Abstract
Children living in Defence families affected by parental mental illness face increased stressors, pressure for self-sufficiency and potential exposure to adverse outcomes, including future mental illness. Little is known about the impact of mental health early interventions on the help-seeking tendencies of military young people affected by parental mental illness. Similarly, the responsivity of this cohort to peer-based mental health literacy interventions aimed to dispel misconceptions and stigmatising beliefs about mental illness and help-seeking is not well understood. The current study measured key aspects of belief-related mental health knowledge and intentions to seek help from a range of formal and informal sources of 236 Australian young people aged 7–18 years, living with parents affected by military-associated mental health problems. Findings suggest that young people in the Defence community are most likely to seek help from their parents and show changes in beliefs about mental illness and intentions to seek help from some help sources following a brief (2-hour) group intervention. Help-seeking intentions from telephone helplines increased for the cohort, and males were more inclined to seek help from a friend after intervention.
Limitations including the absence of a comparison or control group, issues in data measurement protocols and a need for long-term follow-up were future research directions.
Keywords: Youth; Mental health; Stigma; Help-seeking
Introduction
Young people living in military families face novel complexities compared to those experienced by most of their peers.1 These include adjustment to frequent relocations, disruption to friendships, education and community networks, reduced access to parenting resources of the non-deployed parent and stress among family relationships.2,3 Within this context, attributes such as self-sufficiency and toughness are desirable for military-connected young people (and their parents)4 and are reflected in stigmatised beliefs regarding mental illness that are frequently cited as barriers to help-seeking within the Defence community.5
Unsurprisingly, mental health service usage rates by military-connected individuals in need of support can be as low as 15%.6.7 Prior unhelpful experiences with support providers who showed a limited understanding of military-specific issues and difficulty maintaining connection and trust with supports as a consequence of relocation are factors that further negate mental health help- seeking by those within the Defence community.5 Reducing stigma and other barriers to help-seeking and increasing access to adequate mental health supports remain a key priority of the Australian Defence Force (ADF).7,8,9
Strengths and resilience factors are noted for young people living in military-connected families amid an overall increased risk of mental health and other adverse outcomes.10 The risks are compounded for young people living with parents who experience a military-associated mental illness.10,4 In the general community, parental mental illness affects about one in five children.11 Recent data collected in relation to ADF families suggest a 12-month prevalence of mental illness in recently transitioned (ex-serving) personnel of 46%.12 Post-traumatic stress disorder (PTSD) is a complex mental health condition specifically noted within the military context, frequently occurring with one or more other mental health conditions.13
Children who experience parental mental illness face an increased risk of social isolation, disrupted parent–child relationships, carer’s responsibilities, academic difficulties, emotional and behavioural issues, and a higher likelihood of personal mental illness.11,13 Therefore, the provision of programs and supports aiming to reduce these risks by intervening early and effectively is a necessary endeavour.14 A systematic review regarding the impact of mental health early interventions for young people affected by parental mental illness reported a 40% reduction of risks for those receiving interventions, including reduced evidence of internalising and externalising problems compared to young people who did not receive intervention.15
According to Jorm and colleagues, the concept of mental health literacy is a key aspect of effective mental health early intervention.16 Mental health literacy refers to an individual’s ability to recognise mental health problems and maintain helpful attitudes and skills to enable effective help-seeking.16 Previous research has indicated those with higher levels of mental health literacy are more likely to seek help early and appropriately for emerging mental health issues.17 Help-seeking is described as an active form of coping involving communication with others in response to distress or a perceived problem, with the aim of recovery or reducing the problem.18
Interventions delivered to young people across the past decade have integrated psychoeducation and peer-group formats endorsed by program participant groups.19 One example of a peer intervention is the CHAMPS program that has been delivered in Australia.20 There is a paucity of research relating to the impacts of such interventions for military- connected young people affected by parental mental illness who present unique and complex intervention needs as a cohort. It is plausible that the barriers to engagement with mental health early intervention are magnified for military-connected young people, given the specific stigma and related factors outlined above. Extending current peer-group programs by grouping participants with peers who share a Defence community background and embedding the interventions within a recreation-camp format may support greater engagement of this hard-to-reach cohort of young people.
