Barriers to Client Engagement and Strategies to Improve Participation in Mental Health and Supported Employment Services

Much has been published about clients who are difficult to engage. Individuals with borderline or antisocial personality disorder, substance use disorder, or other conditions often exhibit anger, threats, or illegal behaviors; reject help; and refuse to participate in treatment (1–5). Similar challenges are seen among clients with complex behavioral and medical conditions that are often exacerbated by poverty, unemployment, unstable housing, and legal issues (6). Between 33% and 46% of all clients in community mental health centers discontinue recommended services after beginning treatment (7, 8).

In every clinic, a small number of clients present these engagement challenges. Case managers, supported employment specialists, and clinicians may not have experience with managing these situations. In this article, we provide a qualitative analysis of consultations regarding 105 clients in the Social Security Administration’s Supported Employment Demonstration (SED) who were having difficulty accessing and participating in care services. These clients were not representative of all SED clients; the purpose of this study was to identify the barriers to accessing and engaging in care encountered by the small number of clients whose behaviors were the most challenging and to describe strategies that consultants and teams developed to better serve them.

Methods

SED

The SED, funded by the Social Security Administration, used an experimental design to test the hypothesis that providing behavioral health treatment, employment services, and other supports to people at high risk for disability would enable them to become self-sufficient and avoid dependence on disability benefits (9). The project recruited 2,960 individuals in 30 communities across the United States between November 2017 and March 2019. Eligible participants had been denied Social Security disability benefits despite claiming the inability to work because of a behavioral health condition. Clients were randomly assigned to one of three interventions: usual care (client-initiated treatment as usual), basic care (team-based treatment and care management with supported employment specialists and services), and full care (team-based treatment, care management, and supported employment services with the addition of a nurse care coordinator). Teams offered services for 3 years to each participant in the two active treatment groups (basic care and full care) (9).

Early in the project, feedback from treatment teams indicated that some clients were difficult to engage in services. For example, among the 984 clients assigned to full care teams, 41% (N=403) failed to complete intake assessments and engage with a nurse care coordinator (10). Additionally, among the nearly 2,000 clients assigned to the two SED arms that included supported employment services, only about one-third received those services in each of the first 24 months of follow-up (11). The SED implementation leadership team responded to these findings by creating a consultation program in which four experienced psychiatrists and psychologists advised teams working with clients who were challenging to engage. In this study, we examined the content of 105 completed consultations. All SED activities were approved and monitored by the Westat Institutional Review Board; one site also received secondary review by the Los Angeles County Institutional Review Board. All enrolled clients gave informed consent.

Clients

Teams requested clinical consultations regarding 105 individual clients (∼5% of those in active treatment groups). The most common reasons for consultation involved clients who did not complete baseline clinical assessments and initiate psychiatric care or clients who were not meeting with supported employment specialists. Demographic and diagnostic profiles for all SED clients were previously published (12). Most SED clients self-identified as female (57%, N=1,674 of 2,960), White (56%, N=1,580 of 2,823), and non-Hispanic (87%, N=2,558 of 2,928). Many were age 35 or older (58%, N=1,702), reported at least a high school education (81%, N=2,406), lived with relatives (69%, N=2,043), had never married (55%, N=1,615), and were unemployed (81%, N=2,397) (12). The most common diagnoses were anxiety (most commonly posttraumatic stress disorder), depression, and personality disorder (most commonly antisocial or borderline). SED clients averaged more than six behavioral and medical diagnoses, and those for whom consultations were requested generally had even more conditions. Clients also experienced multiple social determinants of poor health, such as lack of income, unemployment, unstable housing, and unsupportive social networks (12). To preserve client confidentiality, the consultants did not document demographic data for the 105 clients about whom consultations were provided.

Analysis Plan

We conducted a document analysis of notes kept by the consultants to better understand the needs of these clients and approaches used by clinicians to engage them in services. Quality monitors for SED implementation teams maintained regular contact with all study sites and invited the teams to submit consultation requests for clients whose behaviors were challenging. The consultants kept notes regarding each consultation, including the reason for and circumstances leading to the consultation, description of the participant, and recommendations to the treatment team. The research addressed the following questions: What were the most common barriers to accessing and participating in behavioral health services? What strategies did consultants and the clinical teams use to address these barriers?

