Implementation of evidence-based group interventions in a community-based mental health service for self-harming and suicidal youth

ABSTRACT

Objective

Suicide and self-harm have significant health, economic and social costs. Group-based interventions can be an effective and advantageous treatment approach; however, there is little evidence on the application of these groups in youth community mental health settings in Australia. Using the RE-AIM framework, we explore the feasibility and implementation of two group programmes in a youth-focussed community-based mental health service: dialectical behaviour therapy-adolescent (DBT-A) and cognitive behavioural Self and Strength (S&S).

Method

Data were collected from an Australian community mental health service that implemented two group therapy treatments over 12 months. RE-AIM implementation framework outcomes included programme reach (patient flow, group uptake, treatment completion), effectiveness (youth and care clinical outcomes), adoption (staff perception), implementation (barriers), and maintenance (service commitment and acceptability).

Results

Groups were considered for all eligible clients (61% of referrals), of these 36% accepted treatment and of those who commenced, 71% completed DBT-A and 82% completed S&S. DBT-A participants reported improved suicidality, emotion dysregulation, self-harm and quality of life (QoL). S&S participants reported improved QoL. Despite small sample size, treatment was acceptable to participants, with high retention and adherence across both groups. Clinicians involved in facilitation noted positive clinical impacts.

Conclusions

Group-based interventions have the potential to reduce emotion dysregulation difficulties for youth. Understanding barriers to uptake of group programmes in community-based mental health settings is essential to improve the reach and implementation of group programmes.

Key Points

What is already known about this topic:

1. Self-harm and suicide are major public health concerns for youth, with self-harm linked to suicide completion.

2. High rates of self-harm and emotion dysregulation occur in youth presenting to community-based mental health services. Clinicians need to be able to offer evidence-based interventions to support these individuals.

3. Group-based interventions for emotion dysregulation present an opportunity to improve health outcomes, facilitate skill development and encourage social help-seeking behaviour.

What this topic adds:

1. Group-based interventions for emotion dysregulation have positive clinical outcomes for young people who engage in these programmes and offer potentially cost-effective solutions to addressing youth mental health; however, reach was low.

2. Staff facilitators and group participants reported high acceptability and provided positive feedback on the group programmes.

3. Further investigation into the barriers for group therapy is required to develop and implement strategies to increase uptake of group therapy.

KEYWORDS:

• Community mental health
• youth mental health
• implementation
• self-harm
• suicide
• group therapy

Suicide is the leading cause of death for young people aged 15–24 years (Australian Institute of Health and Welfare [AIHW], 2023). Suicidal ideation has been reported by 8% of adolescents, and suicide represents 31% and 39% of all deaths in young people aged 15–17 years and 18–24 years, respectively (Australian Institute of Family Studies [AIFS], 2017; AIHW, 2023). The yearly direct and indirect costs of youth suicide are estimated at $511 million (Kinchin & Doran, 2018). Approximately 60% of adolescents who attempted suicide reported previous acts of self-harm, with one in seven adult suicide completers treated for self-harm in the year prior to their death (AIFS, 2017; Hawton et al., 2015; Tidemalm et al., 2015).

The Longitudinal Study of Australian Children (LSAC) reports that approximately 16% of adolescents (aged 14–15 years) think about self-harm, and of those, over 50% engage in an act of self-harm (AIFS, 2017). Up to 20% of Australian young people report a lifetime prevalence of self-harming behaviours (Martin et al., 2017). Risk factors for complete suicide overlap with self-harm including sociodemographic factors (e.g., lower socioeconomic status, non-binary gender, and non-heterosexual orientation), educational factors (e.g., restricted educational achievement), negative life events and family adversity and psychological factors (e.g., impulsivity and poor social problem-solving) (AIFS, 2017; Hawton et al., 2012). Whilst deliberate self-harm may present without suicidal intent, recurrent self-harm increases the risk of suicide by misadventure and increases the likelihood of suicidal related behaviours (AIFS, 2017; Bergen et al., 2012).

High rates of self-harm behaviours and emotion dysregulation (including suicidal thoughts and behaviours) in young people often lead to referral to community-based mental health services (Mehlum et al., 2014; Taylor et al., 2011), presenting an opportunity to provide evidence-based treatment in these settings. Group therapy provides not only therapeutic benefits but also can maximise development of behavioural skills, encourage social help seeking behaviour (Guerreiro et al., 2015; Linehan, 1993) and maximise efficient use of resources.

