Introduction

Disengagement from care is common among clients with serious mental illness and predicts poor outcomes including symptom exacerbations, hospitalization, homelessness, and incarceration.1–4 Studies have identified many predictors of treatment disengagement including demographic factors (e.g., younger age, male gender), clinical factors (e.g., co-occurring substance abuse, poor awareness of symptoms), and service use (e.g., prior history of treatment discontinuation).1,3 Treatment providers are challenged to establish effective therapeutic relationships with clients with serious mental illness because of common complicating factors including impaired awareness of symptoms,5 co-occurring substance abuse,6 housing instability,7 criminal justice involvement,8,9 unemployment,10 and stigma.11,12 Despite and perhaps because of these challenges, effective therapeutic relationships are key predictors of symptom, functional, and service use outcomes in this population.13–15

Previous research has shown that the strength of the therapeutic alliance increases the likelihood of engagement in psychotherapy, medication adherence and improved functional outcomes.16 Clients with mental illness report a desire to share in decision-making, and such participation is needed, in particular, for those who are not content with their treatment.17 Efforts have been made to develop and research tools that may aide in promoting this practice,18,19 but there remains work to be done. Qualitative research offers a useful tool to understand the processes involved when shared-decision making occurs.20

This study examined therapeutic relationships between treatment providers and clients with serious mental illness who had high service needs and a history of recent disengagement from care using a qualitative approach. Prior qualitative studies suggested that clients with serious mental illness are more likely to be in recovery when they develop consistent relationships with treatment providers who support a “normal” rather than “mentally ill” identity.21 Collaborative relationships between clients and treatment providers include a balance of support and encouragement without excessive pressure placed on clients to meet treatment provider expectations,22,23 and treatment providers with successful outcomes place a strong emphasis on a partnership-oriented relationship between treatment provider and client.24 Clients with serious mental illness identify the following priorities for effective therapeutic relationships: sharing common ground, feeling known, the importance of talk, practitioner availability and flexibility, and opportunities for input into treatment planning and choices regarding treatment modalities.25,26

Prior work culminating from the current study outlined reasons for disengagement and characteristics of services that promote engagement among clients with serious mental illness who frequently discontinue services.27 Clients reported that disengagement occurred due to services failing to meet their needs and difficulties with treatment provider relationships and trust, while treatment providers were more likely to view disengagement as part of an ongoing struggle with mental illness, a lack of family or community support and logistical barriers such as transportation. Characteristics of mental health services that support engagement were categorized as: acceptable (caring, strength-based approach); accessible (transportation assistance, flexibility); and available (job training, entitlements). This report expands on findings from prior studies and identifies themes related to effective therapeutic relationships with treatment providers for this population. By drilling down on reports of treatment experiences we sought to determine whether common notions regarding therapeutic relationships would apply to these high-need individuals, and to also identify new directions for potential future research.

Methods

The data presented were collected as part of a larger mixed methods study that investigated service use experiences of clients with serious mental illness who had episodes of disengagement from mental health treatment.27 The study included clients identified in the New York City Mental Health Care Monitoring Initiative, a project jointly sponsored by the NYS Office of Mental Health and the NYC Department of Health and Mental Hygiene that operated from 2009–2011 as an effort to improve service coordination for adult NYC residents with serious mental illness. The project involved reviews of Medicaid claims and other state administrative data for high-need populations to identify clients who may not have been receiving needed services, or who used excessive amounts of acute behavioral health services.28,29 Prior reports have summarized the project’s findings regarding numbers of clients disengaged, reasons for disengagement, availability of treatment provider outreach, and re-engagement rates.27,30,31

Client Recruitment

Recruitment and interviews for this study began in the fall of 2010 and concluded one year later. Potentially eligible clients were identified using data from the NYC Care Monitoring Initiative.27 A stratified purposive sampling approach was used to identify high risk clients who met the following criteria: age 18–65; and had a history of criminal justice involvement, mandated outpatient mental health services, or were young adults (between 18 and 24 years of age). These populations are among the most vulnerable and at high risk for disengagement and poor mental health/functional outcomes.32,33

Staff from the NYC Mental Health Care Monitoring Initiative provided the research team with automated monthly reports listing clients who met the following inclusion/exclusion stated above. In addition to the monthly report, care monitors were also encouraged to identify specific clients who met the eligibility criteria based upon ongoing case reviews with treating treatment providers.