A document analysis approach was used in this study (13). We first collected all available written or electronic documents related to each consultation, including e-mails, case reports submitted by teams requesting consultations, and consultants’ summary notes. The four consultants (T.E.S., D.H., G.M., R.E.D.) met as a team and created an initial list of a priori categories (codes) related to the research questions. Consultants then reviewed their own notes, applied the a priori codes to all 105 consultation cases, and identified new codes through inductive review so that codes and themes were driven by the data. The four consultants then formed pairs, and all rated their partner’s notes, after which the pairs met to discuss their code lists and identify points of overlap as well as disagreement. Each pair reached consensus on a code list, after which all four consultants met to share the two code lists, reach consensus on common terminology for similar codes, and develop a final code list.

Results

Barriers to Accessing and Engaging in Services

Table 1 lists common reasons why clients experienced barriers to accessing and engaging in mental health services.

Complex behavioral health conditions and medical comorbid conditions.

Most consultations addressed clients with multiple psychiatric and medical conditions that were inadequately managed. Although baseline screens of SED enrollees revealed low rates of substance use (14), at implementation and supervision meetings, the treatment teams consistently estimated that substance use disorders affected 50%–80% of clients. Few clients reported significant alcohol or other drug use during initial psychiatric assessments. Moreover, clients rarely reported that substance use presented barriers to accessing health care or finding employment, but treatment teams repeatedly reported that substance use was a primary barrier to engagement in care services and employment. Many clients for whom consultations were requested were taking prescription opioids or medical marijuana, often in combination with benzodiazepines, and one clinic team reported that “100% of our clients use cannabis regularly.”

Clients reviewed by consultants reported legal problems related to substance use, including arrests for selling illicit substances or crimes while under the influence; these barriers limited their access to transportation (e.g., a suspended driver’s license). Many clients reported antisocial behaviors, including repeated violence as well as criminal and exploitative behaviors that led to loss of employment and other supports such as housing.

Common medical conditions included chronic pain, obesity, diabetes, epilepsy, and work-related injuries, such as back and traumatic brain injuries. Several clients reported chronic headaches, dizziness, fatigue, and vague neurological symptoms that neurologists or psychiatrists could not adequately diagnose or treat.

A composite client had bipolar disorder, posttraumatic stress disorder, alcohol use disorder, fibromyalgia, and severe hip problems (requiring >10 prior surgeries) and received services from a primary care provider, neurologist, cardiologist, surgeon, hematologist, and psychiatrist. Other clients reported medical problems but did not have access to either primary or specialty medical care. Fragmented care was a common theme, because many clients were unable to engage with providers for extended periods or received uncoordinated and even conflicting care from multiple providers who did not communicate with each other.

Lack of motivation to work.

Many clients were unwilling to work, including those who were appealing their disability denial and were advised by a lawyer not to work pending the appeal. Some clients participated in the SED only to access funds to pay for housing and other critical needs. Other clients were unwilling or unable to work because of psychiatric symptoms or medical conditions, or the need to care for family members, or because they did not want to lose future opportunities for disability benefits. Many were working “under the table,” for example, performing miscellaneous odd jobs or farming without reporting income, which they preferred to competitive employment. Some clients had accumulated substantial debt and were avoiding authorities and creditors. Some SED sites were in economically depressed locations where employment opportunities were limited. One participant noted, “Nobody in this county works.”

Difficult relationships.

Unstable and traumatic relationships were common. Many clients had significant histories of adverse childhood events and trauma, including mental or physical abuse, and domestic violence in adulthood. Some had lost or been abandoned by their parents, often because of incarceration. Lifelong experiences of trauma contributed to prominent mistrust of authority figures, including health care providers. Many clients reported feeling triggered by providers and reacted with anger, lability, and lack of adherence.

