Healthcare entrepreneur Esha Oberoi shares how AI-led platforms can bring critical support to people who need help, wherever they may be
We are in the midst of a global mental health crisis. In March, the World Health Organisation (WHO) reported that the global prevalence of anxiety and depression increased by 25 per cent in the first year of the Covid-19 pandemic, with young people and women most affected.
All of us have experienced an unprecedented level of stress as we grapple with challenges including social isolation, increased domestic responsibilities and working hours that grow ever longer. Although disruptions to healthcare services have eased, WHO reports that “too many” are still unable to access the mental healthcare they need. Interestingly, it also found that people increasingly seek support online, “signalling an urgent need to make reliable and effective digital tools available and easily accessible.”
Having lived through debilitating depression, I know how important it is to receive effective support in time—which is unfortunately not the case for most people. Even before the pandemic, one in eight people globally was living with a mental health condition, but the minimum waiting time to access mental healthcare is six weeks; this extends to over six months in rural communities. Long waiting times can also lead to other problems. In Australia, where I’m based, we’ve observed more instances of antidepressants being prescribed prematurely because healthcare services are under strain.
The healthcare ecosystem as it is currently can be fragmented, expensive and inefficient. Put simply, it doesn’t meet the inherent needs of its stakeholders. Fortunately, technology like artificial intelligence (AI) can make up for the gaps by complementing human-led support and providing tools in between sessions to scale care. This is why I founded Leora, a virtual self-care companion launching this month.
Bringing mental healthcare to those who need it most
In Asia, cultural stigmas persist against those seeking help for mental issues. A platform that provides anonymous, round-the-clock, AI-powered chat services and clinically-proven therapeutic techniques, can help those in need receive support before they take the bigger step of speaking with a therapist.
AI-led chatbots are particularly useful in countries like India. With a population of more than 1 billion people, some of the world’s highest suicide rates per capita and only 4000 psychologists available, we can see how urgently scalable solutions are required.
So it’s important that tech-healthcare solutions empower people to take charge of their mental health. For example, Leora offers users a generalised anxiety disorder assessment typically used by therapists when they first meet with clients. Completing this allows users to better understand their current mental state and make better informed decisions about which treatments to pursue. Leora is also based on proven therapeutic techniques such as cognitive behavioural therapy, a widely used approach that helps people to reframe negative thought patterns. It also offers mindfulness, meditation and breathwork activities.
Crucially, AI can help with triage and early intervention, detecting and helping cases with low-to-mild anxiety and depression before they become more serious. It can also help people at risk. Leora contains a crisis detection model in which certain words will immediately trigger it to suggest connecting with a therapist or emergency health services.
The rapid evolution of AI-led healthcare
While we’re still a ways from chatbots that are indistinguishable from humans, we are already at a point where we can leverage AI and natural language programming to understand basic user intent.
Digital cognitive behavioural therapy has been around for a long time, but AI can add value by improving the delivery of e-mental health tools so that instead of being static and non-engaging, they are delivered in real-time, and are interactive and engaging for users. With responses and content written by trained psychologists, Leora can “speak” and connect with users in a natural, engaging and conversational way. This ability is constantly improving because AI can be trained to become better at assessing user intent, providing predictive analytics and creating faster, more accurate diagnoses. We also plan to develop algorithms for Leora that can match an individual’s unique needs and preferences (like language or culture) to the right therapist and specialisation.
As the demand for healthcare continues to surge, I see AI transitioning from stagnant, rule-based systems to applications that are complex forms of machine-learning involving deep-learning algorithms. The next wave of transformation will come from using de-identified data (data from which all personally identifiable information has been removed), so that practitioners and government agencies have more data that they can use to identify trends and predict health outcomes more effectively.