Factors Influencing Service Use Among Homeless Youths With Co-Occurring Disorders

Up to 89% of street-living and marginally housed homeless youths in the United States have a psychiatric disorder, a rate four times higher than among housed youths, and 67% have more than one disorder (1). Co-occurring substance use and mental disorders present a significant burden among homeless youths: about 60% meet criteria for a lifetime substance use disorder, and of those, 93% meet criteria for at least one other mental disorder (2). Generally, youths with co-occurring disorders are not adequately served; they have lower satisfaction with care than those with mental disorders or substance use disorders alone, and young people in particular tend to have the highest perceived unmet need (3). Studies of service use by homeless youths are limited; however, they are thought to be disinclined to utilize services and face challenges accessing services above and beyond challenges faced by the general population (4). Barriers to and facilitators of care among homeless youths with co-occurring disorders are not well understood.

Among substance-using homeless youths, use of addictions services has been positively correlated with illness factors such as a history of mental illness and more severe drug use and negatively correlated with service factors such as poor accessibility (for example, a long waiting list) (5). Facilitators of service use identified in interviews of substance-using homeless youths include the availability of youth-specific agencies, outreach services, social supports, peer mentoring (6), and perceived empathic care (7), whereas barriers include embarrassment, concerns about the service, and a negative perception of providers (6). Among homeless youths with mental health problems, the therapeutic relationship has been found to enhance engagement (8).

Although some studies have elucidated determinants of service use among substance-using homeless youths, very few have examined service use among youths with co-occurring disorders. Given the significant burden of co-occurring disorders among homeless youths, their underutilization of services, and the knowledge gap in regard to factors influencing their service use, we sought to increase understanding of facilitators and barriers among homeless youths with co-occurring disorders. We report findings from a series of focus groups designed to identify factors influencing service use among this high-risk group of youths.

Methods

This study was part of a larger project, Evidence on Tap, for the Ontario Ministry of Health and Long-Term Care, that examined services for youths age 16 to 24 with co-occurring disorders. The focus group format was chosen to allow observation of group interactions and consensus building, generate rich discussions, and normalize talking about mental health problems. Focus groups were conducted across Ontario with youths with co-occurring disorders, their families, and service providers. This subproject focused specifically on homeless youths with co-occurring disorders and interviewed youths only. Inner-city agencies that offered mental health services to homeless youths with co-occurring disorders in Toronto, the largest city in Canada, were invited to participate. Agencies were selected to reflect a breadth of service users: a shelter with short- and long-term facilities where substances were banned on site, an emergency shelter that used a harm-reduction approach, and a drop-in center and health clinic that served many street-living youths who avoid shelters. At each agency, mental health and addictions services were available on site, including consulting psychiatrists. All four agencies were invited and agreed to participate.

At each agency, mental health workers were asked to identify clients who met criteria for co-occurring disorders as assessed by the psychiatrist. The workers asked 26 consecutively presenting eligible clients whom they judged clinically stable enough to participate, and all agreed to participate; however, three did not attend on the day of the focus group (two from one group and one from another). Twenty-three youths age 18 to 26 participated in the groups, a sample slightly older than the Evidence on Tap target population and reflecting the age demographic served by the four agencies. The study was approved by the Research Ethics Board at the Centre for Addiction and Mental Health in Toronto.

Data were collected from April 2010 to April 2011. Four focus groups, each with five to seven youths, were conducted at the recruitment sites; each participant attended one group. Groups were led by two authors (NK and either AC, HW, DI) and followed a semistructured group interview format with follow-up questions. The standardized questions were based on input from clinicians who work with homeless youths with co-occurring disorders. [A list of the questions is available in an online data supplement.] Participants were asked about how they got connected with services, facilitators to initiate and maintain service use, and barriers for people with co-occurring disorders. They were not asked about personal substance use or service use. Participants provided basic demographic information.

