohol abuse which is six to eight times higher than that of non-runaway peers (Robertson, 1989; Warheit & Biafora, 1991). In a sample of 300 runaway shelter youths, Koopman, Rosario and Rotheram-Borus (1994) found that about half of runaways who drink alcohol report having three or more drinks on each occasion when drinking. Runaway/homeless youths use alcohol at a younger age and experience greater impaired social functioning owing to alcohol use compared to non-homeless adolescents (Kipke, Montgomery, & MacKenzie, 1993). Yet, given these youths’ high use and need for intervention, one study determined that only 15% had ever received treatment for alcohol problems (Robertson, 1989).

Although health and psychological problems of runaway and homeless teens are pressing national health concerns, runaway and homeless youth are an under-studied and under-served population (Rotheram-Borus, Feldman, Rosario, & Dunne, 1994). The few studies to date have focused on youths’ reasons for running and on categorizing types of runaways and their behaviors. Such research is helpful for identifying areas in which runaway youths are especially vulnerable and help guide interventions that target those areas. However, more research identifying effective treatment components and subsequent outcome is needed. Other than an HIV intervention with shelter youths (Rotheram-Borus, 1991) and a case management intervention with street youths (Cauce et al., 1994), no published reports were found that identify potent treatment interventions with homeless youths. Shelters, and shelter-based programs, have been the main intervention noted in the literature, and virtually no data exist that specify the treatment received by youths while in these shelters (Teare, Peterson, Furst, Authier, Baker, & Daly, 1994). Moreover, shelters for runaways are overcrowded, and many shelters are not equipped to treat youths for substance use and family problems beyond crisis intervention.

Even with the clear need for intervention, several studies have documented very low service utilization by these youths (De Rosa, Montgomery, Kipke, Iverson, Ma, & Unger, 1999; Kipke, Simon, Montgomery, Unger, & Iversen, 1997; Robertson, 1989). Only 30% of homeless youth access shelter services, indicating that youths most in need of services do not receive them (U.S. House of Representatives, 1992). In addition, only 9% of homeless and runaway youths surveyed in one study had ever accessed mental health services (De Rosa et al., 1999) and only 10–15% had ever received treatment for alcohol or drug problems (De Rosa et al., 1999; Robertson, 1989). As Marshall and Bhugra (1996) note, most homeless individuals do not receive help from any service agency, and their distrust of statutory health services and the inflexibility of these services are contributing factors to their continuing poor physical and mental conditions.

The general consensus is that runaway youths are difficult to engage and maintain in therapy (Morrissette, 1992; Slesnick, Meyers, Meade, & Segelken, 2000; Smart & Ogborne, 1994) and this is a population perceived as “difficult to work with” (Kufeldt & Nimmo, 1987). Indeed, numerous barriers impede successful intervention with these youths including the youths’: (1) lack of insurance or knowledge of how to use insurance, (2) lack of transportation, (3) few available services, (4) confidentiality issues, and finally, (5) unrealistic stereotypes of these youths held by care providers (De Rosa et al., 1999; Stango, 1995). Furthermore, given the methodological challenges in locating, treating, and retaining homeless youths in treatment, it is not entirely surprising that there are few treatment outcome studies that focus on these youths. However, this is a population sorely in need of societal attention. Evaluations of both service utilization and outcome will assist in the development of innovative methods for intervention. Such evaluations will also assist in the identification of youths’ service needs which, in turn, may guide providers in removing barriers and resistance to treatment.

This article will examine the relationship between service utilization (three months prior to the interview) and substance use in a sample of shelter-bound runaway youths. These youths may include those who have left home voluntarily (runaways), were thrown out of the home (throwaways), or were removed from the home by the state (system kids). It was expected that service utilization would predict less substance use by youths at the pretreatment assessment.

METHOD

Participants

Runaway youths were recruited through two runaway/homeless shelters in a southwestern city as part of a larger study examining family therapy outcome with substance abusing runaway adolescents. Hence, all youths in this sample agreed to the possibility of family treatment. The sample included 51 youths aged 12–17 (M = 15.22, SD =1.35) of which 23 were male and 28 female. Ethnicities were reported as Anglo (N = 17), Hispanic (N = 24), Native American (N = 6) and African-American (N= 3) and Other (N= 1). Youths reported 2.73 lifetime runaway episodes (SD = 2.69), and 56% of youths reported that they had accessed medical services, had attended 12-step meetings and/or attended psychological, drug, or alcohol counseling sessions in the three months prior to the interview. To be eligible for participation, adolescents had to satisfy DSM IV diagnostic criteria for alcohol abuse or dependence, and must have had the legal option of returning to a home situation. The shelter referred substance abusing teens, with the option of returning home, to project research assistants. Only two eligible youths referred to the program refused to participate (95% engagement). Table 1 presents selected characteristics of the study sample.