The current study was exploratory and sought to gather feedback on the help-seeking intentions of young people (aged 7–18 years) whose parents experience military-service associated mental illness collected before and after intervention. Additionally, we explored whether young people from a designated military cohort experience changes in mental health knowledge related to myths and misconceptions about mental illness by comparing pre- and post- intervention feedback. We hypothesised there would be improvements in mental health knowledge and help-seeking intentions from pre- to post-camp assessments. Gender effects were assessed for changes in help-seeking intentions, with positive changes hypothesised as less likely for males who are reluctant help-seekers in general21 and potentially more so within the context of self-sufficiency noted in Defence communities.4,5
Materials and methods
Participants
The sample comprised 236 young people aged 7–18 years (M = 11.59, SD = 2.48) from four Australian states/territories (NSW 64.2%, QLD12.1%, ACT 17.7%, and NT 6.0%). All were participants in the Australian Kookaburra Kids Foundation (AKKF) ‘Defence Kids’ 2-day camp program between February 2017 and December 2019. Young people from ex- serving (44%) and serving (56%) Defence families were referred by their parents to the Defence Kids program. Invitation to participate in the evaluation research was voluntary, with participation having no bearing on program inclusion or access to any AKKF service or support. Parent and participant consent prior to commencement was required for inclusion in the study, and individual consent from each young person at the time of data collection. Data were collected from 89% of the program attendees. Proportionally, 59.7% of the sample identified as female, 39.6% male and 0.7% as other. The intervention components were delivered to age- matched participant groups to ensure content was developmentally relevant, with the percentages of participants in each age category presented in Table 1.
Table 1. Participant number and percentage by age category
| Age category (years) | N | Per cent (%) |
|---|---|---|
| 7-9 | 49 | 20.8 |
| 10-12 | 84 | 35.6 |
| 13-15 | 58 | 24.6 |
| 16-18 | 15 | 6.4 |
| Missing | 30 | 12.7 |
Design and measures
The study received ethical approval from the Department of Defence and Veterans’ Affairs Human Research Ethics Committee (ID 028-18) and adopted a pre-post design investigating variables of mental health knowledge and help-seeking intentions. The measures were selected based on evidence of their use in previous research. In the case of the mental health knowledge measure (Children’s Knowledge Scale of Mental Illness, 23), the measure was derived from specific feedback from young people living in families affected by parental mental illness regarding unhelpful stigmatising myths and misconceptions of mental illness, therefore explicitly deemed relevant to the current study. Measures were scored as per published methods, with higher scores representing more of the measured attribute. Participant experience items, including satisfaction with the program and intent to continue, were also administered. Table 2 provides descriptions and references for the study measures (excluding client experience items).
Table 2. Study measures, references and internal consistency (Cronbach’s Alpha) values
| Dependent variable | Measure | Description and information |
|---|---|---|
| Mental health knowledge | Children’s Knowledge Scale of Mental Illness18 | 7-item measure with response format (true, false, don’t know) tapping young people’s knowledge of unhelpful myths and misconceptions experienced by children whose parents have a mental illness. Sample items include, ‘A mental illness can be caught by the flu’, or ‘Many parents get a mental illness because of the way their children behave’. Authors: Grove, Reupert & Maybery, 2015. |
| Help-seeking intentions | General Help-Seeking Questionnaire19 | An 11-item scale asking respondents, ‘If you were having a personal or emotional problem, how likely is it that you would seek help from the following people?’ Responses are made using a 7-point scale from ‘extremely unlikely’ to ‘extremely likely’. Used extensively > 10 research projects investigating helpseeking intentions. Authors Wilson, Deane, Ciarrochi & Rickwood, 2005. |
Table 3 Overview of the aim, methods and sample differentiations within the manualised Chat Group early intervention component of Defence Kids
| Therapeutic aim/intent | Intervention/method | Differentiation samples |
|---|---|---|
| Access to a safe, relatable peer social context Opportunity for supported and successful peer interaction Proactive and scaffolded opportunity to communicate about mental health and wellbeing Provision of developmentally relevant knowledge and skills-building opportunities Opportunity for exploration and growth in healthy helpseeking attitudes and behaviours Durable and ongoing opportunity to ‘subclinical’ mental health support, triage and referralon opportun |