Taylor et al. (2011) evaluated a manualised version of cognitive behaviour therapy (CBT) for deliberate self-harm, primarily focusing on a problem-solving-based approach, which informed the Self and Strength (S&S) group programme in the current evaluation. Adolescents who completed the programme reported improvements in self-harm, depressive symptoms and trait anxiety. Despite CBT being recognised as an evidence-based modality for a variety of mental health presentations in youth, programmes targeting self-harm have reported limited efficacy with a small effect for self-harm and a moderate effect for suicidal ideation (Labelle et al., 2015). Further investigation of CBT-informed programmes for deliberate self-harm and suicidal ideation is required.

There is strong empirical research on dialectical behaviour therapy (DBT) for the treatment of emotion dysregulation and suicidal behaviours (Linehan, 2015; Williams et al., 2010), including with young people (Cook & Gorraiz, 2016; Woodberry & Popenoe, 2008). DBT aims to replace problem behaviours with skilful behaviours and allow the individual to experience a range of emotions without acting upon them (Linehan, 1993). Miller et al. (2007) developed a structured multi-family skills DBT group programme (DBT-A) in which a parent/caregiver and adolescent learn skills together in the group to enable increased application of skills to the home environment. Adolescents who participated in DBT-A reported improvements in suicidality and self-harm (Kothgassner et al., 2021), and clinical improvements were observed for parent participants (Flynn et al., 2020) in terms of burden, grief, stress and use of DBT skills to meet their needs and the needs of their adolescent (Flynn et al., 2020).

We examine the feasibility and clinical outcomes for two group programmes delivered in a community-based mental health care service: the S&S protocol for young people that self-harm and the DBT-A protocol for suicidal young people and their families. The primary aim was to evaluate the real-world sustainability and success of these programmes within a local health service to guide health service planning. Given an existing evidence base, this represents an implementation study and we adopt the reporting criteria of the RE-AIM framework (Glasgow et al., 1999): reach, effectiveness, adoption, implementation and maintenance. In terms of reach, we report the number of referrals and demographics. For effectiveness, we report clinical outcomes of emotion regulation, suicidal ideation, self-harm, and QoL. For adoption, we report clinician adoption, including their perception of group programmes. For implementation, we report on barriers. For maintenance, we examine service commitment and treatment acceptability.

Method

Setting

The Southern Youth Mental Health Service (SYMHS) covers southern Adelaide and is provided by a community mental health team staffed by mental health nurses, social workers, psychiatrists, clinical psychologists, psychologists (including a student clinic and clinical registrars), and occupational therapists. The SYMHS offers early intervention for some of the most vulnerable, at risk, and disengaged young people aged 16–24 years. These young people have often experienced family breakdown; physical, emotional, or sexual abuse; homelessness; drug and alcohol abuse and forensic involvement. Consequently, many disengage from school and work, lose structure in their life and experience reduced social supports. These experiences often produce overwhelming emotions without the necessary coping skills. Fifty percent of the young people registered with SYMHS presented with significant emotion dysregulation at the time of development of the programme, and 60% were engaging in self-harm. Internal auditing suggested that approximately 40% of individuals would accept group therapy if offered.

Design

The study employed a naturalistic design; group allocation was based on eligibility criteria, clinical need and severity of symptoms following discussion with the clinical team, which included consideration of client presentation and additional factors such as client motivation to change, risk stabilisation and co-morbid mental health conditions requiring prioritisation (e.g., substance use). Participants were able to complete both groups consecutively if clinically indicated; however, data from the second group were not included in the current evaluation.

Two referral pathways exist to enter the group programmes. The primary pathway involves eligibility for groups being assessed by the allocated SYMHS clinician. Eligible participants are then allocated an internal care coordinator to organise and oversee client care. This coordination includes developing an integrated care plan, liaising with other service providers and helping clients to understand and manage their condition. A secondary pathway was introduced 8 months into the 12-month recruitment period, allowing external providers to refer participants to the groups. The external provider is required to commit to ongoing individual treatment, thereby eliminating the requirement for an internal care coordinator and reducing service demand. In both pathways, suitability for the group is determined following a comprehensive assessment and presentation to the clinical team.

Participants

Individuals were recruited from sequential referrals between October 2020 and September 2021. Eligibility criteria included sufficient English language, learning and literacy skills, willingness to effectively engage in group therapy and commitment to work towards treatment goals. Exclusion criteria for both groups included active psychosis, mania, or an episode of severe depression, significant antisocial or narcissistic personality traits, or substance dependence as primary diagnoses.

At least one episode of self-harm in the previous 12 weeks was required for the S&S group. Additional eligibility criteria for DBT-A included ≥3 borderline personality disorder criteria (one being recurrent suicidal or self-harm behaviour) according to DSM-5 (American Psychiatric Association, 2013), a suitable parent or family member able and willing to attend group sessions and family relationships as a perpetuating factor in the young person’s distress.

