The Choice Project: Peer Workers Promoting Shared Decision Making at a Youth Mental Health Service

Providing optimal care for young people with mental illness is essential given the prevalence and impact of mental disorders in this age group. Most mental disorders begin between the ages of 12 and 24, although help seeking is poor among young people, and those who visit a provider often do not stay long enough to receive adequate care (1). Working together with young people to understand and address these gaps is critical (2).

Over the past 20 years, consumer participation in mental health services has been recognized as essential for both service providers and consumers. Although several local and international strategies have been developed to promote consumer participation among recipients of adult mental health services (3–8), few strategies exist in youth mental health services (9,10).

The World Health Organization (11) and the United Nations (12) have both stipulated that young people have the right to make informed health care decisions. There is a clear responsibility to explore individual needs, values, and preferences for young people who seek care from mental health services. In addition to the emphasis on youth participation in service development and provision, there is also recent interest in strategies that promote the involvement of all clients in making decisions about their care (13–17). The most commonly suggested strategy is shared decision making (SDM), a collaborative approach to treatment decision making that incorporates evidence-based practices and client preferences and values (18,19).

Despite the appeal of SDM, studies in general practice and adult mental health services demonstrate relatively low levels of SDM (20–22). To date, trials of SDM interventions for mental health have been conducted among adults (4,23–26) and among children for whom adults (for example, parents) are the decision makers (27). Although the results are promising, these interventions tend to support a specific treatment decision (for example, treatment for schizophrenia). Two interventions have taken a broader approach and can be used for any decision, regardless of presenting problem (26,28,29). One of these, CommonGround, is used in mental health clinics and inpatient settings. Based on the recovery model, CommonGround employs peer workers to help clients use an online decision support tool in the waiting room (28). This tool allows clients to explore their preferences and values in relation to treatment options, and a report is prepared and taken into the time-limited session with their clinician. In clinics that use this SDM tool, clients arrive 30 minutes before their appointment with the clinician and are invited by peer workers to use the tool. Following the consultation, peer workers are available for further support. The tool helps clients convey complex information in a report that can be quickly reviewed by clinicians. This program has been implemented in both adult and young adult settings (29).

This novel combination of SDM and peer work is yet to be tested in younger populations, such as adolescents. Peer work among young people needs to consider developmental stage, social and educational factors, and stage of mental illness when relevant. A small number of youth participation models have been described, and the models highlight facilitators and barriers to youth involvement in the context of delivering peer-led decision support. Monson and Thurley (30) described a youth peer work service largely driven by young people themselves. Peer workers are former clients of a specialized youth mental health service, who use their lived experience of mental illness to promote recovery for current clients of the service. One barrier related to the use of peer-led electronic decision support is the lack of availability of online tools, even though clinicians and clients want to use technology in youth mental health care (31).

To investigate the usefulness of peer work and SDM with online decision support tools in a youth mental health setting, we took the basic principles of CommonGround (delivered by peer workers and completed in waiting rooms, with a report taken into clinical session and a focus on promoting SDM) and applied this to youth mental health care. The Choices About Healthcare Options Informed by Clients Experiences and Expectations Project (Choice Project) employed youth peer workers to support other young people to make informed decisions about treatment options. The purpose of this study was to evaluate this intervention, which included an online tool codesigned with peer workers to facilitate SDM.

Methods

Setting

The study took place at a youth mental health service in New South Wales, Australia, known as headspace Gosford. Young people ages 12–25 are assessed by the triage team—the Youth Access Team (YAT)—which is staffed by allied health professionals. A revised version of the HEADSS assessment instrument (Home, Education, Activities, Drug use and abuse, Sexual behavior, Suicidality and depression) (32) is used to determine client needs and the most appropriate treatment. Treatment options at the time of the study included one-on-one counseling with a clinical psychologist; an appointment with a general practitioner or nurse; a counseling service for cannabis use; a general support service, including housing assistance; a vocational support service; and a welfare service.

Intervention

The intervention has been described in detail elsewhere (Simmons MB, Coates D, Batchelor S, et al., unpublished manuscript, 2017). In summary, peer workers welcome clients before their appointment with the YAT. Peer workers use an online decision support tool delivered via electronic tablet (an iPad). The tool provides decision support based on the Ottawa decision support framework (33) and was designed in line with the International Patient Decision Aid Standards (34); however, the tool is not a traditional decision aid because it does not address a specific decision or disorder. Peer workers were involved in codesigning the tool to ensure quality, usefulness, and acceptability. [More information on the intervention is provided in an online supplement to this article.]