Research staff contacted treatment providers who previously served potential participants as confirmed by project care monitors and/or the monthly lists. The treatment providers were asked to present the project to the potential participant, who if interested could then either call the research team or sign a permission to contact form that allowed a member of the research team to contact the client directly. Eligible clients were told that the research study involved audio-recorded interviews focusing on their experiences with the mental health system and their reasons for choosing to seek or discontinue care. Eligible clients were offered a $50 cash card for completing the interviews. The project was reviewed and approved by the New York State Psychiatric Institute’s IRB.

A research team member scheduled interviews with eligible clients at a setting convenient to the client. Interviews were typically conducted in a diner or café in the clients’ communities. Some clients were residing in inpatient settings at the time of the interview; these interviews were conducted in a quiet and private area of the hospital. Interviews were also conducted in outpatient mental health facilities when treatment providers offered a space, or at offices where the research team was based, if eligible clients preferred a more private setting.

Two interviewers from the research team attended each interview. Before the interview began, interviewers obtained informed consent from the client. Clients were given a description of confidentiality procedures and assured that all identifying information would be removed from reports of their interviews. They were also told that they could stop participation at any time and that their health care would not be affected by their participation. All clients were informed that any indication of harm to self or other (e.g., suicidal ideation) might be reported to ensure the safety of the participant or others. Once clients consented to participate, interviewers conducted a short-answer survey to collect descriptive data prior to the semi-structured qualitative interview. The qualitative interview portion was audio-recorded. Field notes documented any additional relevant information. All recorded files were transcribed verbatim for analysis.

Treatment Provider Recruitment

Recruitment for treatment provider interviews began in the winter of 2011. Permission to interview treatment providers was first obtained from agency directors or high-level supervisors. When permitted to do so, research staff contacted treatment providers who had been identified having clinical experience with clients identified in the Care Monitoring Initiative. Researchers interviewed no more than two treatment providers from the same treatment program and aimed to reach treatment providers with various types of clinical background including case managers, social work treatment providers, therapists and clinical supervisors.

Potential participants were contacted by research staff and informed that involvement would include audio-recorded interviews focusing on their experiences with high-needs clients such as those identified in the Care Monitoring Initiative. Individuals were offered a $25 cash card for completing the interviews, and these were given to treatment providers or offered as a contribution to the agency according to agency policy.

With the exception of one interview conducted at an agency, all treatment provider interviews were conducted over the phone. Two interviewers from the research team participated in each interview. Prior to interviews, consent was obtained from participants, usually returned by fax to a member of the research team. Qualitative interviews were audio-recorded and transcribed verbatim for analysis.

Participants

Over a 12-month recruitment period, 2361 eligible clients were identified, 1647 of these clients had an identifiable treatment provider, researchers attempted to contact treatment providers for 1431 clients, for 983 clients contact with treatment provider was successful, treatment providers for 971 clients agreed to attempt to make contact with the identified client, 126 clients were successfully contacted, 49 clients refused the offer to participate, 77 clients granted permission for the research team to contact them, the research team contacted and scheduled interviews with 60 clients, and 57 clients provided informed consent and completed interviews. Table 1 lists demographics for client participants.

Table 1 Demographic and treatment characteristics of clients with serious mental illness who completed qualitative interviews
Full size table

The treatment provider interviews took place over a 7-month period, overlapping with client interviews. Twenty-one agencies were contacted and requested to allow treatment providers to participate in the study, 15 agencies gave permission to contact treatment providers and 25 treatment providers from 14 agencies provided consent and completed interviews. The sample of treatment providers participating included 21 females (84%) and 4 males (16%). The sample included 3 bachelor’s level case managers (12%) and 22 individuals with master’s or doctorate training (88%).