TABLE 1.

Characteristics of the Adolescent Sample: 90 Days Before Study Recruitment

	Mean	SD	Comments
Highest Grade Completed	8.55	1.39	Range 5–11
Annual Income	$27,370	$27,907	Range 0–100,000
Percent Days Alcohol Use	21.04	14.66	Range 3–66%
Percent Days Illicit Drug Use	27.03	26.94	Range 0–97%
12-step Meetings Attended	.52	1.39	12% report attending
Days Medical Care	1.60	6.93	39% report care
Days Alcohol Counseling	.19		2% report care
Days Emotional Counseling	3.33	10.39	29% report care
Days Drug Counseling	2.06	7.72	1 3% report care

Procedure

A project research assistant approached youth identified by shelter staff as potentially eligible for the project. Youths were engaged and screened for participation utilizing the eligibility criteria as a guide. Those passing the screening criteria were then scheduled for an assessment at the shelter within 24 hours. Assessments began with a review of the nature and conditions of the study, a formal review of the elements of informed consent, and a signing of the consent statement. The interviewer then proceeded to administer the Diagnostic Interview Schedule for Children (DISC; Shaffer, 1992) sections on alcohol, marijuana, and other substances. Those not passing the diagnostic screening continued with treatment as usual through the shelter. Those meeting the criteria for participation in the study continued with the assessment battery. They were told that the assessment would take up to 3 hours and that they would receive $25 at the end of participation. Once a youth was found eligible and agreed to participate in the project, the youth’s primary caretaker was contacted and consent was obtained.

Materials

The assessment included interviews and self-report questionnaires. Assistance in completing forms was provided to youth as needed, and participants were given the option of completing the assessment in one session or in two shorter sessions on separate days. Only the assessment instruments utilized in the current study are described.

The examiner administered a demographic questionnaire assessing a set of core variables used to characterize and compare samples included in this study. These demographic items include age, gender, self-identified ethnicity, parent and sibling information, education, and income.

The Form 90-I, developed for NIAAA funded Project Match (Miller & Delboca, 1994), was used as the measure of quantity-frequency of adolescent substance use in the current study. The Form 90-I combines the timeline follow-back method (Sobell & Sobell, 1992) and grid averageing (Miller & Marlatt, 1984) and has shown excellent test-retest reliability for indices of drug use in major categories (Tonigan, Miller, & Brown, 1997) with kappa’s for different drug classes ranging from.74 to 1.0. This is a structured interview which yields a daily reconstruction of all drug classes and provides a percentage for days of all alcohol use (in the prior 90 days), the primary dependent measure. In addition, the Form 90-I assesses service utilization (psychological and medical, inpatient and outpatient), employment, school attendance, and primary residence.

RESULTS

Multiple regression analyses (MRA) were conducted to assess the joint relationship between five help-seeking activities and substance use in the 90 days before study recruitment. The five help seeking variables were: (1) days medical care, (2) days 12-step attendance, (3) days emotional counseling, (4) days alcohol counseling, and (5) days drug counseling. The first MRA used the percentage of alcohol use days (of 90) as the dependent measure and the second MRA used the percentage of days (of 90) that illicit drugs were used. Contrary to our expectations, adolescent alcohol and illicit drug use were not significantly related (r=.14, p <.15), leading us to examine the measures separately. In both MRA, the five help-seeking activities were entered into the full model, and a backwards elimination technique was used to derive the most parsimonious model. Regarding alcohol use first, the five predictors accounted for 27% of the variance in reported alcohol use in the 90 days before study recruitment, F(5, 46) = 3.51, p <.009. Elimination of non-significant help-seeking variables led to a three predictor solution wherein lower alcohol use was found with increased medical care (p <.02) and emotional counseling (p <.003), but higher alcohol consumption was reported with increased 12-step attendance (p <.001). These three help-seeking activities accounted for 26% of the variance in alcohol consumption indicating that days of formal counseling for alcohol or drug problems were largely unrelated (but present) to alcohol use. Regarding illicit drug use (excluding tobacco and alcohol), the five help-seeking activities were jointly, as well as separately, unrelated to illicit drug use, F(5, 46) = 1.22, p <.32, and F(1, 50) = 1.52, p <.22.

DISCUSSION

The aim of this study was to examine the relationship between runaway youths’ service utilization and substance use in the three months prior to their arrival at a runaway shelter. Overall, alcohol use was differentially associated with service utilization, although illicit drug use was not. The finding that drug use was not related to service utilization is perplexing. It is possible that youths do not identify their drug use as a problem, as alcohol was identified as their primary drug of abuse. That is, youths did not identify themselves as having drug problems, and hence may be less likely to be influenced by interventions in that area. The ineffectiveness of interventions which focus on drug use was also supported by the finding that drug treatment (treatment with the focus primarily on drug use) was not related to increased or decreased substance use. It should be noted that this finding is not due to a lack of drug use in alcohol abusing and dependent youths. All but three youths reported drug use in addition to alcohol use (with 90% reporting marijuana use). These findings suggest that alcohol and drug use are affected differently by service utilization in this sample of youths.