Participants who were not eligible for groups or declined to participate were offered treatment as usual which involved an individual youth-informed, supportive recovery focused intervention on identified areas of development (e.g., psychoeducation, goal setting, early intervention, skills building, structured psychological therapy, family work and liaising with service providers) with a member of the SYMHS team.

The project was classified as a continuous improvement by the Southern Adelaide Clinical Human Research Committee who deemed that ethics approval was not required. All participants provided informed consent prior to their participation.

Interventions

Self and strength (S&S)

This 10-session programme focussed on social problem-solving to improve emotional coping and reduce self-harm. Content included orientation to the CBT model, psychoeducation, coping and distress tolerance skills, problem-solving and relapse prevention, adapted from “Cutting Down: A CBT workbook for treating young people who self-harm” (Taylor et al., 2015). Additional content included DBT distress tolerance skills and sensory skills training. Sessions were conducted for 90 min each week, with a maximum of 10 participants. Homework activities were assigned each session to encourage participants to practice skills learnt. Treatment completion was defined as attendance at seven sessions. Three family sessions were scheduled during the 10-week group, run as separate drop-in sessions for family members during week 3, 6 and 9 and were not a requirement for participation. In the final session, group participants attended a “graduation ceremony” where they received a certificate and completed a qualitative questionnaire about their experience of attending groups. All S&S groups were facilitated by two clinicians with post-graduate training in delivering evidence-based CBT interventions.

Dialectical behaviour therapy for adolescents’ (DBT-A)

This 21–24 session group programme was based on the manual, Dialectical Behaviour Therapy with Suicidal Adolescents (Miller et al., 2007). The therapy included five modules: mindfulness, distress tolerance, emotion regulation, interpersonal effectiveness and walking the middle path. Sessions ran for 2 hours per week with a maximum of six families per group. The support person attended all group sessions with the young person, and weekly homework exercises were assigned to encourage both to practice skills learnt each week. Rolling entries allowed new participants to enter the group at the start of each module. When new participants entered, the introductory mindfulness session was repeated, resulting in a variation in treatment duration (21–24 sessions) dependent on the number of rolling entries. Participants were permitted to miss a maximum of six sessions over the programme. In the final session of the group, participants attended a “graduation ceremony” where they were presented a certificate of completion and invited to complete a qualitative questionnaire about their group experiences. The groups were facilitated by two clinicians trained in DBT-A.

Measures

Clinical outcomes for young people were assessed at baseline and post-treatment using the measures described in Table 1. The low Cronbach’s alpha for the EQ5D indicated that results need to be interpreted with caution. High standard deviations reflected the limitation of the small sample sizes. Higher outcome scores for the SIDAS, ISAS, DERS and DASS indicated higher pathology, whereas higher scores for the EQ5D indicated better quality of life.

Table 1. Measures for young people and carers participating in DBT-A and S&S groups.

Measures	Description	Cronbach’s alpha in current sample
Young Person
The Suicidal Ideation Attributes Scale (SIDAS) (van Spijker et al., 2014)	Suicidal ideation and risk. Five questions (scored 0–10) assess the severity of suicidal thoughts over the previous week	.91 DBT-A; .91 S&S
Inventory of Statements About Self-Injury (ISAS) (Klonsky & Glenn, 2009)	Lifetime frequency of 12 non-suicidal self-injury behaviours in the past 3 months	.70 DBT-A; .83 S&S
The Disorders of Emotion Regulation Scale – Short Form (DERS-SF) (Kaufman et al., 2016)	Emotion regulation deficits. Responses on the 18-item measure ranged from 1 (“almost never”) to 5 (almost always’)	.82 DBT-A; .88 S&S
EuroQol Group (EQ5D) (Brooks, 1996)	Health-related QoL. Five items scored from 1 (“no problems”) to 3 (“major problems”)	.43 DBT-A; .50 S&S
Carer
The Depression Anxiety Stress Scale-21 (DASS-21) (Lovibond & Lovibond, 1995)	Depression, anxiety and stress. Responses on the 21-item measure range from 0 (“did not apply to me at all”) to 3 (“applied to me very much or most of the time”) over the past 7 days.	Depression (.81) Anxiety (.79) Stress (.85)
Young Person and Carer
Qualitative post-group evaluative questionnaire	Evaluation of group experience, including content, materials, scheduling and logistics of group.	NA

DBT-A, dialectical behaviour therapy-adolescent; QoL, Quality of Life; S&S, Self and Strength.