Interviews and Analysis

The research team met weekly for three months to develop the interview guide. The client interview guide included questions about starting, engaging in, and disengaging from mental health services; how relationships with therapists, doctors, or other treatment staff helped (or interfered with) treatment, best (and worst) mental health treatment experiences, and experiences related to mental health in the criminal justice system. This paper reports findings on treatment provider-client relationships, which included questions such as: Can you tell me how relationships with therapists, doctors, or other treatment staff might have helped your treatment/made treatment less helpful? and Does your treatment provider(s) involve you in making treatment decisions? If so, how? Other questions that often led to a discussion of treatment provider-client relationships include: What was the best mental health treatment experience you ever had? How has your current treatment been different from previous experiences?

The treatment provider interview focused on access to treatment, engagement, disengagement, and outreach and reengagement. Treatment providers were asked to describe their relationships with clients in each phase of engagement in services, to define what they considered a positive working relationship, and to report how they approached goal setting.

The interview guide was modified based on feedback from practice interviews among study team members and discussion from weekly training sessions. The team continued to meet weekly after the interview phase began to review interviews and debrief as needed. Additional probes were added to the interview guide based upon the research team’s consensus regarding areas where more specificity was desired.

A modified grounded theory approach34 was utilized to understand clients’ reasons for choosing to discontinue and resume mental health services. Interview transcripts were analyzed using the Atlas.ti 6.2 software system, which aids coding, organization, and retrieval of text for qualitative analysis.35 Themes were developed and refined through an iterative process. Data was examined and used to generate open codes to form hypotheses, which were then re-compared to the data and codes. This analysis led to modification and focus of interview questions and further analysis. During each stage of the analysis researchers recorded thoughts, ideas, and hypotheses about the process. This constant comparative analysis helped further refine the themes. Open codes detailed primary concepts, whereas axial codes related categories to each other. Three researchers coded the interviews and consensus was sought to ensure consistent and reliable data. Reliability across researchers was 83% for both client and treatment provider interviews based on blinded double coding of a 10% sample of text.

Results

Qualitative interview data analyses identified 3 broad themes related to the development of meaningful therapeutic relationships for this population: (1) Being present—availability, tuning in, and trust are necessary for a meaningful client-treatment provider relationship; (2) Clients and treatment providers often do not share the same treatment goals; and (3) Approaches and interventions that have the greatest impact.

Theme 1: Being present—availability, tuning in, and trust are necessary for a meaningful client-treatment provider relationship

The most common theme identified by clients related to the time and availability of treatment providers (45%, n = 25). One client expressed awareness of the demands on treatment providers’ time but felt treatment providers must make an effort to give more focus to individual clients: “… you have to treat everybody like there’s only one patient. Because if you treat people like there are lots of patients, it’s going to be difficult for you. It's going to be difficult for both the patient and the therapist.” Another client felt the negative result of the lack of availability: “…I guess when [treatment providers are] very busy, they cannot help me.”

Clients described a sense of security and value in having ready access to treatment providers. For example, one client with a history of self-harm explained: “I did want to get help, and I want to be able to call somebody that I know I can call and speak to someone.” Another client explained how she was not interested in treatment with a treatment provider who seemed to be too busy: “I just wouldn’t be interested in no one looking at their watch. I’d be like okay … you started my session late because of an extra session you were in, then you wanted to short me 15 minutes, and then after you short me 15 minutes, you still want to make it seem like as though I’m putting some stress on you.”

Both clients (13% n = 7) and treatment providers (44%, n = 11) discussed the importance of being “in tune” with one another. Not only is it important for treatment providers to tune into clients, it is also important to recognize that clients are tuned into their treatment providers and report being keenly aware of treatment provider attitudes and receptiveness. Perceptions of subtle cues from treatment providers strongly impacted clients’ attitudes and experiences. One client explains her expectations of her treatment provider: “I want to feel like you want to know what's going on. You know? Like everything in the body language counts, the eye contact counts. …You know, you could tell. You could feel it, that vibe. You could feel the vibe if they're interested or not, or whether…they're just bullshit or whatever.” In another instance, a client’s perception of the treatment provider’s lack of interest quickly led to a decision to discontinue care: “Come on, I really don't want to sit here and tell you about my problems because you want to go get on this highway right now and go home. If it's mandatory, I would go, but I would just shut down and not really say what needed to be said.”