Peer workers use the tool with clients in the waiting room before their appointment with a YAT clinician. Clients complete the “What matters to you?” section, which explores their needs and preferences. A report based on this information is generated for the YAT clinician to view at the start of the appointment. The tool is also available for use during the appointment, with a section on treatment options (“What are my choices?”) to be discussed at the end of the session after the standard HEADSS assessment. The treatment options are described briefly in the tool and are complemented with three key questions that promote an SDM approach to treatment: “What are my options?” “What are the possible benefits and harms of those options?” “How likely are each of those benefits and harms to happen to me?” (35). After the appointment, clients can see a peer worker for further support.

Participants

Young people ages 16–25 years attending headspace Gosford for an appointment with the YAT were invited to participate in the study.

Outcome Measures

Before and after their assessment with the YAT clinician, participants completed the Decisional Conflict Scale (36), a self-report measure that assesses the degree to which a person is conflicted about a decision that he or she faces. Higher scores reflect higher decisional conflict, an undesirable outcome. After the assessment with the YAT clinician, participants completed the nine-item Shared Decision Making Questionnaire (37) and four items from the headspace Service Satisfaction Survey (38). Higher scores on these two instruments reflect higher perceived SDM and satisfaction, respectively. The nine-item Shared Decision Making Questionnaire was administered once to the comparison group (with reference to the YAT clinician only) and twice to the intervention group (with reference to the YAT clinician and the peer worker) to measure the level of perceived SDM. Participants were asked to cite one or more reasons that they were attending headspace Gosford. Six options were available, with an additional “other” option.

Design

To evaluate the intervention, a historical comparison group design was used. During the 26-week period between January 8 and June 23, 2014, all clients ages 16–25 attending headspace Gosford to see a YAT clinician (that is, for an initial assessment) were invited to complete study measures before and after this assessment. No additional interventions were used during this time, and YAT assessments were completed as usual. This group is referred to as the comparison group. Between June 24 and December 1, 2014, peer workers began to work at headspace Gosford and invited clients to use the SDM tool and complete the study measures. This group is referred to as the intervention group. Informed consent was obtained from all participants, and this study was approved by the New South Wales Human Research Ethics Committee (LNR/13/HNE/346).

Statistical Analysis

We present frequencies and percentages of responses to binary variables and means, standard deviations, and nonparametric statistics for continuous outcomes. One-way analyses of variance (ANOVAs) were used to compare demographic variables. Paired t tests were used to test for change in continuous outcomes over time within the intervention group; repeated-measures analysis of covariance tested for change in continuous outcomes over time between the comparison and intervention groups. A multiple linear regression tested for factors that were associated with a continuous outcome. When significant differences were found, effect sizes (Cohen's d, η², and Cohen’s f²) were calculated to assess the magnitude of the difference between groups. Statistical analyses were conducted with IBM SPSS Statistics, version 22.0 (39).

Results

Participants

In total, 229 young people participated in the study, with 80 participants in the comparison group and 149 in the intervention group, although response rates for each measure varied. The groups did not differ significantly in age, gender, or reasons for coming to headspace (Table 1).

Perceived SDM

Participants’ scores on the Shared Decision Making Questionnaire for the YAT clinician were significantly higher in the intervention group than in the comparison group (p=.015). For individual SDM items, scores were significantly higher in the intervention group on four of the nine items (Table 2).

In the intervention group, SDM ratings were significantly higher for the YAT clinician than for the peer worker (p=.015). Significantly higher ratings for the YAT clinician were observed on five of the nine items (Table 2).

Decisional Conflict

Across both groups, a significant decrease in decisional conflict was observed from before the YAT assessment to after the assessment, with significant changes observed on each of the decisional conflict subscales (Table 3). However, no difference in the decrease of decisional conflict scores was observed between the intervention and comparison groups.

Satisfaction

Overall, both the comparison and intervention groups reported high satisfaction levels, and no participant endorsed “disagree” or “strongly disagree” on any of the individual items (Table 4). For the comparison group, the total score (sum for the four individual items) ranged from 5 to 20 (maximum possible score of 20), with a mean score of 18.07±2.61. For the intervention group, the total score ranged from 15 to 20, with a mean score of 18.56±1.76. No significant between-group differences were found in satisfaction levels (Table 4).

Factors Associated With Satisfaction

A model for measuring associations with satisfaction (total satisfaction score) was developed by using the level of SDM (total score on the nine-item Shared Decision Making Questionnaire) and the level of decisional conflict (total score on the Decisional Conflict Scale after intake appointment with YAT clinician). The final model was statistically significant (F=14.21, df=2 and 71, p<.015, R²=.286, f²=.40), with results indicating that higher SDM scores (β=.333, t=2.86, df=130, p=.006) and lower postassessment decisional conflict (β=–.295, t=−2.54, df=130, p=.013) were significantly associated with higher satisfaction.

