Mental health support for adolescents worldwide is a priority: depression is a leading cause of illness and disability among adolescents, suicide is the second leading cause of death among 15–29-year-olds and half of all mental disorders start before 14 years old. Barriers to accessing psychological interventions include a lack of services, difficulties accessing existing services and stigma.
One step in addressing barriers to psychological interventions is through the development of digital self-help, accessed through smartphones and computers. Digital interventions have the advantage that they can be accessed discreetly, and the user can choose where and when to use them. Smartphone and social media use is rising in low- and middle-income countries with adolescents regularly using these. Hence, digital psychological interventions offer a unique opportunity to reach distressed adolescents.
The aim of STARS project, which began in March 2018, is to create a digital psychological intervention for adolescents aged between 15-18 years old experiencing high levels of psychological distress that impairs daily functioning, for example ability to study, or carry out their duties. However, it is expected that the intervention will be useful to adolescents experiencing lower levels of distress too. A key aim of the project is to develop an intervention that is highly usable and engaging for adolescents. To achieve this, human-centered design approaches have been applied, which put the user at the centre of the design process. Adolescents and community members, many of whom reported high levels of psychological distress and had experienced substantial adversity from five settings (Jamaica, Nepal, Pakistan, South Africa and West Bank and Gaza Strip) have been heavily involved in the design process.
The project is currently in the development phase. The initial stages of development explored the needs and situational context of adolescents by reviewing existing literature and conducting interviews with the adolescents and community members described above. We found that adolescents have regular access to a smartphone (though this access is sometimes shared), they frequently use technology to communicate with others (e.g. WhatsApp, social media), play games and watch videos. For example, adolescents reported watching videos to learn any new skill (whether putting on make-up or learning about engineering). This highlights the potential use of videos to teach psychological skills to distressed adolescents.
Adolescents reported experiencing distress from diverse situations (for example, school, family relationships and adverse experiences, such as bullying). Most adolescents were interested to learn how to manage strong negative emotions. Community members emphasised the stigma associated with mental health. We completed literature reviews and consultations with experts to understand which evidence-based psychological interventions could potentially be used. We also explored what techniques are likely to encourage adolescents to use the intervention. Ideas for maintaining continued use amongst adolescents included provision of badges following completion of a module or task, providing a digital character to guide the adolescent through the intervention and allowing the adolescent to create or choose a pathway in a a narrative story. To co-develop ideas for the intervention content and design, we held workshops with adolescents.
Development of prototypes
Informed by these findings, we developed prototypes for the intervention modelled on evidence-based psychological techniques. The first prototypes were very simple. For example, we displayed different psycho-educational stories on PowerPoint slides. Adolescents chose which story they preferred and why. We then revised the stories to make them more relevant and thus more engaging for adolescents.
Inspired by the high reported use of messaging applications (e.g. WhatsApp and Facebook Messenger), we explored the use of ‘chatbots’ (a pre-programmed messaging agent which gives the ‘sense’ of speaking to a person). A chatbot design allows for the use of videos, stories and other activities within the intervention. Chatbots in general are popular with adolescents and allow for adaptation of the intervention to the user’s needs, which is known to be essential for engagement. Indeed, many of the interviewed adolescents recommended use of chatbots.
Our STARS intervention will use a ‘chatbot’ base, with the inclusion of videos to communicate information. So far, we have created and gained feedback on over 30 prototypes, exploring different ways to present information and engage adolescents.
Adolescents can experience distress in various ways. They may be anxious, angry or deeply sad. To make the intervention relevant for many adolescents, we are considering adapting the information provided depending on their reported problems or distress presentation. For example, if an adolescent tends to feel anxious, they may be provided with an emotional regulation technique (such as deep breathing) with content explaining how this can help reduce anxiety. An adolescent with a different problem may get different information. If and how this will be done, will be explored through further prototyping.
The chatbot-based intervention will be developed using iterative cycles where prototypes will be created, tested with adolescents and refined to learn how best to engage adolescents, and to establish the content of the intervention.
“It’s like I’m having a one on one conversation with something…I can express myself… it helps me a lot”.
– Adolescent in Jamaica while using a “chatbot” prototype
Current questions we are trying to answer include how to best identify different presentations of stress and tailor content accordingly, what level of human guidance is required, and how to use the chatbot to help adolescents learn skills. Through this process, we hope to develop an intervention that is so engaging that adolescents will not only benefit from it, but that they will enjoy using it and come back to it in times of need.
We are aiming for a version of the STARS intervention to be ready for testing in state-of-the art randomised controlled trials by March 2020. If these research trials find that the intervention has a positive impact on adolescent emotional wellbeing, STARS will be made available as a free of charge global public good for helping adolescents around the world.
WHO acknowledges the generous support of Fondation Botnar for this project.