Didactic and interactional teaching (psycho-education) Direct instruction Skills modelling Social narratives Scripting Cognitive–behavioural interventions Self-management Peer-mediated instruction and intervention Reinforcement Clinical triage In-vivo clinical support (as required) |
Activity: ‘Identifying signs and symptoms of mental illnesses’. In this activity, young people learn about symptoms associated with a range of mental illnesses and apply their knowledge via an interactive activity. It utilises psychoeducation (didactic and interactional teaching), peermediated instruction and intervention, and reinforcement. To apply and demonstrate their knowledge, young people listen to a symptom/characteristic and move to a placeholder label (e.g., depression) to indicate their response. Questions (peer-led) and conversation to clarify and deepen understanding is facilitated around the response activity. Differentiation 1: Males Increased activity and physicalisation is provided by the differentiation. Individuals each stand at a placeholder. A ball is provided to one participant, when the symptom/characteristic is described the ball-holder throws (or rolls) the ball to the person standing at the appropriate place. Differentiation 2: Developmental stage (age) Individuals from older age groups tend to engage less with this active task than younger participants. An option to allow interactional learning but reduce physical mobility is to equip participants with ‘true’/’false’ cards or flags. The facilitator reads out symptom/characteristic and raises a placeholder label. The participating young people signal either ‘true’ or ‘fa |
Procedure
Participants were administered the study measures shortly after arriving at the Defence Kids camp and againattheendofcamp. Theinterventionwasdelivered in two 1-hour long ‘Chat Groups’ on Saturday and Sunday morning, comprising two trained facilitators and age and gender-matched participants in groups of 8–10. Participants undertook a range of other social and recreational activities (e.g., canoeing, art/crafts, giant rope swing, meals) while grouped with the same peers and facilitators from their Chat Group. Interventions were delivered by trained AKKF personnel who work from standardised Chat Group Facilitator Manuals with delivery methods adapted to match the age group of participants. Facilitators are AKKF employees, volunteers and mental health qualified clinical consultants aged 21 years who have received formal program induction and training, and are eligible to work with children (i.e., have successfully completed working with children and police checks). Participants work from and receive a take-home resource (‘Kookabook’) adapted to the developmental stage and aimed at promoting intervention consistency and generalisability. Chat Group content and activities comprised evidence- based mental health literacy components including psychoeducation and skills-building to promote active help-seeking. Table 3 details the aims, approaches and examples of differentiation of early intervention components in Chat Groups.
Onsite support to participants, including proactive check-ins with Chat Groups, in-the-moment support to manage emotional distress and triage acute needs, and liaison to support referral of any ongoing mental health needs, were provided for the duration of the camp program by the mental health qualified clinical consultants. Across the programs described here, targeted support by clinical consultants was provided to 16% of participants.
Table 4. Pre- and post-test means and rank order for mental health knowledge and help seeking intentions from different sources
| Pre-test | Post-test | ||||||
|---|---|---|---|---|---|---|---|
| Variable | Mean | Rank | SEM | Mean | Rank | SEM | n |
| Mental health knowledge | 4.79 | nab | 0.18 | 5.89 | na | 0.13 | 123 |
| Intentions to seek help | |||||||
| Friend | 4.78 | 2 | 0.17 | 4.66 | 2 | 0.16 | 111 |
| Parent | 5.20 | 1 | 0.19 | 5.26 | 1 | 0.18 | 110 |
| Other relative | 4.57 | 3 | 0.17 | 4.23 | 4 | 0.18 | 107 |
| Teacher | 3.45 | 6.5 | 0.18 | 3.24 | 8 | 0.18 | 109 |
| Other adult (e.g. coach) | 3.22 | 8 | 0.16 | 3.14 | 9 | 0.15 | 112 |
| School welfare team | 3.54 | 5 | 0.17 | 3.33 | 7 | 0.17 | 111 |
| Mental health professional | 4.16 | 4 | 0.19 | 4.24 | 3 | 0.16 | 112 |
| Phone helpline | 3.06 | 9 | 0.17 | 3.82 | 5 | 0.17 | 112 |
| Doctor/GP | 3.45 | 6.5 | 0.19 | 3.54 | 6 | 0.17 | 101 |
| Religious leader/chaplain | 2.25 | 10 | 0.16 | 2.29 | 10 | 0.15 | 111 |
| No onea | 2.53 | na | 0.19 | 2.57 | na | 0.18 | 108 |
| Note: Minor variation in sample sizes for each variable due to missing data resulting ranging from n = 101 to n= 123, a indicates reverse scored item with scores closer to 1 indicating higher help seeking intentions b “na” indicates not applicable. |
|||||||
Results
Data from hand-written questionnaires were entered and analysed using IBM SPSS. Participant experience feedback indicated 89% experienced the Defence Kids program as ‘good’ or ‘great’, with 90% endorsing ‘yes’ to return to the program and likely to recommend it to a friend or someone else with similar needs.