Statistical analyses

Clinical outcomes were analysed using paired sampled t-tests comparing pre- to post-treatment. Within groups, effect sizes and 95% confidence intervals were calculated using the sample standard deviation of the mean difference at each time point, adjusting for the correlation between measures (Cuijpers et al., 2017). Reliable change indices were calculated to examine the clinical significance of statistical changes using the formula X₂ - X₁/SE_diff, where SE_diff = √(2[SD₁√1-r]²), where SD₁ represents the standard deviation at the earliest observation and r is the internal reliability correlation for the current sample (Jacobson & Truax, 1991). Changes greater than 1.96 times the SE_diff are considered to indicate reliable change within 95% confidence.

Framework

RE-AIM is a planning and evaluation model which addresses dimensions of programme impact and sustainability. Table 2 presents the main outcomes for each RE-AIM dimension.

Table 2. RE-AIM framework categories.

	Description	Measure
Reach	Absolute number, proportion and representativeness of individuals willing to participate in initiative	Patient flow Number of group Participants Group uptake Treatment Completion
Effectiveness	Impact of intervention on outcomes, including negative effects and QoL	Young person clinical outcomes ISAS SIDAS DERS-SF EQ5D Support person clinical outcomes DASS-21
Adoption	Absolute number, proportion and representativeness of settings and intervention ages who are willing to initiate a programme	Staff perception
Implementation	Intervention agents’ fidelity to various elements of intervention protocol	Barriers to implementation
Maintenance	Extent to which programme or policy becomes part of organisation polices/practices; intention and ability to continue with group programmes	Service commitment Acceptability

DASS-21, The Depression Anxiety Stress Scale-21; DERS-SF, The Disorders of Emotion Regulation Scale – Short Form; EQ5D, EuroQol Group; ISAS, Inventory of Statements About Self-Injury; QoL, Quality of Life; SIDAS, The Suicidal Ideation Attributes Scale.

Results

Reach

Patient flow

Over the 12-month recruitment period 173 clients were referred to the service of which 109 clients (63%) were potentially suitable for one or both groups (4 [3.7%] were external referrals). Upon assessment, 67 (39%) were subsequently deemed eligible (Figure 1) and 49% completed baseline measures and commenced DBT-A; while 50% completed baseline measures, but only 25% commenced treatment for S&S. The current study reports on the 13 young person-parent pairs (92% female young people) that completed the DBT-A group, aged 16–18 years (M = 17.1 years; SD = 0.8 years) and the nine young people (100% female) who completed the S&S group, aged 16–18 years (M = 17.4 years; SD = 0.8 years). Participants in both groups were from areas of relative socioeconomic advantage (Index of Relative Social Advantage and Disadvantage DBT-A = 3.92; S&S = 3.11; range 1 to 5) (Australian Bureau of Statistics, 2018).

Figure 1. Participant flow chart.

BPD tx, borderline personality disorder treatment; DBT-A, dialectical behaviour therapy-adolescent; S&S, Self and Strength.

Group uptake

The relatively low uptake of the group was noted throughout the recruitment process. Reasons for participant ineligibility were sourced informally from clinicians and included: residential location outside of service area, no adult available to attend group (DBT-A specific), scheduling clashes, unwillingness to work towards treatment goals and clinician prioritisation of risk stabilisation. Internal research on barriers to group uptake conducted after the conclusion of the present study indicated that most of the youth met at least one exclusion criterion, expressed a preference for individual therapy and were unwilling for parental involvement (Wood & Num, 2023).

Of those 67 participants eligible for a group programme, 52% declined to participate. Primary reasons indicated motivation towards treatment and/or disengaging from the service (i.e., failure to attend appointments or maintain contact with a care coordinator) which accounted for 43% of withdrawals from the group recruitment process and a preference for individual treatment (i.e., treatment as usual) which accounted for 36% of withdrawals.

Treatment completion

Retention

Of those who completed baseline assessment and commenced treatment, completion rates were 71% for DBT-A and 82% for S&S. Within the DBT-A group, one family missed six sessions and opted not to reapply. The remaining seven families that withdrew cited varied reasons (e.g., prioritise work and study commitments; did not enjoy the group dynamic) with no emerging themes. Follow-up surveys were conducted 3 months after group completion and were returned by 21% of participants. Given the small number and low interpretability, results for this timepoint have not been reported.

Adherence

Average attendance for participants enrolled in the DBT-A group was greater than 85%, and average attendance for participants enrolled in the S&S group was greater than 80%. However, no parents attended any of the optional drop-in support sessions for the S&S group. Homework adherence (i.e., practicing skills of previous week) was high for young people and support people in the DBT-A group. Only one young person completed less than 80% of homework, and two support people completed less than 70% of homework. Homework adherence was variable for young people in the S&S group; one-third completed over 80% of homework activities, and one-third completed less than 20%.