Attendance at 12-step meetings predicted higher alcohol use while attendance at psychological counseling sessions as well as medical visits predicted less alcohol use. Of the youths that attended 12-step meetings, the average number of meetings attended in the previous three months was 3.38, whereas the average number of sessions for youths attending therapy for emotional or psychological problems was 11.53. The 12-step model requires a long-term lifestyle change with a commitment to abstinence and regular attendance, especially at the beginning of recovery. The number of 12-step meetings attended by these youths would not be sufficient to evoke change. It is possible that youths attending 12-step meetings are characterized by more denial and resistance, which may be associated with more alcohol use. Alternatively, these youths may be unwilling to accept abstinence as a goal. Many youths do not accept that they have a alcohol problem severe enough to warrant abstinence, beleiving that they are too young to be alcoholics.

The differing structures of 12-step and counseling sessions may also play a role in maintaining youths in treatment. Youths are likely to require transportion to meetings by their parents; the level of chaos in the family may preclude the youth’s regular attendance at 12-step meetings which occur daily. Traditional counseling requires standing appointments, and if one is missed, the therapist will likely call and re-engage the youth and family. Hence, the therapist works to track the family and will encourage the parents and youth to follow through. In this way, denial and resistance are addressed. Moreover, as the focus of the psychological/emotional counseling sessions are likely to be on issues that affect youths’ alcohol use, including problems with family, peers, and school, it is not surprising that more therapy sessions predict less alcohol use. Without regular attendance and the introduction of a sponsor, 12-step meetings cannot be as effective as they would otherwise be.

Medical care visits also predicted less alcohol use in youths. Medical appointments are likely to be made by parents and these youths are also likely to be transported to appointments by their parents. It is possible that agreeing to receive medical care indicates a receptiveness to care in general. However, more research will be needed to explore this finding.

Many youths in this sample engaged in help-seeking behaviors which included behavior change interventions and medical services. Although 56% of youths accessed services, which is more than reported by De Rosa et al. (1999), only 2% of this sample of alcohol abusing youths received counseling for alcohol problems. However, all youths in this study accessed the shelter system and so are not representative of street youths in general. That is, shelter-bound youths may be more amenable to accessing services in the community given that they were willing to stay at a runaway shelter. The findings suggest that youths’ service utilization prior to their stay at the shelter was not successful as measured by youths’ removal/flight from their homes, and continued problem alcohol/drug use at the time of the assessment.

Several limitations should be considered when interpreting the results of this study. First, the cross-sectional design precludes conclusions regarding causality. These youths may have accessed more services than they would have otherwise in the three months prior to arriving at the shelter, as that was a high-risk period of time. Second, treatment outcome research with an operationalized intervention and several follow-up assessments is needed to evaluate the effectiveness of treatment for alcohol abusing runaway youths. Third, the results are based on self-report questionnaires and interviews, with no convergence from parents or primary caretakers. Although great care was taken in the current study to address concerns youths had regarding confidentiality and the purpose and nature of the data collection, youths’ self-report of sensitive and illegal behaviors may underestimate or overestimate their true prevalence. Finally, it should be noted that runaway/homeless youth constitute a diverse group. Researchers are noting that shelter youths show less severe substance use and high-risk behaviors than do street youths, and that within each subgroup, those who have been “thrownaway” by their family, engage in greater substance use and high-risk behaviors (Ringwalt, Greene, & Robertson, 1998). Thus, it is likely that more severe patterns of substance use and less service utilization would be found in street youths who did not reach one of the two shelters sampled by this project.

Given these limitations, this article’s findings have several treatment and research implications. Many individuals in this sample of shelter-bound youths accessed services in the community, although these services were relatively unsuccessful in stabilizing the youths and families. This finding suggests that youths and families are not resistant to intervention and will indeed seek help during difficult periods. Empirical evaluation/development of interventions for runaway and homeless youths has received little attention, and more work is needed to tailor interventions to address the unique needs of runaway substance abusing youths. This work is especially important to the therapists who may effectively intervene and prevent future runaway episodes and increased substance use. Moreover, future research should explore this study’s finding that among runaway youths who identify alcohol as their drug of choice, other drug use was not influenced by the five help-seeking interventions. Differing mechanisms for change may be involved in alcohol versus drug use behaviors for alcohol-abusing and dependent runaway youths.