Discussion

This study demonstrated the feasibility of implementing a peer worker intervention promoting SDM in a youth mental health setting. Clients actively engaged in the intervention, demonstrating a willingness to connect with peers in this setting. Clients in the intervention group reported feeling more involved than those in the comparison group in making treatment decisions with their YAT clinician, although the magnitude of the effect was small. However, this finding is of critical importance to services that seek to promote client-centered care, as well as to youth mental health services in general, where help seeking and clinical engagement are significant barriers to timely treatment of mental illness (40).

In the intervention group, clients felt significantly more involved with their YAT clinician than with their peer worker in making decisions about treatment. Given that clients make treatment decisions with the YAT clinician rather than with the peer worker, this finding supports the proposition that peer work, SDM, or the combination of peer work and SDM result in clients’ feeling more involved in treatment decision making with their clinician. The role of the peer worker was to focus on promoting involvement and engagement in the service, and thus it was possible for clients to feel just as involved, or more involved, with peer workers and to have blurred perceptions of the roles of peer workers and clinicians. Had clients in the intervention group felt equally as involved in treatment decision making with peer workers as they had with YAT clinicians, uncertainty would remain about whether clients merely felt more involved in the service in general rather than at the critical point when decisions about treatment are made. However, it is also possible that peer workers primed clients to feel more involved by promoting SDM through both the tool and motivational support directly before their appointment with the YAT clinician.

Clients in both the comparison and the intervention groups were highly satisfied with their care, and no difference in satisfaction was found, possibly because of either a ceiling effect or the fact that only four items measured client satisfaction. A validated measure of client satisfaction specifically designed for youth mental health services has since been developed based on the items used in the current study, and the full version should be considered for future studies (38,41). Similarly, both groups experienced a significant reduction in decisional conflict after their appointment with the YAT clinician, and no differences in this reduction were found between the groups. However, regression analysis showed the importance for client satisfaction of both perceived involvement in decision making and lower decisional conflict directly after the decision assessment, which represented a small-medium effect. This finding highlights the importance of interventions that promote SDM and focus on increasing client satisfaction, such as by use of decision support tools.

These findings add to the growing fields of SDM and peer work in mental health, which have largely omitted young people. The intervention resulted in clients feeling more involved in making decisions, which is consistent with SDM interventions for adults diagnosed as having depression (42,43) and schizophrenia (4,44). However, studies in adult populations have also demonstrated effectiveness in terms of reducing decisional conflict (42,45,46), increasing client satisfaction (42,43,47), and improving knowledge (an outcome that we did not examine in this study because of the diversity of treatment decisions that we were seeking to support) (4,42). Most important, this study demonstrated that the combination of peer work and SDM can play an important role in a youth mental health service with a focus on early intervention, as it has in adult medication clinics for individuals with severe mental illness (26,28,29).

This study also contributes to the understanding of how technology can be used in youth mental health services. Both young people (48) and clinicians (49) are enthusiastic about the use of technology to promote mental health and well-being; however, there are few well-tested tools for this purpose, particularly tools for use in the clinical consultation (31). With the increased use of smartphone technology, mental health clinicians need evidence-based Web sites and applications to fully engage digitally connected young people and maximize the chances of providing appropriate treatment in a timely manner (50).

The study had several limitations. The design was not randomized; funding limitations precluded our undertaking a large cluster randomized or stepped-wedge randomized trial. More clients in the intervention group than in the comparison group participated in the evaluation component, and there were missing data for several measures. It is likely that the presence of peer workers made the service more welcoming and that clients were less likely to participate in the research if asked by reception staff, with whom they may not have built a relationship. This was the first large research study at headspace Gosford, and data collection procedures became more refined during the intervention period. Also, it was not possible to tease apart the impacts of the two main components of the intervention—SDM and peer work. Finally, there was no formal measure of fidelity to ensure that the decision support tool was used in full or that it facilitated SDM as measured by audio-recording sessions or by use of an observed rating scale (51).

A considerable strength of the study was the real-world nature of the design, which showed that the intervention can be readily adapted by other services. In addition, both the peer work roles and online tool were codesigned with young people, ensuring acceptability and integrity.

Future research should focus on the adaptation and effectiveness of a combined SDM and peer work intervention for tertiary youth mental health services, where the effects are likely to be more profound. It is also vital to better understand the mechanisms by which SDM interventions lead to improved outcomes. Determining the role of mediating and moderating factors related to more positive experiences of services, better engagement, and improved clinical outcomes will help define the role that SDM and peer work can play in youth mental health services. By involving young people in multiple ways, it may be possible to promote help-seeking behaviors and clinical engagement thereby improving outcomes for young people.

Conclusions

This study demonstrated that involving young people in youth mental health services with peer workers and SDM was feasible and led to participants’ reports of feeling more involved in making decisions about their care. Interventions that target perceived involvement and reduction in conflict about treatment decisions are likely to improve client satisfaction with care.