Effectiveness

Young people in the DBT-A group demonstrated large within-group improvements in suicidality and emotion dysregulation over the course of the programme, together with moderate improvements in self-harm and smaller improvements in QoL (Table 3). Reliable change indices demonstrated that most participants reported a reliable decrease in emotion dysregulation (N = 11) and suicidality (N = 8). One participant reported an increase in suicidality. Although active self-harm was not an eligibility criterion for the DBT-A group, all but one participant had engaged in self-harm in the previous 12 weeks, and over treatment half reported a decrease of 75% and a further third reported entirely abstaining from self-harm behaviours. Overall, two participants reported a reliable decrease in self-harm. Young people (38%) in the DBT-A group reported reliable increases in QoL over the course of the treatment and there was a small within-group effect for QoL over the course of the group. There was no within-group improvement for carers involved in DBT-A (Table 4). Individually, over half of the carers involved in DBT-A reported a reliable decrease in two or more scores on a measure of depression, anxiety and stress scores. One participant reported reliable increases in both depression and stress, and one-third reported no reliable change in any outcome.

Table 3. Means (standard deviations) and within group effect sizes of standardised measures and reliable change indices for young people.

				RCI
	Baseline	Post-Tx	Within group ES 95% CI	↓ n (%)	↑ n (%)
DBT-A (n = 13)
Suicidality (SIDAS)	5.15 (2.52)	2.32 (1.62)	1.11 (.31 to 1.91)	8 (62%)	1 (8%)
Non-suicidal self-harm (ISAS)	311.72 (353.98)	59.18 (82.11)	.68 (.30 to 1.34)	2 (15%)	0
Emotion Dysregulation (DERS)	3.99 (.33)	2.65 (.76)	1.64 (.74 to 2.53)	11 (85%)	0
Quality of Life Index (EQ5D)	.33 (.15)	.59 (.22)	−1.26 (−2.03 to − .49)	0	5 (38%)
S&S (n = 9)
Suicidality (SIDAS)	6.40 (2.48)	4.24 (3.12)	.7 (−.21 to 1.60)	5 (56%)	1 (11%)
Non-suicidal self-harm (ISAS)	310.00 (233.06)	101.38 (8.55)	.96 (−.01 to 1.93)	3 (33%)	0
Emotion Dysregulation (DERS)	3.90 (.55)	3.28 (.84)	1.01 (−.34 to 2.35)	5 (56%)	0
Quality of Life (EQ5D)	.23 (.22)	.46 (.25)	−.72 (−1.63 to .18)	0	0

Bold indicates significant effect sizes.

CI, condifence interval; ES, effect size; RCI, reliable change index; Tx, treatment.

Table 4. Means (standard deviations) and within group effect sizes of standardised measures and reliable change indices for support people (DBT-A group only: n = 12).

	Effect sizes			RCI
	Baseline	Post-Tx	Within group ES 95% CI	↓ n (%)	↑ n (%)
Depression	.90 (.55)	.72 (.44)	.35 (−.24 to .95)	1 (8%)	1 (8%)
Anxiety	1.01 (.55)	.77 (.51)	.45 (−.16 to 1.06)	3 (25%)	1 (8%)
Stress	.93 (.59)	.68 (.49)	.45 (−.06 to .97)	2 (17%)	0

ES, effect size; RCI, reliable change index; tx, treatment.

In the S&S group, as reported in Table 3, within-group improvements were not observed. Reliable change indices showed that one participant reported an increase in suicidality over the course of the programme. Contrastingly, five participants reported decreased suicidality over the course of the group. Three participants reported a decrease in emotion dysregulation and self-harm. No participants reported an increase in self-harm, with 78% of participants reporting a reduction of self-harm between 59% and 92%. No reliable changes in QoL were reported.

Adoption

Four group facilitators (including two Group Coordinators, roles described below) answered qualitative questions regarding their experiences in group planning and provision. Overall, clinicians provided positive qualitative feedback about training and materials provided to facilitate groups (“The training was comprehensive and intensive”. “I feel the training has increased my skills … ”). Clinicians described the time commitment required as a group facilitator (“Group facilitation is exhaustive(sic) and can be time-consuming”, “ … it did require a significant amount of planning and preparation … ”), with clinicians reportedly preparing for group 5 days ahead of facilitation (e.g., consult group, administration, material preparation) and dedicating approximately 4 hours on the day of group.

