Commodifying mental health: apps, care and profitability

Apps claim to improve our mental health – but maybe their business models should be making us anxious all over again. Eoin Fullam looks behind the scenes.

Mental health apps already feel like a ‘thing’, we’ve all probably had a go with some of the freely available Cognitive Behavioural Therapy (CBT) and mindfulness apps out there, many of us even paying for subscriptions for those like Headspace which offer increased access to their various mental health ‘tools’. Computerised mental health now feels both novel and mundane, both exciting and safe: technologically automated self-treatment via scientifically verified ‘evidence-based’ treatment methods. Some also even offer companionship and coaching via chatbot interlocutors. My own research focuses on these ‘chatbot therapists’, but much of what follows applies to automated computerised interventions in general.

Searching ‘mental health’ on the Google Play Store brings up hundreds of options, but they can be broadly put into several categories depending on the modes of treatment that the apps employ and the mental health treatment methods that inform them. The vast majority of mental health apps are simple (and sometimes crude) meditation or mindfulness apps, or CBT apps. Meditation and mindfulness apps help the user to make time during the day to survey their mental state and to try to sort through their immediate sensations or recent experiences. Meditation can be guided or non-guided; both offer the same techniques, but guided meditation offers a video, audio, or text explainer to provide context and motivation whereas non-guided provides, usually in list form, the steps which the user must follow to perform the meditation task. It is difficult to clearly assess the popularity of apps, as they are available on different formats (Android and iOS), but Headspace and Calm, both automated meditation-based apps, have received the most media attention, downloads and user reviews, indicating that they are the frontrunners in smartphone-based treatment.

Different ways to make a profit

The broad economic context in which mental health app companies seek ways to make profits from their products involves two elements. The contemporary app market tends towards free-trial and subscription-based models, and mental health apps are no different. Attracting users to the apps is of course important, but equally important to the makers of mental health apps is attracting private and public investments and partnerships. The drive to attract and capture not just users but also investments, and the consequences that this drive has on the interventions provided is the focus of this article.

How do mental health software companies make money? The seemingly obvious approach, creating an app and selling it on an app-store, is a rarity in an era of ‘free’ software. We have become accustomed to having immediate access to apps which then go on to seek monetary recompense in other ways, and mental health software companies understand this. We cannot be expected to fork out money for an app that might be at best gimmicky, and at worst perhaps even detrimental to our already off-kilter mental health. Software companies also understand this, and so the most common approach involves a ‘trial’ method. For example, the therapy chatbot Wysa can be downloaded and used for free, but use is limited to a smaller range of all mental health activities included in the app. For a fee, users gain access to the premium version of the app, which includes more mental health activities and access to a human counsellor who is available to chat via text-based conversation.

This commercial approach depends on recruiting and retaining users who might eventually buy a subscription. This approach is risky in that the whims of users – and of the app stores from which software is purchased – must be contended with. Mental health software companies have learned that in order to build a sustainable business model, directly targeting customers, even using trail-based methods to attract discerning customers, has shortcomings.

Away from the app store

What other options are there? Mental health apps seek more sustainable funding models in three ways: 

1. Attracting investor funding from startup incubators, venture capitalists, etc.
2. Commercial partnerships – leasing the app to a company which then offers it to its employees 
3. Health care partnerships – leasing the app to a public health utility

We can understand option number one as a common ‘startup’ strategy in which a software company attracts large funding injections which can be used for rapid expansion projects and technical development. For example, the chatbot-based app Woebot generated $90 million in investor funding in 2021, bringing their total funding at that point to $114 million. Some startup apps generate vast sums over long periods with the promise that the investments will eventually pay off, this ‘uber-model’ investment style relies on a starker promise: that any competition will be demolished, and the dominant app will eventually rule the market.

A less ‘move fast and break things’-style business model involves developing public and private partnerships. In the drive for sustainable business models, mental health app companies seek subscription-based partnerships. This is potentially a more sustainable business model in that subscriptions involve long term contractual agreements, higher value software purchase deals, more entrenched customer loyalty, and more predictable buying cycles. The primary form of commercial partnerships involves providing a mental health app to another company, which then provides that app to its employees. All of the most popular mental health apps – including Headspace, Calm, Wysa, Woebot, Betterhelp, Talkspace, and Thrive – are promoted as benefiting employers who provide their app because this leads to employees taking fewer sick days. Mental health apps aggressively pursue business partnerships in the form of ‘EAPs’, Employee Assistance Programs.

Selling the product

How do these app companies convince employers or health institutions that their app should be chosen above the others? I’ve discovered that mental health software companies conduct and publish vast amounts of ‘research’, which is used essentially as promotional materials to convince potential partners of the superiority of their apps. While research is clearly needed not just into the effectiveness of this technology but also their impact on the broader fields of mental health treatment, what happens when these companies are conducting their own research into their own products? Every developer wants to prove that their own technology is not just as effective or superior to traditional treatment, but also that their own product is superior to their competitors’ rival products. Mental health software company X2AI conducted a technical report on their own app, Tess. It is worth quoting their abstract, which reads like a promotional blurb:

This technical report highlights how one mental health chatbot, or psychological artificial intelligence service named Tess, has been customized to deliver on-demand support for caregiving professionals, patients, and family caregivers at a non-profit organization. This low-cost, user friendly, and highly customizable service allows emotional support to be scaled to thousands of people at a single time.

