Enhancing an online cognitive behavioural therapy intervention for depression: Harnessing the feedback of sexual and gender minority youth to help improve SPARX

Objective SPARX is an online cognitive behavioural therapy self-help intervention for adolescent depression provided in serious game format. Since 2014, it has been freely available in Aotearoa New Zealand (NZ) due to funding from the NZ government. In 2020/21, feedback from sexual and gender minority youth (SGMY) was used to refine and update SPARX. Method Three online focus groups and follow-up email consultations involved 12 SGMY (16 to 25 years old) in NZ. A general inductive approach was used to analyse data. Results SGMY had specific needs as well as preferences and four themes were identified: attend to our contextual realities; portrayals of sexual and gender minority people in games; envisaged ideals for serious gaming and appraisals of SPARX. SGMY feedback was used to improve SPARX for this unique population, with the updates launched in October 2021. Conclusions SGMY are underserved in terms of their mental health needs. Refining or tailoring existing interventions proffers a potential way forward in terms of addressing these needs.

S exual minority (e.g. lesbian, gay and bisexual) youth are at an increased risk of depression, 1 with gender minority (e.g. transgender/trans) youth having particularly high mental health needs. 2 Unfortunately, sexual and gender minority youth (SGMY) are frequently stigmatised and mistreated and, as a result, they are more likely to experience compromised mental health. 3 Limited intervention options specific to SGMY are available; 4 however, prior research indicates that these young people value the online delivery of psychological supports for their mental health. 4 One such online intervention, SPARX (Smart, Positive, Active, Realistic, X-Factor thoughts) is provided in serious game format; it has been used by over 9000 adolescents and evaluated amongst SMGY for almost a decade. 2,5 Serious games (in brief, digital games for health, education or another 'serious' purpose) may be useful for increasing the reach and impact of online interventions given the popularity and non-threatening nature of games. 6 However, results to date indicate that this seven-module online cognitive behavioural therapy intervention should be refined with SGMY in mind. 4,5 SPARX has been made freely available in Aotearoa New Zealand (NZ) since 2014 with 2.3% of its users (n = 207) identifying as transgender. 2 In 2020, funding from the NZ Ministry of Health was obtained to update the intervention. As part of the co-design work associated with this upgrade, we obtained additional funding and sought the input of SGMY. In this small qualitative study, we asked SGMY to draw upon their experiences of SPARX, and computer games more generally, to consider: · In which ways should a serious game for mental health, like SPARX, be refined so that it better meets the needs of SGMY?

Methods
The consolidated criteria for reporting qualitative research (COREQ) 7 has informed this study's write-up.

Sample and data collection
Participants were recruited via a staff member that was known to them from a SGMY organisation in NZ, whereby staff posted online messages about the study and those young people that expressed an interest in it were sent additional information. This information included online promotional material about the study, and this was sent to those aged 16-25 years old that were keen to take part. The participants were from one of two SGMY organisations based in major cities. However, they came from a range of locations in the North Island of New Zealand.

Ethics and consent
Ethics approval was granted by the Health and Disability Ethics Committee in NZ. All participants provided written informed consent prior to taking part in the study.

Focus groups
Three online focus groups (N = 12) consisted of ten participants aged 16-19 years and two participants aged 20-25 years; all participants were SGMY (see Table 1). Focus groups comprised of the participants (in private spaces in various locations) and ML, who facilitated all the sessions. These groups lasted between 59 and 65 minutes. A semi-structured format (questions available from the corresponding author) was used. Initial questions (and prompts) focused on participants' experiences of games generally and then (after a short demonstration of SPARX) questions about this serious game, for example: The facilitator (ML), who is based in the United Kingdom, is an academic experienced in clinical and group-based youth mental health work. The participants had not previously met ML, but they were aware that he was a co-developer of SPARX. They also knew ML is a gay male and gender role non-conformer, and participants seemed at ease knowing this about him. Focus groups were audio-recorded and professionally transcribed. Basic field notes were taken.

Data analysis
We used a general inductive approach (GIA) for data analysis. 8 GIA is a method of content analysis which seeks to build understandings from participants' comments on a specific research question (or questions) as opposed to testing pre-existing hypotheses. The transcripts were read with the research question in mind. ML read and re-read the transcripts and identified lower order units of meaning which were then clustered with similar units. Units were reduced to address overlap and redundancy among the categories before the final themes and subthemes were agreed. A preliminary summary of results was sent to participants.