Group coordinators reported an impact of group on their clinical duties, especially during low staffing periods (“ … when there was a lack of staff trained to deliver groups, there was less time for these more senior aspects of the role”, “ … having to use additional time to manage clinical duties due to decreased availability (of other clinical staff on the team)”). Group facilitators reported little impact of group facilitation on their other clinical duties, and overall, there was a sense of teamwork and camaraderie amongst clinicians involved in group facilitation (“ … with the support of other team members and the leadership teams, this was managed”, “ … the sense of teamwork in co-facilitating”). Despite the time commitment to training and facilitating groups, clinicians reported a positive experience of group facilitation (“Facilitating the groups was the most enjoyable part of my week”. “I enjoyed the facilitation … ”). Clinicians also noted a positive clinical impact (“groups can increase clinical output for teams with less clinical duty demand”, “ … increased my effectiveness, competence, and confidence as a clinician … ”, “ … provided me with advanced skills … ”). Finally, clinicians observed an overall positive impact on clients (“I was able to see a reduction in some of the presenting problems … ”, “ … received more positive feedback and thanks from participants in the group program than any other program or intervention I have been in involved in … ”, “Some have even said that the program has saved their lives … ”).

Implementation

Sessions delivered

No session for either programme was cancelled or rescheduled due to staffing or organisation reasons. Groups were maintained in a hybrid format (online/face to face) during COVID-19 health restrictions and held online during lockdowns.

Barriers

Over the course of the groups, four barriers to implementation arose. The first was low group uptake (presented in Reach). The second was related to staffing continuity at the service, with high turnover of staff and gaps in staffing identified throughout the project timeframe. Third was the service model of care, which also provided ongoing options for individual treatment. Finally, there were ongoing disruptions related to COVID-19.

Staffing continuity

Throughout the evaluation, issues relating to staff continuity impacted the project implementation resulting in periodical holds and restrictions on referrals to the service due to under-staffing. Fundamentally, this impacted the number of referrals accepted into the service, and subsequently the number of clients that could have been eligible for groups. Additionally, the service had a dedicated Group Coordinator, whose role was to implement, deliver, and monitor the ongoing group programmes. This included group-related intake assessments, treatment and care planning and reviews, leading group therapeutic interventions and liaising with internal and external service providers. It was difficult to continuously staff the Group Coordinator position, which limited the team’s capacity to administer groups at an optimal level.

Low staff and high clinical staff turnover more broadly had a critical impact on implementation throughout the project. The expert nature of the DBT-A group requires clinical staff to initially observe a full course of DBT-A and then undertake 80 h of specialised training over 16 weeks. The S&S group requires clinicians to have post-graduate training in delivering evidence-based therapy intervention (e.g., Master of Cognitive Behaviour Therapy, Master of Psychology) and to be specifically trained in the S&S programme. This in turn impacted the scheduling of both groups, which was cited as a barrier to group attendance.

Service model of care

Until the commencement of the group interventions, the service model of care was centred on individual intervention, with a clinical focus on one-on-one sessions with clients and caregivers tailored to target symptomatology, therapeutic processing, and increased coping strategies. This initial focus on individual interventions may have impacted recruitment into programmes, potentially due to limited awareness and understanding of group programmes by clinicians who were not trained facilitators or client preference for working individually.

COVID-19

The implementation of the DBT-A and S&S groups was affected by COVID-19 health directives (October 2020 to September 2021), including state-wide lockdowns in November 2020 and July 2021.

Maintenance

Service commitment

Following the 12-month recruitment period (October 2020-September 2021), the SYMHS continued to run the DBT-A and S&S groups. Since October 2021, 141 individuals have commenced groups, with 24 individuals completing DBT-A and 7 completing S&S. SYMHS was also successful in obtaining additional funding to continue to coordinate and deliver the programmes. Subsequently, the team had applied to make the Group Coordinator role a permanent role within SYMHS once grant funding ended to ensure the groups operate. There has been significant in-kind support from SYMHS clinicians, including incorporating training in therapeutic modalities.

As of March 2023, the SYMHS runs two weekly DBT-A groups with rolling entries and S&S groups on an as needed basis. Seven clinicians have been trained to deliver DBT-A, and nine have been trained to deliver S&S, increasing the capacity to deliver both group programmes. Both groups have additional staff undertaking training.