Treatment providers also gave examples of the importance of tuning in. One treatment provider explained that clients “know what to say and what not to say, so in that you have to keep reading between those lines and looking at those lines and trust [your assessment].”

Both clients (25%, n = 14) and treatment providers (56%, n = 14) described trust and honesty as key aspects of successful relationships. Many clients described a difficulty in trusting others in general and especially opening up to treatment providers, despite reassurances that treatment providers would maintain confidentiality. One client explained:

If I get the first inkling that you aren’t trustworthy, I don't deal with you, and maybe that’s what’s kept me out of treatment, dealing with people like that. The first time I feel, I run to the hills. … A lot of times people with mental illnesses, they tend to hold back, not sure how much they can let go, how much can they really tell? The worst thing you can do with a person that has mental illness is throwing stuff back at them.

This touches on the importance of trust as well as a tendency to towards not fully opening up to treatment providers when there is a sense that the treatment provider is not an ally. Another client spoke to the importance of confidentiality and awareness of exceptions that might lead to a break of confidentiality and the consequences that would ensue: “That’s why I said honesty is very important,[my current treatment providers] let me know right off the bat, they say listen, if you say this, this is going to happen, if you say that, that's going to happen, so they tell me to be very careful what I'm saying and think about what I’m going to say, give myself time to process what I’m getting ready to say.”

Theme 2: Clients and treatment providers often do not share the same treatment goals

The goal for treatment most commonly reported by clients was to find adequate housing (41%, n = 23, identified this as a primary goal). The next most commonly reported goals were to find a job (38%, n = 21) and to work towards educational goals such as pursuing a GED and enrolling in a university or vocational school (25%, n = 14). Clients also reported goals related to treatment including reducing mental health symptoms (25%, n = 14) and taking medication as recommended by their doctor (18%, n = 10), but the majority of clients did not identify as being ill and placed more emphasis on goals related to tangible services.

Treatment providers (32%, n = 9) confirmed that clients often entered treatment seeking stable housing as a primary goal—one case manager reported that housing was “normally number one” in terms of client priorities. Only 16% (n = 4) of treatment providers, however, reported that they worked on housing-related issues with clients. Many treatment providers identified denial as an area that needed to be addressed and worked through (68%, n = 17): “I think some people don’t want to [receive treatment]. Some people don’t think they’re sick.” Another treatment provider noted: “There are many people in denial that say ‘I don’t need treatment.’” Clients did not emphasize acceptance as an important part of treatment or their relationship with treatment providers.

Clients reported that rifts in the client-treatment provider relationship were more likely when the treatment provider did not acknowledge their goals: “Like if I told them that I need help with certain things, they'll just deny it [and say I could do it on my own]…… so I really didn't trust nothing. I just listened to them and barely talked to them.” Another client, when asked if treatment providers were aware of his goals, stated: “They know that I’m homeless. They know that I have an animal. And they know that I need permanent housing. And they’re not giving me any help at all.”

Assisting clients in attaining tangible services such as housing, benefits and health care may be integral not only in the client-treatment provider relationship, but also in keeping clients engaged in services. One client explains why she disengaged from services after failing to receive assistance from her treatment provider:

“I had the stroke…I was paralyzed on the left side. I couldn’t move it, or anything like that. And then when I came home, I needed help. I figured he was my social worker; he should be able to help me find certain things that I needed so I could take care of myself. And he was telling me that I had to advocate for myself. So my thing is, well, if I have to advocate for myself, what do I need you for? So, I cut my ties.”

Nearly all treatment providers interviewed (96%, n = 24) reported using shared decision strategies to develop goals. To better attain the goals set, some providers endorsed the idea of working as a team with clients (32%, n = 18) and employed strategies of working towards short-term goals (28%, n = 7) to help clients work towards long-term goals in small, attainable increments. However, many clients felt that their needs and opinions were not being addressed (34%, n = 19) and even fewer reported feeling that they were making progress towards their goals (25%, n = 14).