Focus groups lasted for 60 minutes. Participants provided written informed consent and received a small cash honorarium and snacks. The sessions were digitally audio-recorded and transcribed verbatim. Transcriptions were checked for accuracy, and identifying information was removed. Transcripts were examined iteratively with thematic content analysis. Two authors (NK and DI) independently read and coded the transcripts, identifying repeated ideas and phrases relevant to factors influencing service use among homeless youths with co-occurring disorders. They then discussed and labeled the ideas and phrases as themes, grouping them within broad categories (individual, program, and systemic factors). They then reread and independently coded the transcripts by categories and themes. Finally, they discussed their coding and resolved differences by consensus.

Results

Most of the 23 participants were male (N=20, 87%), and the mean±SD age was 22.2±2.1 years. Of the 20 who answered questions about education, eight (40%) had not completed high school, five (25%) had completed only high school, and seven (35%) had attended some college or university. Fourteen (61%) identified themselves as white, followed by Asian (N=4, 17%), African American (N=2, 9%), Hispanic (N=2, 9%), and Aboriginal–North American Indian (N=1, 4%).

Factors perceived as influencing initiation and ongoing use of services for co-occurring disorders were grouped into three categories: individual (motivation, support, and therapeutic relationship), program (flexibility and comprehensiveness of services and availability of harm reduction services), and systemic (stigma and accessibility) factors. [Themes are described with accompanying quotations in a table in the online supplement.]

Many youths cited personal motivation and readiness for change as mandatory for engaging in services: “It’s up to you to go and make use of the help.” Youths were more likely to take action in accessing services when “you're the one deciding to go,” as opposed to being coerced. Participants also cited the importance of having a support network of family and peers. Supportive figures made the youths feel cared for, encouraged them to seek treatment, and advocated for them. As one participant said, “A third time [I got clean], I had a lot of help from friends and family. Like, if it wasn’t for them, I don’t know what I would have done.” Another said, “You have to get references from people you know who’ve used the service to know which ones are more likely to help.”

Youths cited the relationship with service providers as a key influence. One participant remarked, “She’s cool, it’s confidential, and I get an hour to just talk to someone who has actually gone through training . . . and might just have some options for you. Participants also noted the importance of getting to know their practitioners over time and the value of persistence. “[My caseworker] . . . went above and beyond . . . calling me when I wouldn’t return her call. . . . It took a long time for me to come around. And she was really persistent with me . . . helped me get through a lot of stuff.” Providers who were easy to relate to—particularly peer support workers—encouraged use of services.

Youths generally endorsed services that catered to their individual needs. Service use was facilitated by program flexibility, including the ability of staff to see youths who lacked valid identification. Services that were comprehensive and met basic needs of homeless youths tended to be valued as a first step toward dealing with mental health and addictions issues. “Sometimes I didn’t wanna talk about it, maybe I wanted to, like, solve something else that’s an issue. . . . Just having mental health [problems] and having nowhere to live is hell.”

Furthermore, services that offered recreational activities and vocational services were successful in engaging youths and helped them abstain from substance use. Participants also reported being more likely to attend agencies that successfully integrated services—a “one-stop shop” for services for co-occurring disorders as well as for basic needs and health care.

Participants had mixed opinions about a harm reduction approach. Supporters argued that an abstinence model created too many barriers. As one youth said, “If . . . I’m high and they know it, I would get kicked out, right? But how is that really helping me? . . . If you come home . . . drunk or under the influence, that’s the time you need somebody to take you in the most.” However, some argued that harm reduction services, especially those based in shelters, attracted intoxicated and drug-dealing youths, creating a temptation to use substances. Some participants advocated for separate facilities for those seeking a harm reduction model and those aiming to be abstinent, “geared [to] where you are in your life at that particular moment.”

Participants spoke emotionally about stigma as a barrier. “You feel like a freak. [The psychiatric hospital] is associated with all these people walking around talking to themselves. . . . There’s no place where you still feel kinda normal.” Service providers were a source of stigma, sometimes making the youths feel labeled: “It’s difficult for me to ask for help. . . . I don’t want to tell people that I have mental health issues or that I have a substance abuse problem. . . . They’re gonna think that I’m dirty, they’re gonna think I’m a prostitute, they’re gonna think that I use dirty needles.”