Acceptability

Feedback on the programmes was provided by treatment completers only. Treatment completers in both groups provided positive feedback. For DBT-A, treatment completers (N = 13) liked the size of the group (100%), materials (100%) and content (100%). Clinicians and group coordinators also indicated that the materials were appropriate and well structured (“ … the materials seemed appropriate in content and complexity for the client group”). A few participants reported a preference for session times outside of school/work hours. All responders reported that they found the groups helpful and that they would recommend them to others. Both young people and parents in the DBT-A group provided positive qualitative feedback regarding the involvement of parents in session, with some young people and adults specifically referring to improvements in the relationship. Young people who completed the S&S group also indicated that the group had been beneficial. Overall, qualitative feedback from graduation sessions was overwhelmingly positive from both adolescents and adults.

DBT-A participants who provided feedback rated their learning, the likelihood of using the skills learnt and their child’s learning (parents only) highly (8–10). Young people in the S&S group rated learning highly (7–8) and likelihood of using skills as medium to high (6–10).

Discussion

This study examined the implementation of DBT-A and S&S groups delivered in a youth community-based mental health care service, suggesting that group interventions for emotion regulation in community mental health settings may be effective in reducing self-harm and suicidal behaviours. Initial reach of the programmes was good; however, the actual uptake of the group programmes was poor. For individuals who engaged, results indicate clinical effectiveness of the DBT-A group, and, despite the small sample size, reliable changes also suggested positive outcomes for the S&S group. The adoption of the programmes by the clinicians was favourable. While barriers to implementation were identified, maintenance of the group programmes continues given high acceptability by clients and ongoing service commitment.

Although initial reach of the programme was broad, only 30% of those approved for clinical review were eligible to complete group therapy. Barriers for implementation of DBT programmes and access to group treatment in community settings can be classified as organisational challenges and client barriers (Carmel et al., 2014; Leeuwerik et al., 2022). Two organisational challenges may have impacted the program’s reach of both groups. First, the established model of care in the SYMHS focussed on individual treatment as opposed to a dedicated treatment pathway for group therapy. This may have impacted the number of individuals being directed through to group programmes, which could have minimised the therapeutic modalities which were delivered as a group as a treatment option due to competing therapeutic priorities (Carmel et al., 2014). Second, low staff numbers and high staff turnover impacted periods of referral intake and consequently the number of individuals being referred through the service. Further research into organisational approaches to overcome these challenges and implement solutions may assist in the reach of similar programmes in the future (Flynn et al., 2019; Swales et al., 2012; Swenson et al., 2002).

Identified client barriers to group therapy include group processes (e.g., fear of sharing, belonging and accountability, continuity, individual attention and therapist support), general therapy themes (e.g., understanding how to overcome a specific mental health condition, personal relevance of treatment), attitudes towards treatment and extra-therapy factors (e.g., personal circumstances) (Leeuwerik et al., 2022). Similar client barriers have been identified for DBT, along with time-commitment (e.g., 6-month treatment), focusing on a “life worth living” and an explicit commitment to change problematic behaviours (Swenson et al., 2002). Programmes similar to S&S have reported client barriers including lack of agency for participants, severity of depression and recency of crisis (Sayal et al., 2019). However, these barriers have largely been established in adult populations. Further understanding of client barriers and implementation of solutions in the youth-focused community mental health settings may increase the reach of similar group therapy programmes.

Acceptance of the treatment programmes by clients was moderate, although this needs to be considered with caution due to the small sample sizes. A recent systematic review and meta-analysis of DBT-A programmes reported a 7% average drop-out rate, ranging between 0% and 51% (Kothgassner et al., 2021). The DBT-A drop-out rate in the current study (29%) was within this range although higher than the average reported by Kothgassner et al. (2021), which may reflect hospitalisations or changing availability of clients (Kothgassner et al., 2021). CBT-specific interventions for self-harm in youth are less commonly reported in the literature, with several programmes taking a combined approach with other therapy modalities or focusing on specific population groups (e.g., inpatient) (Ougrin et al., 2015). A systematic review of self-harm interventions for low- and middle-income countries reported drop-out rates for CBT programmes to vary between 0% and 69.5% (Aggarwal et al., 2021). The S&S drop-out rate in the current study (18%) fell into previously identified ranges. Documented evidence of reasons for drop-out was not collected, however anecdotal feedback indicated issues with client scheduling and attendance criteria.

While the small sample sizes across the two groups limit conclusions that can be drawn regarding effectiveness, youth participants showed promising improvements in their levels of suicidal ideation and risk, emotion regulation and self-harm over the course of the DBT-A group. This is consistent with a meta-analysis and systematic review of DBT-A which has shown consistent small to moderate reductions in self-harm and suicidal ideation (Kothgassner et al., 2021). While research has indicated that when evidence-based mental health treatments result in improvements in youth functioning, their parents’ psychosocial functioning may also improve (Kazdin & Wassell, 2000), we saw no improvements to parental function in the current study.