The report concludes that:

There is evidence that using psychological artificial intelligence to provide customized support for caregiving professionals, patients, and family caregivers is a feasible service delivery method. This report suggests that the Tess service may offer an affordable and scalable solution that accommodates the busy schedules of caregivers while helping them reduce burnout and improve resilience. Furthermore, Tess’ capacity to expand support to patients further reduces the caregiver burden and has the potential to relieve feelings of depression, anxiety, and loneliness.

While each company conducting their own research can claim to strictly follow procedures such as randomised controlled trials and double-blind procedures, the context within which these procedures are undertaken cannot be ignored. The blurred line between research and promotion is readily apparent in the language used in the research papers produced by therapy bot companies, which often read more like promotional adverts than scientific studies. Software companies conduct research which ostensibly shows that their apps are successful at treating mental health on the one hand, and on the other, do this better than the competition – but also that they are more accessible, more visually appealing, easier to navigate, are more fun – basically anything that adds to their appeal which can be included in a ‘scientific’ study.

They conduct these studies in two ways, by asking their users about the apps, and by directly measuring their engagement with the app. User behaviour is tracked, evaluated and transformed into documents which are used to illustrate the effectiveness of the product and thus its marketability: its monetary value on the open market. Mental health app companies compete with each other to recruit and retain more users, develop new markets, to generate investments and partnerships. The most successful mental health app is deemed to be the ‘best’ app, but this is not necessarily because it is better at helping people with their mental health, but because they have out-competed the other apps. We see this with Headspace, which produces a remarkable amount of its own research. The app is most popular because, on one hand, it is very slickly designed, it has a lot of features, etc, and on the other hand, they can produce the best and most numbers.

Profitability overtakes effective care

For mental health apps, competition imposes a requirement, among other requirements, to create more and more promotional research to demonstrate the effectiveness in more and more ways – by identifying new ways that the app might be useful to the users. Ultimately, for computerised mental health treatment, marketability and therapeutic effectiveness become inseparable. New ways to engage and retain the users through the treatment itself will be sought, and the mental health intervention will be transformed in the process. Treatment effectiveness, while still maintaining a necessary function, will lag behind the compulsion for user-retention and expansion. There is a glaring paradox in this process.

Mental health app companies compete with each other to attract and retain more and more customers, whether individual users or commercial partners, and will modify their apps over time to facilitate this. Modification takes place on two levels, the surface ‘manifest’ level and the structural ‘latent’ level. The manifest level involves adding more features, making the app easier to use, improving the visual qualities, etc. The latent level involves modifying the codebase so that the app can be more easily integrated into a commercial partner’s assistance plans. One would assume that redesign of a mental health app would tend to mean improving the mental health intervention offered, but due to the competitive commercial impositions, intervening into mental health becomes subordinated to ensuring that, on the one hand, apps are made more attractive to commercial customers, and on the other hand, that they are made to be available to more users. These two aspects feed into each, with the former leading to more users, and the latter leading to more promotional research potential.

The demand for data means that mental health app companies must become more creative in how they generate more data, but also in how they can render user-activities as value-laden data. The makers of mental health apps will be compelled to constantly discover new uses for their apps. This involves researching the various forms of mental and more recently, physical, suffering that the app can help to alleviate. This research is then directed towards institutional recognition with the aim of reaching and recruiting an ever-broader cohort of consumers. Mental health app-makers are in the business of finding evidence that their app can not only help with dealing with mental issues such as postpartum depression but also with physical issues such as chronic pain. You can anticipate a scenario where these apps purport to deal with a much wider range of human suffering than simply ‘mental’. CBT, which currently claims to operate by pinpointing specific problems (such as ‘panic disorder’) will do the opposite: stretched to its limit, it will become an increasingly generalised and universalised ‘system’.This state of affairs, while paradoxical, can be observed in the rise of computerised mental health intervention and its potential for hegemonic spread: the competition between different apps creates the conditions for both their consumption and their production. The very fact that they are in competition with each other appears to affirm a need for their provision. But this competition is provoked, not because there is already a demand for automated therapy, that it in some way satisfies some ‘need’ – but because competition is required by the economic structure of capitalism. Marx noted: ‘Production… creates the consumer. Production not only supplies a material to the need, but it also supplies a need to the material… Production thus not only creates an object for the subject, but also a subject for the object’. The demand for computerised mental health treatment can be postulated as having presupposed its development, but only after it has reached a level of commercial success so as to be socially ubiquitous – to be encountered as an already viable way to attend to one’s mental health.