Items from the Children’s Mental Health Knowledge Scale were recoded to reflect ‘correct’ or ‘incorrect’ responses, with scores of 7 indicating correct responses on all items. Mean scores for total mental health knowledge were created for pre- and post- responses and each help-seeking item. Paired samples t-tests were used to determine whether hypothesised changes in dependent variables were evident. Table 4 presents the mean scores before and after intervention along with findings of paired t-tests for significance. There was a significant increase in mental health knowledge from pre- to post-test, t(123) = -5.90, p <.01.
A 2 (pre-post) by 2 (gender) mixed ANOVA with age entered as a covariate was calculated for all intentions items. Given that only one participant indicated they identified as neither male nor female, this case was excluded from the analysis. There were also four cases for which data for gender were missing. Prior to analyses, assumptions were tested. There were only two significant ANOVAs; for intentions to seek help from friends and a helpline. Intentions to seek help from ‘friends’ had a moderate negative skew at both pre- and post-test time points. The variable was squared, resulting in the transformed variable meeting the normality assumption. Intentions to seek help from a helpline were normally distributed at both time points and did not require transformation. The results for intentions to seek help from friends were highly similar for the transformed and untransformed variables, so results from the untransformed variables are reported for ease of interpretation. For both analyses Box’s test was nonsignificant, and both met the assumption of sphericity (Bartlett’s test). Levene’s test was also nonsignificant, indicating analyses met the assumption of homogeneity of variances.
There was a significant time by gender interaction for intentions to seek help from a friend, F(1101) = 9.88, p =.002. Means and standard deviations are provided in Table 5. This indicates that males increased their intentions over time while females decreased their intentions to seek help from friends over time.
For intentions to seek help from a helpline there was no significant interaction effect (F(1102) = 0.99, p =.32) but there was a significant main effect for time (F = (1, 102) = 14.46, p <.001). This indicated a significant increase in intentions to seek help from a helpline over time.
There were no other statistically significant effects on intentions to seek help for any of the other eight help sources.
In theory, an increase in mental health knowledge may contribute to increased help-seeking intentions. In order to explore the relationship between changes in mental health knowledge and changes in help-seeking intentions, change scores were calculated by subtracting the pre-test scores from post-test scores for total mental health knowledge measure and for help-seeking intentions from each source (and no- one). Only one of the 11 Spearman’s correlations was significant between changes in mental health knowledge and intentions to seek help from a school welfare officer or school nurse (r(110) =.27, p =.002 [1-tailed]). This is a small to moderate positive relationship indicating that improvements in mental health knowledge are associated with increases in help-seeking intentions from school welfare officers or school nurses.
Table 5. Means and standard deviations for help seeking intentions which were significant in time by gender ANOVAs
| Pre-test | Post-test | ||||
|---|---|---|---|---|---|
| Help-seeking intentions | N | Mean | SD | Mean | SD |
| Friends | |||||
| Males | 41 | 4.27 | 1.88 | 4.83 | 1.65 |
| Females | 63 | 4.98 | 1.72 | 4.56 | 1.75 |
| Helpline | |||||
| Males | 41 | 3.12 | 1.76 | 4.12 | 1.78 |
Discussion
Support was found for hypothesised improvements in mental health knowledge throughout the peer support group. The mental health knowledge measure captures the capacity of young people to correctly identify misconceptions and unhelpful beliefs about mental illness known to be relevant to this cohort (i.e., children whose parents are living with mental illness). This change occurred over a relatively brief mental health intervention (i.e., 2 x 1-hour sessions). Without a control group, we cannot causally attribute such changes to the intervention; however, these initial correlational findings are promising.