Theme 3: Approaches and interventions that have the greatest impact

Clients (20%, n = 11) expressed the need to feel that they will not be judged if they are going to open up to treatment providers. One client explained, “[My treatment provider] just tells you to always be straightforward with her because she won’t judge you. She just wants you to be straightforward, so I can build up a relationship. And that's what we kind of did.” Clients also endorsed the need for positive feedback from treatment providers (21%, n = 12). One client reported, “I thought I was trying my best and everything I did wasn’t good enough for her to give me any kind of credit.”

Treatment providers were more likely to value their psycho-educational role (52%, n = 13), though very few clients made mention of this aspect of treatment (9%, n = 5). Psycho-education typically includes informing clients about their diagnosis, symptoms and treatment options, and although it may be part of most treatment plans, it did not emerge as an important aspect of treatment for the clients interviewed.

Peer support was not included in the interview guide, but some clients (11%, n = 6) brought up these relationships as meaningful and positive. Although no peer counselors were interviewed, the treatment providers interviewed (20%, n = 5) also emphasized the importance of peer counselors as a positive component of treatment. One client explained very clearly the benefit that peer support provided:

…and, the people that was there, they was really friendly. You know, they was like, they could relate to what you’re going through. Some of them had been through the same thing that you’re going through. Some of them used drugs in the past. Some of them….was ex-convicts…It was a big difference because, you know, even though me and my therapist got along real well, and me and my psychiatrist got along real well, I think the fact that they can relate to you more, because they’ve been in those situations before.

Clients reported that peer support contributed to their willingness to be honest and comfortable talking about their experiences: “One of the peer counselors here, he always gives me good advice, really key advice…maintaining…a positive attitude, and that will heighten my self esteem.”

Discussion

The findings of the current study indicate that mutual trust, respect and availability represent the building blocks of a meaningful client-treatment provider relationship. This is consistent with a substantial prior body of research highlighting the importance of the therapeutic alliance, the client perspective, and person-centered approaches.36–39 Clients and treatment providers agree that meaningful relationships require mutual respect for the client’s dignity and personhood. This respect is conveyed both verbally and nonverbally, and interviews suggest that clients may base their opinions of treatment providers by tuning into subtle cues. These findings suggest, along with previous research26 that treatment providers who are both reliable and able to meet clients where they are build trust (e.g., through consistency of staff assignments, flexibility with appointments, and consideration of schedules). Although research has shown the importance of these factors, it is clear from the current study that some treatment providers are unable or unwilling to forge a positive therapeutic alliance with clients.

On the other hand, these findings indicate that clients and treatment providers often disagree regarding goals of treatment and technical elements. Many treatment providers maintain an approach to therapy that emphasizes an illness-based model and the importance of the client accepting his or her symptoms and limitations. Treatment providers see themselves as well intended “psycho-educators,” providing expert consultation regarding diagnosis, course of illness, and treatment options. Treatment providers also view themselves as assisting clients in addressing psychological barriers to understanding and awareness of specific symptoms, the presence of an illness, and functional limitations. This approach draws upon classic psychodynamic principles that emphasize the primacy of insight and self-awareness.40 Clients, on the other hand, were more likely to view care as an opportunity to access resources. Clients are much less likely to endorse an illness model and are more likely to view treatment providers as facilitating access to important supports and services, especially related to housing and employment. Although this study did not include questions regarding peer support directly, both clients and providers suggested the importance of such relationships. Contrary to treatment providers that may follow an illness model, qualified peer support workers support increased hope, empowerment, self-esteem, reduced stigma and engagement.41–47 However, some findings are less positive and bring into question the efficacy of peer support programs on outcomes for people with severe mental illness, suggesting that experimental trials evaluating peer support are needed.48

These differences extend to client and treatment provider perceptions of goals for treatment: clients are more likely to value care when they see a direct link to acquisition of services and supports. Treatment providers also view these as good outcomes but are more likely to view them as natural consequences of improved insight and self-awareness. However, research indicates that insight increases the risk of suicidality and lowers quality of life.49 Treatment providers are therefore less likely to value the role of facilitating access to services and supports.

