Mental health is at a global tipping point. Over 970 million people worldwide live with a mental health disorder. The associated economic costs are expected to exceed $16 trillion by 2030. Meanwhile, healthcare systems are facing critical staff shortages.
The OECD projects that, by 2030, there will be a decline in the ratio of working-age adults to retirees from three to two, putting further strain on already limited care structures. This development is unfolding alongside the exponential pace of technological innovation, ongoing political crises and the mounting impact of climate change. Digital innovations are on the rise, yet the question is not whether we need more technology, but how we combine digital scale with human depth.
Why analogue or digital alone fall short
Many people still consider in-person psychotherapy to be the gold standard. It offers trust, nuance and human presence. It is resource-intensive, however, and, depending on the healthcare system, it is often unaffordable and inaccessible to many.
At the other end of the spectrum, fully digital mental health apps can be scaled up, but many fail to sustain engagement or adapt to the complexity of an individual’s therapy journey. Global organizations, such as the WHO and the OECD, highlight the growing treatment gap. Blended care addresses this by integrating structured digital care pathways with human-led therapy, bridging the gap between accessibility and personalization.
The promise of blended care
There is growing evidence in favour of hybrid models. A recent University of Cambridge study found that video-call-based cognitive behavioural therapy (CBT) is as effective as in-person CBT in reducing symptoms across diverse groups.
A feasibility trial involving university students in Europe demonstrated improved depression scores after a six-week blended intervention. Students received brief weekly therapy sessions via video conference, combined with app-based CBT exercises, resulting in medium to large reductions in depression symptoms. This represents the next stage of care evolution: not merely incorporating technology into care, but redesigning care around continuous digital-human interaction.
Let’s look at this case example: A patient enrolled in a blended care programme begins treatment for persistent anxiety. She works through a digital health application on her smartphone. In between, she schedules therapy sessions with a therapist. Alongside these sessions, a digital coach based on artificial intelligence (AI) is available 24/7. She can use it to talk about her thoughts and feelings or to receive additional guidance on concepts introduced in the treatment programme or the exercises discussed with her therapist.
The coach processes voice and text input, creates helpful summaries for the therapist and ensures that her next talk therapy session builds on the progress she’s made on her own since the previous session. This establishes a continuous dialogue between the digital health application, the AI-powered digital coach and the personal therapy sessions.
Making blended care work at scale
To scale blended care responsibly, systems require secure real-time data flows, user-friendly interfaces and practitioners who are AI literate. Governments must act now. In 2020, Germany’s DiGA law was the first to establish a reimbursement pathway for prescription digital therapeutics. South Korea followed suit and others, such as Belgium, France and Austria, are working on setting up viable pathways. The FDA’s AI-COA pilot is a promising start. Regulation, however, generally lags behind innovation. Equitable design, diverse training datasets and market access, based on RCT data and real-world evidence, must become the norm.
Beyond the clinic: Societal impact
Mental healthcare doesn’t start in the clinic. With smart prompts, digital check-ins and nudges and self-guided reflection, the journey to recovery starts earlier and feels less anxiety-inducing. When data confirms progress, it reinforces motivation. Blended care shifts the narrative around mental health from seeking help only in a crisis to maintaining mental health, just as you aspire to maintain physical fitness.
AI and the human-in-the-loop: Augmenting, not replacing
AI has the potential to dramatically improve the early detection and personalization of mental healthcare. And, in the long term, it will allow technology to provide just-in-time support.
Studies show that combining heart rate variability (HRV), speech and even electroencephalogram (EEG) data can signal mental distress with increasing accuracy. For example, combining ECG/HRV and EEG signals has achieved up to 95% accuracy in distinguishing between high- and low-stress states.
While AI can identify patterns and suggest interventions, it may lack the relational context and ethical judgement offered by human professionals. This is why ‘human-in-the-loop’ architecture is essential. AI gathers signals and suggests actions, while therapists contextualise, adapt and take responsibility for their patients, ensuring an optimal and personalized process.
Mental health vision 2030
By 2030, blended care should provide personalized support in rural areas, reduce the average cost-per-case and lower dropout rates by offering engaging, personalized programmes. Early intervention becomes possible when physiological signals prompt help before a crisis erupts. Patients gain agency by understanding their own patterns. Therapists gain time to focus on complex cases. The result? More resilient systems, more responsive care and more empowered lives.
This article is republished from the World Economic Forum under a Creative Commons license. Read the original article.