Potentially, such changes in this element of mental health literacy could positively impact the developmental and psychological trajectory of young people in this cohort who may have otherwise retained inaccurate beliefs such as ‘many parents get a mental illness because of how their children behave’ or ‘I am probably the only person in my school who has a parent with a mental illness’. Previous research, including economic modelling research on mental health early interventions, suggests the impact of these changed beliefs may be substantial and lifelong for the young people themselves and the broader community.15,18
Our results indicate that young people living with parents affected by military-associated mental health problems are most likely to seek help from their parents, followed by friends, other relatives and then a mental health professional. This order appears to remain relatively stable from pre- to post- test. Previous research has highlighted Defence communities’ close-knit and self-sufficient nature due to active service challenges. The findings highlight the critical importance of supporting positive help- seeking attitudes among military parents and the need for adequate resources to equip these parents in understanding and supporting their children in times of need. While generalisability and familial information sharing are indirectly supported in the Defence Kids program, the findings demonstrate considerable scope for holistic family interventions, as highlighted in previous research.7,10
Reliable changes in help-seeking intentions following intervention were evident for only two help-seeking sources. The first was for telephone helplines (such as Kids Helpline). There was no interaction effect for gender, with result patterns suggesting both males and females increased their intentions to seek help from this source. Although these changes cannot be causally attributed to the intervention, it is notable that one of the key Chat Group activities related to help-seeking focused on calling a helpline. This activity involved a real-time demonstration of placing a call to Kids Helpline. The skills demonstration, modelling and practice aimed to increase participants’ generalisability and the likelihood of positive behavioural change. This result reminds us of the merit in skills demonstration and role-plays, which may be integrated with other aspects of Chat Group or similar early interventions for greater effect.
There was a significant time by gender interaction for intentions to seek help from a friend while controlling for the effects of age. Males have previously been indicated to deny the presence of mental health problems and feel compelled to keep problems to themselves.25 In this study, males were significantly more inclined to seek help from a friend after the intervention than they were before. In contrast, females were slightly less likely to seek help from friends (but not significantly so when age was controlled). This finding may highlight the importance of cohort-specific peer groups in allowing males to gain confidence to discuss mental health and other topics that can be experienced as ‘personal’. By sharing ideas and skill-building in a safe and relatable group, male participants may feel more confident and able to share with their ‘everyday’ friends. However, study limitations prevent more definitive conclusions regarding mechanisms of change and signal the need for further qualitative and follow-up research.
This research provides valuable preliminary insights into the impact of mental health literacy interventions and help-seeking intentions on a group of young people previously not studied. However, there are notable limitations in the current research, additional to those noted above. Data collection was hindered by a pen and paper methodology that may be burdensome for participants and lack suitability for the diverse ages of participants. A significant issue regarding the appropriateness of the standardised measures for the younger children must be acknowledged. While all data was screened prior to manual entry and analysis, it is reasonable to acknowledge that younger participants did not have an equitable opportunity to present their feedback due to this limitation. Improvement of data collection protocol and processes is highlighted for future directions.
The current study is a real-world application of a mental health promotion program aimed at a target group of young people and, therefore, focused primarily on ensuring intervention delivery. The absence of a comparison or control group is a notable limitation of this study that needs to be addressed in future research (e.g., via a quasi-experimental replication of the current research, possibly utilising a time-delayed method). In addition, no information about the long-term impacts of the intervention is known, which highlights the need for longitudinal follow-up that addresses changes to actual help-seeking behaviour. Future research addressing these issues will ensure continued progress in understanding and supporting the mental health early intervention needs of young people affected by military-associated parental mental health concerns.
Corresponding Author: Virginia Williams, virginia.williams@kookaburrakids.org.au
Authors: V Williams1, F P Deane2,C Giles1, L Knight1
Author Affiliations:
1 Australian Kookaburra Kids Foundation – Research
2 University of Wollongong