A composite client was a woman who was constantly relocating to shelters or to friends’ residences across two states, with a history of bipolar disorder and chronic substance use, loss of custody of her child, multiple arrests for misdemeanors related to substance use, and long-standing conflictual relationships with family members. Treatment team members struggled to establish consistent and trusting relationships with clients who experienced long-standing and severe difficulties with basic needs.

Social determinants.

In addition to behavioral health, medical, and relationship difficulties, these clients were experiencing multiple social determinants of poor health. Having been denied disability applications, they had limited income as well as food and housing insecurity. Many had no health insurance. Each day, the goal was survival rather than long-term employment.

Staff burnout.

Treatment team members expressed anger and frustration toward clients who rejected help, were threatening, or created dissension among staff. Some staff experienced repeated threats of physical violence from clients and feared for their own safety. These experiences contributed to team members feeling helpless, “burned out,” and unable to sustain outreach efforts for extended periods. Consultations often involved helping the staff feel protected, supported, and capable of continuing outreach.

Strategies for Addressing Barriers

Table 2 lists themes related to strategies for addressing the identified barriers to treatment and supported employment services.

Be persistent.

A common theme involved teams committing to ongoing, indefinite outreach to clients who were not participating in care or pursuing employment. Unlike in typical community mental health care, all SED clients remained on teams’ caseloads, and team members continued outreach throughout the 3-year follow-up period. This approach differed from most clinic policies, which require patients to be discharged after defined periods of lack of engagement. Some SED clients showed significant improvements after long periods of disengagement, often for unexpected or unclear reasons. Clients with significant trauma histories and difficulties trusting authorities sometimes responded to persistent, noncritical, strengths-based outreach and support before engaging fully in clinical and employment support services.

Coordinate services.

Consolidation and coordination of care were critical. Many clients improved when as much of their care as possible was brought into a single clinic. This approach decreased the fragmentation of care, improved provider communication and care coordination, and removed obstacles to job searches. When consolidating care was not possible, a team member (typically a care manager or nurse care coordinator) would accompany the client to appointments with medical specialists or off-site psychiatric prescribers. Despite the time burden, teams noted that accompanying clients to these visits often resulted in better service engagement.

Provide trauma-informed care.

Clients who repeatedly expressed no interest in engaging in care services or who responded to team outreach with anger and hostility were especially challenging. The SED implementation team leaders and consultants advised team members on trauma-informed approaches, including meeting client needs in a safe and compassionate manner, avoiding practices that retraumatize clients, and building on the strengths and resilience of clients in the context of their environments and communities (15).

Calibrate goals.

SED consultants and team members noted the importance of calibrating client goals in a person-centered manner. Many consultations were initiated because client service plans were ineffective due to mismatches between the team’s goals and the participant’s immediate circumstances and goals. Team members learned to apply principles of harm reduction and motivational interviewing (16) in their efforts to engage clients with active substance use. They also recalibrated service goals to validate clients’ immediate needs and focus on achievable short-term objectives.

Understand complex medical conditions.

SED care teams appreciated education and guidance regarding clients’ complex medical care needs. As noted earlier, many clients, and especially those for whom teams requested consultations, had multiple psychiatric and medical conditions, many of which were poorly controlled. Team members benefited from education regarding screening for and the detection and management of substance use and chronic pain syndromes. Educating staff facilitated care coordination efforts and allowed team members to identify realistic goals. Related activities included educating clients and support persons to address self-stigma or public stigma that contributed to clients’ reluctance to see providers or follow service recommendations. Team members described many instances in which they assumed active roles in educating their clients and advocating for them when interacting with families, health care providers, and social service agencies.

Foster team morale.

Team members reported feeling angry, resentful, and helpless when their recommendations were consistently rejected or their competence was challenged by clients. These responses were magnified when clients experienced poor outcomes, including death: ∼3% of the 2,960 SED enrollees died during the study (unpublished data, Drake RE, 2021). Many SED staff did not have formal training in individual and group psychodynamic processes, such as projective identification and splitting. These staff benefited greatly when consultants and team leaders validated emotional reactions to clients with complex behavioral conditions that were challenging to manage. Staff needed reassurance and reinforcement of their clinical judgment and skills from other team members and supervisors in these circumstances.