Improvement in the S&S groups was less marked with only clinical improvement for suicidality, emotion dysregulation and self-harm. S&S is a lower intensity programme, with less than half the therapeutic input and resources than DBT-A. Whilst CBT-based interventions (such as S&S) have been effective, DBT appears to be more effective in reducing suicidality and self-harm. DBT was specifically designed to address the needs of individuals with borderline personality disorder (BPD), such as the emotion dysregulation and interpersonal difficulties, which are characteristics of BPD (Linehan et al., 2006). In contrast, CBT-based interventions may not address these issues directly. It may be that the increased emphasis on regulating intense emotions, tolerating distress and improving interpersonal relationships may increase the effectiveness of this treatment (Linehan et al., 2006). In the present study, two participants commenced the DBT-A group after successfully completing the S&S group. Lower-intensity groups like S&S may present an opportunity to socialise youth with groups and encourage transition to DBT-A if families are willing and able; however, these conclusions need to be considered with caution due to the small sample size.

Regarding adoption, clinicians reported positive impacts through their participation as group facilitators and observed a positive impact on clients. Staffing continuity issues were noted, with low staffing and the Group Coordinator role periodically vacant, impacting the service’s ability to develop and implement groups and other service opportunities. The Group Coordinator role included intake and discharge clinical review documentation, as well as communicating with external care teams. When this role was vacant, these duties were shared amongst existing team members, which increased external client care coordination and increased time spent on groups by other staff. Minimum requirements for one group required two staff facilitators, each dedicating approximately 4 hours per week. Given that these group programmes typically demand specifically trained staff, effective implementation of group programmes requires careful consideration of resource training and allocation, particularly accounting for staff leave and turnover. Individual therapy was the focus in the service model of care initially. The impact of COVID-19 throughout the project was also noted, with hybrid and online groups offered when required due to public health recommendations. Staffing is an ongoing issue across healthcare generally and in mental healthcare, which was exacerbated by the COVID-19 pandemic (Johnson et al., 2018; Walton et al., 2020).

This project was delivered over a 12-month period, not including post-project evaluation. While implementation was challenging, the continued commitment of the SYMHS to the groups and ongoing client group completions post-project suggest that these groups continue to fill a treatment gap for youth with emotion regulation difficulties, with the DBT-A group providing the greatest improvements in family functioning.

Limitations

This implementation study provides unique information on the delivery of group programmes for adolescents in a community mental health setting; however, findings should be interpreted in the context of the following limitations. First, the small sample size across both groups limits the conclusions that can be drawn both from a clinical and implementation perspective. Second, staffing continuity may have impacted potential participant uptake. Consequently, the findings reported here may underestimate the real-world sustainability and success of similar group programmes. Third, the clinical effectiveness data refer to the final week of treatment. This reflects practical limitations of achieving acceptable response rates with longer and less frequent questionnaires compared with shorter and briefer questionnaires (Johnson et al., 2022; Radunz et al., 2021). Finally, treatment effectiveness has likely been over-estimated given the poor treatment uptake and the estimation of treatment effects only in patients fully completing treatment.

Future directions for research

Group therapy provides therapeutic advantages such as a sense of universality and connectedness and increased learning opportunities from other group members (Kealy & Kongerslev, 2022). However, it is important to note that whilst these benefits are clear to clinicians, clients may not be aware of or appreciate them. Despite these benefits, there may be a difference between individuals that elect to engage in group therapy and individuals that decline involvement in group therapy. Evaluation of different approaches to tackling specific barriers to group therapy engagement (e.g., psychoeducation, using testimonials of previous participants from their graduations) could provide insight into strategies to increase group programme involvement. Additionally, due to positive experience of clinicians, future research could investigate group training opportunities for workforce recruitment and retention. Self-harming and suicide have a high economic cost as well as a high personal and social cost. Further investigation into the cost-effectiveness of group interventions such as DBT-A and S&S compared to treatment as usual would aid in optimising health resource allocation. Finally, the reason for individuals not engaging in group therapy is a research gap in which understanding may assist with the increased uptake of group programmes.

Acknowledgments

We would like to acknowledge funding from the James and Diana Ramsay Foundation and Breakthrough Mental Health Research Foundation. We would also like to acknowledge the effort and support of the clinicians and leadership at Youth Mental Health Services, including Group Facilitators and Group Coordinators. Finally, we would like to acknowledge the clients and their families for taking part in this research.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The deidentified data that support the findings of this study are available from the corresponding author upon reasonable request.

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Additional information

Funding

This work was supported by the James and Diana Ramsay Foundation and Breakthrough Mental Health Research Foundation.