The (Urgent) Inclusion Problem With Digital Mental Health

Illustration by Kaixin Wang

Digital mental health apps are increasingly getting popular but are not necessarily inclusive

Digital mental health apps are getting increasingly popular and are being used by millions of people across the world and in India. These are early days, and while the jury is still out on the efficacy of these apps, they do hold much promise when it comes to their scope to minimize the treatment gap in mental healthcare. 

An increased reach and wider use mean that digital mental health apps are being used by more diverse people from different classes and social groups than offline services. This is both an opportunity but also a substantial risk if apps are not designed inclusively from the outset. 

The risk emerges because these digital mental health apps which focus on improving the well-being of users rely on three pathways primarily – self-help modules, AI chatbots and providing a directory of therapists – in their product and service design. However, all of these three pathways very often exclude (or even actively discriminate against) people of a range of identities in terms of gender, caste, race, disability, religion, ethnicity, sexuality, and other salient markers. These errors of omission and commission are a huge risk for users using these apps and might exacerbate the problems they purportedly seek to solve. 

Some reasons why this problem is pronounced are as follows: 

1. Apps can reproduce, reinforce and even worsen social inequalities. 

If intersectional inclusion is not made a priority in app designing, digital mental health services in general, and apps, in specific, simply act as temporary fixes to the large and ever-present looming threat of the country’s mental health burden that is rooted in and maintained by social hierarchies that afford power to some and rob them from others. Kvasny (2006) outlines this ability of transformations that occur in the digital space, to reproduce social inequities. In other words, digital exclusion not just mirrors, but could worsen social exclusion from healthcare. 

2. Digital mental health apps are used by more people and are therefore bound to be used by more diverse users. 

Unlike offline mental healthcare services, especially non-clinical services, which are largely accessed by people from a certain class group who have the time, and are not held back by mental health stigma, which are factors that often correlate with having majority identities, digital mental health apps are available to all those who have access to technology and data. 

3. Digital fixes to social problems like mental health are susceptible to overgeneralization as often, designing for wider reach is traded for meaningful reach. 

In order for digital services to live up to their marketed promise of reaching more people than offline services, they offer one-size-fits-all content. Digital mental health apps too, have resorted to generalizing, offering vaguely personalized content, instead of diversifying their content. 

4. Online help-seeking has implications for offline help-seeking behaviour. 

Help-seeking behaviour is linked to improved mental health. Online help-seeking can inspire individuals to also seek support offline (Kauer, Mangan & Sanci, 2014). But for users of mental health apps to be motivated to transfer their online help-seeking behaviour to the offline space, the online care they received must be satisfactory. This level of satisfaction is correlated to the sociocultural relevance of the content and the ability of apps to anticipate and address their identity-related mental health needs. As offline mental healthcare is more culturally situated and nuanced in its design and delivery, it is an important pit stop for individuals to truly address their mental health needs and achieve well-being but if apps fail to provide them with meaningful satisfactory content, users may be discouraged from seeking mental healthcare. Hence, if apps aren’t inclusive, they can set a limit to the amount and quality of care an individual will receive. 

Findings from the first Inclusive Digital Mental Health Bootcamp

Recognizing the growing prevalence of mental health apps evidenced by the large number of downloads these apps receive on app stores and the ever-increasing funding to digital mental health companies, and the aforementioned problem of lack of inclusivity and the extreme need for it, Belongg hosted the  ‘Inclusive Digital Mental Health Apps Bootcamp’ on 28.09.22. 

The Bootcamp was attended by app developers, user interface designers and therapists who work with persons with disabilities, women with psychosis and others. The group tackled the aforementioned problems and discussed the scope of apps to be truly inclusive. Therapists also unpacked what inclusive mental healthcare means and looks like. The Bootcamp brought to the fore, the challenges that app developers face in making their services accessible and inclusive, some actionable solutions to take the first steps towards making an app inclusive, and more. 

The major highlights from the pre-research and the Bootcamp included the following:  

• Many product managers and startups would like to do the right thing but bandwidth and resource shortages in small companies and startups are key barriers to adapting apps to be more accessible to users with disabilities. Systematic support to these companies can make a substantial difference. 
• Most mental health AI bots are not sensitive to users’ socio-cultural-biological identities. Adopting a conscious approach to creating training data sets that are intersectional is essential otherwise these AI bots run the risk of being plainly unhelpful at best and putting users in danger at their worst. 
• Enabling constant learning about intersectional inclusion is critical for therapists and app developers to make their practices and products more inclusive. 
• The value of co-designing with marginalized community members and constantly engaging with user feedback to make apps more inclusive is essential but happens rarely. This barrier must be addressed urgently through systematic processes that invite such lead users into product design settings for these apps. 
• Using innovative ways to capture user identities while still protecting their privacy in case of a data breach is key in designing mental health experiences that adequately cater to the unique needs of individuals with marginalized identities while ensuring that no user faces risks because of disclosures made on the app. Such data protection protocols need to be enforced diligently by regulators and civil society actors. 
• The therapeutic value of a diverse and representative roster of therapists in digital apps is high. The psychological gains of users being matched with the therapist on grounds of social identities are significant and apps should make efforts to curate more intersectional therapists as part of their offering. 
• The risk of virtue signalling needs to be managed. Therapists questioned what it means to be an inclusive therapist and raised important questions about the validity of certifications to claim one is sensitive to the issues of a certain group such as the queer community. 

Belongg is hosting a series of discussions and interactive workshops to share findings from the short user research conducted with users with diverse sexuality, gender, religion, disability and caste identities, and hopes that this will over time, contribute to systemic change in the manner in which digital mental health apps get designed to be inclusive of people with a range of marginalised identities. A community of practice focused on intersectional mental health (link) helps connect practitioners and innovators with each other and with knowledge on this topic. 

Notes: 

Kvasny, L. (2006). Cultural (re) production of digital inequality in a US community technology initiative. Information, Communication & Society, 9(02), 160-181.

Kauer, S. D., Mangan, C., & Sanci, L. (2014). Do online mental health services improve help-seeking for young people? A systematic review. Journal of medical Internet research, 16(3), e66. https://doi.org/10.2196/jmir.3103