The evident discord between client and treatment provider accounts of goals is an important obstacle to improving relationships and engagement in services and the literature supports the importance of goal consensus as a key component of positive patient outcomes.50 Treatment providers reported the importance of shared decision-making, but they may be ill equipped to address the most important goals of their clients, which extend outside the perceived clinical relationship. Although case managers are often tasked to address these needs, the client may be without these additional services or the services received may be inadequate, leaving the client without access to many services even while some mental health treatment is in place.

Previous research has shown the value of training treatment providers in goal setting interventions34; however, the training and qualifications of many treatment providers focus primarily on mental health goals and outcomes rather than tangible services. Furthermore, many treatment providers may not see it in their scope of service to provide support of certain client goals such as housing. Clients, however, are looking to their treatment providers to connect them with services and help them attain the goals that are most critical to them. The resources and knowledge about supporting clients in reaching these goals should be made available to treatment providers who otherwise may not feel prepared to address these needs. Treatment providers often report the burden of large caseloads and paperwork, and treatment provider burnout is well documented.51–54 Research investigating the variables involved in treatment provider burnout37 has found a positive relationship between hours of client contact per week and a sense of accomplishment, suggesting that both clients and treatment providers may benefit from their time together and bringing to question the burden of administrative duties required of treatment providers. Research of burnout on individuals with borderline personality disorder and their treatment providers has shown that both clients and providers experience burnout. Of particular importance to this study; treatment provider burnout appears to be related to the level of frustration clients reported with treatment providers in their recent past rather than with the current provider.55

The importance of collaboration between treatment providers and consumers has been well documented,24–26 but the problem persists: participants in this study frequently cited a lack of collaboration and communication in the services available. Policy changes that might serve to improve the relationship between treatment providers and clients include mandated lower case loads, internal support of treatment providers to prevent burnout, and staffing positions dedicated to housing and/or education to better support clients in need of these services. The challenges of implementing evidence-based practice56 may be an important area to consider when addressing the deficiencies in services offered reported in this study.

The conclusions of this study should be of interest to policy makers, researchers, and clinicians, as clients with a history of disengagement and/or incarceration, as well as young adults, are often those most in need of services. Mental health workers have the opportunity to address issues related to homelessness and unemployment. Clients in this study reported a desire to work towards both housing and job stability, but they often find little help in accomplishing these goals. It may be necessary to address these needs first before expecting clients to engage fully in their mental health treatment and recovery.

The findings reported here suggest many areas for further investigation, including the following:

  • Can client-treatment provider relationships characterized by high levels of trust and regard lead to engagement in care when there is overt or covert disagreement regarding technical interventions and goals?

  • To what degree does active non-adherence with treatment recommendations (e.g., refusal to take prescribed medications or active substance use) impact the development of the trust and mutual respect necessary for a meaningful client-treatment provider relationship?

  • Is the trust/regard dimension of the therapeutic relationship more predictive of sustained engagement in care than agreement regarding approach and goals for treatment?

  • Are there differences in the training, support or experiences of providers who offer positive reinforcement and non-judgmental attitudes? How can treatment providers be further-encouraged to employ these approaches to care?

Research addressing these questions will be very helpful in understanding determinants of engagement in care for this population.

There are limitations to the present study’s findings. Although the sample size was appropriate for an exploratory qualitative study, a larger and more diverse sample would be needed to support generalization of these findings related to content analysis. The data collected in this study relies on self-reported demographic and clinical information. It should be noted that some clients may have been unable or unwilling to offer accurate reports of their mental illness and clinical history, and researchers were unable to confirm the reliability of such data.

Implications for Behavioral Health

Clients receiving mental health treatment report trust, respect and mutual understanding as primary to positive relationships with their treatment providers. Treatment providers similarly report these as factors in successful relationships, and this finding is consistent with person-centered approaches to care. Clients, however, are more likely to view mental health services as a pathway to other services that meet tangible needs such as housing, education and employment. This study suggests that client-treatment provider relationships might be improved if treatment providers are better equipped to meet the tangible needs of their clients. Future research should focus more closely on the extent to which agreement with regard to approach and goals for treatment is critical to client-treatment provider relationships.