Timely access to resources was a major influence. Youths felt frustrated if they did not meet narrow intake criteria or had to wait months for an appointment. Youths also described not being able to access services for co-occurring disorders through related sectors, such as the criminal justice system. Even when youths with drug charges were released on bail on condition that they enter addictions counseling, they felt that the system failed to connect them with services. Participants reported significant gaps between services they had received, notably between acute withdrawal management (“detox”) and residential treatment. One participant described the consequences: “When I wanted to go to rehab, I basically had to go to detox, then wait until there was an opening at the rehab centre. . . . I relapsed in the time I had to wait. . . . I had a friend that basically relapsed, got so depressed he [hanged] himself. If only there was a place in the rehab.” The same youth summarized the impact of these barriers: “Don’t make it hard to look for a better way out.”

Discussion

These results begin to elucidate factors influencing service use among homeless youths with co-occurring disorders. Many factors are consistent with the literature on the general youth population with co-occurring disorders and on homeless youths with a range of mental health problems.

A review of co-occurring disorders among youths underscored the role of motivation in seeking treatment and the potential of motivational interviewing (9). This strategy demonstrated mixed results in a homeless substance-using youth population (10), but it requires further study. The therapeutic relationship has been identified as an important factor influencing help seeking (6,8). Its importance to homeless youths may reflect the lack of adult support structures in their lives, an idea explored in the qualitative literature (7). Our participants also cited the importance of peers in role modeling help-seeking behavior and suggesting resources. Although peer-led education strategies are effective among homeless youths (10), surprisingly little has been written about the importance of informal peer support in facilitating service use in this population. Provider and peer support may also offset stigma, a well-documented barrier for youths with mental health problems and specifically for homeless youths (8). However, peer-based interventions are susceptible to iatrogenic peer-group influences thought to increase substance use and other problem behavior and must be planned carefully (11).

Participants’ concerns that harm reduction services exposed them to substance-using peers may also reflect this peer-contagion effect. This finding suggests the need to diversify local services, with separate facilities (especially shelters) offering harm reduction and abstinence-based services. Programs that address youths’ basic needs and vocational and life skills have demonstrated important mental health outcomes (10), and case management has been linked to increased mental health service use among homeless youths (12). The fragmented nature of services described by participants is widely noted in literature on homeless youths (8) and youths with co-occurring disorders, underlining the importance of integrated and comprehensive services (9). Ontario has adopted the philosophy that “every door is the right door” (13). Our participants suggested that system entry could be better facilitated.

Accessibility is a well-described service barrier for homeless youths. Problems with accessibility both within services and between treatment stages can lead to further fragmentation and worse continuity of care. Such barriers can have serious implications for youths with co-occurring disorders, who typically benefit from longer-term treatment (14). Some barriers may be unique to homeless youths, underlining the need for a flexible and comprehensive approach.

This small, exploratory study had several limitations. Participants were a convenience sample recruited from local agencies, and youths were well enough to communicate in a free and goal-directed manner. Thus our sample was likely better served and perhaps less ill than typical homeless youths with co-occurring disorders. Also, we did not use an operationalized definition of “homeless youths” or diagnostic tools; rather, we relied on providers’ knowledge of participants. The gender distribution varied in each group, but most participants were male. A Toronto survey of homeless adults found a gender breakdown of 72.7% male and 26.4% female (the remainder were transgender or transsexual youths) (15). Although we tried to focus the discussion on services for co-occurring disorders, there was inevitably some overlap with general services. Thus our conclusions cannot be applied to any one specific diagnosis, intervention, or service model.

Conclusions

Homeless youths with co-occurring disorders identified factors influencing service use that are common to other populations of homeless youths but also highlighted factors that may be more specific to their diagnostic group. These included the role of peers, the importance of integrated and comprehensive services, and the need to have both harm reduction and abstinence-based services. Findings suggest the need for more research on interventions such as motivational interviewing, peer- and group-based interventions, and integrated services. Randomized controlled trials are needed of interventions to engage homeless youths with co-occurring disorders in treatment.