All the strategies listed earlier helped to improve team morale, engagement of clients, and outcomes. Informally, consultants heard from teams many times about clients who began to engage over time and achieved positive outcomes when goals were more realistic.

Discussion

Clients with personality disorders, substance use disorders, extensive trauma histories, and multiple medical comorbid conditions are more likely to have difficulty managing personal, clinical, and employment relationships (1–5, 17–20). Specific psychotherapeutic approaches are effective for these clients, such as trauma-informed care (15) or dialectical behavior therapy (21, 22). However, many mental health care managers and supported employment specialists have not been trained in these approaches, and they are more likely to become angry or dismissive toward clients and to blame themselves when experiencing difficulties engaging them (23, 24).

The clients for whom teams requested consultations were not representative of all SED clients; they had the most challenging behaviors, likely resembling a minority of clients in community mental health centers who cause similar challenges or who reject care. This population is nevertheless important: up to 60% of individuals with serious mental illness either are not engaged in care or have discontinued recommended services (7, 8, 25, 26), and it is reasonable to assume that many individuals who fail to receive care resemble the clients whose behaviors were challenging as described in this article.

Recent national and state health care reforms emphasize a focus on population health and encourage community mental health providers to identify and reach out to individuals who are not consistently engaged in care. Many individuals in the community experience barriers related to social determinants of health, such as food insecurity, unstable housing, unemployment, and poverty. Rather than waiting until individuals experience crises, a population health approach requires that providers monitor patients' conditions, sustain outreach, and offer prevention and early intervention services to clients in order to limit the complications of chronic illness and disability (27). When providers are responsible for the overall health of the population, they have incentives to assess and address common social determinants of health (28).

This study’s findings can be used to identify standards of care related to outreach and engagement for clients with complex behavioral conditions who have high needs. These clients need care, provided by multidisciplinary teams, that is acceptable, accessible, and available (29). Team members should be trained to follow principles of trauma-informed care (15), person-centered planning (30), shared decision making (31), motivational interviewing (16, 32), and harm reduction (33) in their outreach and engagement efforts. Rather than allowing or encouraging clients with complex behavioral conditions who have high needs to disengage from care, clinic teams should document their repeated efforts and be persistent. Clinics and clinicians will have better protection from liability if there are clear standards of care related to outreach, documentation, and engagement efforts for these clients.

The main limitations of this study included selection bias and the use of documents as the primary data source. The lack of demographic information for the study sample, which represented a small number of the overall SED participants, further underscores the need for caution when generalizing results to community mental health clinic populations. Consultation notes also do not provide comprehensive information for research. We were unable to obtain alternative perspectives from the clients discussed in the consultations, which would have allowed for triangulation of themes to strengthen the findings. We also did not attempt to account for key social determinants known to affect access to services, such as transportation availability, provider density, insurance status, and individual or structural racial-ethnic biases. We note, however, that an interim SED process report (14) identified similar themes based on ethnographic interviews and observations of enrolled clients.

We did not systematically collect data on the use and impact of the identified engagement strategies. Our qualitative analysis was meant to identify specific barriers to accessing and engaging in care services and strategies for mitigating those barriers that can be formally tested in future studies. We received much anecdotal feedback, however, that some clients responded well to the suggested strategies, and team members felt safe and more empowered to conduct outreach after they received training on motivational interviewing, shared decision making, and trauma-informed approaches. Clients with complex medical problems often improved after staff consultations and referrals to specialists, and many clients engaged in treatment when posttraumatic stress disorder or substance use disorders were identified and directly addressed.

Conclusions

A small number of clients are challenging to engage in community mental health services. Strategies for staff to meet engagement challenges include persistence, team-based coordination, trauma-informed care, shared decision making, motivational interviewing, calibration of goals, and consultations. Offering expert consultations on clients who are difficult to engage creates opportunities to build staff skills, increase morale, and improve clinical outcomes.