Results
Twelve SGMY participated, and a further potential participant wanted to take part but could not do so given the timings of the focus groups. Almost all (10/ 12, 83%) identified as gamers and two-thirds (8/12, 67%) had used SPARX. The remaining participants (4/ 12all in Focus Group 3) who had not used SPARX engaged with the short demonstration of the program during their focus group. Table 2 summarises the results of the focus groups.

Discussion
Building on earlier work, we had the opportunity to enhance SPARX, in consultation with SGMY. Prior research has already focused on improving the intervention for M aori users [e.g. 9 ], and exploratory research has begun amongst young people that have a long-term physical health condition. 10 However, feedback from SGMY suggested that the intervention be updated as it was 'old fashioned'. Participants also provided specific ideas about how SPARX could be made more acceptable to SGMY. Focus group findings and follow-up email consultations   'Just make it a bit more accessible to give more masculine body shapes a dress or more feminine body shapes a suit just so you can create characters that fall in between a bit more'.
Abilities to explore and make sense of gender (or sexuality) Opportunities for exploration were recognised as important, as Rylee noted: '…a big part of being…this minority [i.e., SGMY] is that you constantly change, you learn new information and you constantly change'. For instance, being able to pick and change gender pronouns allowed for exploration: '…if the pronouns were used and they were able to be changed, then it's a cool way for trans people or questioning people or anybody like that to try out pronouns without having to endanger themselves or do anything else…' (Keegan). and I think having it as a text just is a very ineffective way to show it off…' (Nat). Cobi stated: '…it's just the mechanics are confusing, it's a bit cliché as well, because the first level is like a chest which says, hope lies within, and then a bird called Hope comes out and then someone's like, oh, is that the Hope bird, I thought we had lost it!' SPARX should be more inclusive Inclusion issues were identified regarding SPARX, with Cobi saying: SPARX can be appealing There was no clear consensus in terms of SPARX's appeal, as some participants did not find it engaging or anticipated that it would only be useful after it was sufficiently updated and modified. However, several participants did think SPARX was appealing. As Carter said: 'I think overall it's a really good way of giving people the techniques to listen with stuff that they've got going on and I acknowledge that it's not meant to be like a therapist'.
after the focus groups led to several changes being made to create the updated version of SPARX (see Table 3 and Figure 1 for details).
Earlier work has already investigated the importance of identity-affirming avatars in gaming for gender minority youth. 11 Similarly, we also found that games can provide valuable opportunities for young person to explore their gender identity, and yet gaming often constrains diverse gender expression, an issue that can be overcome with enhanced customisation options. 11 We have attempted to address some of the issues raised by SGMY but did not have sufficient funds to address these issues as fully as we would have liked.
Strengths of this study include using an established qualitative method to help improve a freely available evidence-based psychotherapy together with SGMYan underserved population. Notable limitations include this being a small-scale exploratory study which only recruited from SGMY-specific organisations. ML, who facilitated the focus groups, is part of the development team that created SPARX and participants knew this, and as a result, some participants may have felt inhibited to express criticisms of the program. Recruitment was challenging, as is often the case for studies focused on this unique population 12 and even more so during a global pandemic. We attempted to include intersex young people, given earlier SPARXrelated work identified their elevated mental health needs, 13 but unfortunately, we were unsuccessful in this endeavour. Ideally, more online focus groups would have taken place; however, given the Reduce dialogue and make it 'less like reading from a book'

3
Enhance the game controls (e.g. ability to use a mouse) 3 a Glooming Negative Automatic Thoughts (Gnats). b Changes we would prioritise for SGMY if we had been given additional funding.
potential issues related to research fatigue associated with marginalised populations 14 and the time pressures to launch the updated version of SPARX, this was not possible. Despite these factors, this study is worthwhile because it has sought to listen to the voices of a marginalised group so that SPARX better meets the needs of SGMY; this listening to underserved populations is an important initial step towards a necessary process of change. 15 Conclusion SGMY are more likely to experience mental ill-health and interventions such as SPARX can make a difference to their wellbeing. We have demonstrated that SGMY have valuable insights and perspectives which can help enhance the acceptability of tools such as SPARX for SGMY. It is important to respond to the needs of SGMY, and this can be done by better tailoring 'mainstream' interventions. Ongoing monitoring of the uptake of SPARX will be useful to establish if we have increased engagement in terms of this group and other important underserved populations (e.g. M aori rangatahi/adolescents and young people with longterm physical health conditions).