A Quiet Revolution: Tackling Global Mental Health Epidemic with AI

COVID-19, combined with the ongoing economic crisis and the war in Ukraine that started in 2022, has brought severe consequences with which we still struggle. And no, it is not just about empty office spaces. It is a global mental health crisis that is unlikely to be resolved with traditional measures simply because of a lack of proper human resources on the scale needed. Can AI be a silver bullet able to tackle the challenge?

Let's cut to the chase right away: the answer is no. Surprised? Well, although it is common nowadays to think AI will save us (or doom us), nothing is that easy. AI adoption - despite its huge potential - is not something that happens overnight because of issues with data quality, quantity, lack of AI strategy, security concerns, etc. The mental health area is not an exception. Sure, AI can be used in battling the global mental health crisis, but hailing it as a "silver bullet" would be a huge and dangerous simplification considering that sensitive area of medical science.

Are We In the Midst Of a Global Mental Health Epidemic?

2022 can be called many things, but it would not be an exaggeration to say it was - among others - the year when mental health challenges ceased to be silenced and shamed.

Ryan Reynolds, Ed Sheeran, Dwayne "The Rock" Johnson, Meghan Markle, and Michael Phelps shared their personal battles with anxiety, eating disorders, depression, and self-worth. Selena Gomez produced a documentary, "My Mind and Me," highlighting the mental toll of fame, and Megan Thee Stallion launched "Bad Bitches Have Bad Days Too," an online platform offering resources for therapy and mental health assistance… And that is just the tip of the iceberg, as - probably - at least a few people in our close circle attend therapy.

Sure, all of these "coming outs" can be easily diminished, called a woke-ish trend, or even attention-seeking, but - unfortunately - stats also prove that we are far from being well, and mental problems are not the ones that can be "jogged out." 

According to the World Health Organization:

• 1 in every 8 people, or approximately 970 million people globally, were living with a mental disorder. Anxiety and depressive disorders were the most common.
• Around 301 million people, including 58 million children and adolescents, were living with an anxiety disorder.
• About 280 million people, including 23 million children and adolescents, were diagnosed with depression.
• Approximately 40 million people experienced bipolar disorder.
• Around 14 million people, including almost 3 million children and adolescents, experienced eating disorders.
• About 40 million people, including children and adolescents, were diagnosed with conduct-dissocial disorder.

The COVID-19 pandemic made things even worse. The global prevalence of anxiety and depression surged by a staggering 25% in the first year of the pandemic due to unprecedented stress caused by social isolation, constraints on people's ability to work, and limited access to support from loved ones and communities. 

Loneliness, fear of infection, grief after bereavement, and financial concerns have also been identified as significant stressors leading to anxiety and depression. The pandemic also coincided with severe disruptions to mental health services, leaving significant gaps in care. Many countries reported disruptions in life-saving services for mental health, including suicide prevention. 

Were the problems new? By no means, yet the pandemic underscored the chronic global shortage of mental health resources, which - again - should come as a surprise as governments worldwide spend just over 2% of their health budgets on mental health in 2020, many low-income countries reported having fewer than 1 mental health worker per 100 000 people.

The Lack Of Access to Mental Health Services and Professionals

Healthcare experts warn that countries must act urgently to ensure that mental health support is available to all, as the consequences will be severe. The sharp rise in suicides among teenagers proves it in the most terrifying way we can imagine, but there are multiple other implications, with "great resignation" and "quiet quitting" leading the pack.

Yet, the recent statistics on the lack of access to mental health services and professionals stand in stark contrast to the demand volume that is needed to address these issues. According to the report from the US National Council for Mental Wellbeing, 74% of Americans believe that mental health is as important as physical one, and 54% have reached out to mental health professionals for themselves or their loved ones. Still, 74% of the survey participants see mental health services as not accessible and affordable for all. 47% claim that the options are limited when it comes to asking for help. Meanwhile, Europe has long been struggling with health worker shortages, with mental health professionals being one of the most sought-after specializations.

And again, these can be pinned to the "spoiled" Gen Z, who take their avocado toasts for granted. However, when we step away from our old-fashioned work ethics, we must see that Gen Z taking care of their mental health is, in fact, a sign that nature is healing despite the lack of systemic solutions.

Also, money plays a role.

Gen Z refusing to join the hustle culture can be challenging for the economy, which - ironically - took people's engagement for granted. But where demand occurs, supply follows. Addressing mental health issues can be - no matter how "Succession"-like it sounds - an excellent business idea, as Gen Z is a very much desired target audience.

Big Tech already sees that coming.

How Could AI Help Tackle This Epidemic?

...that is the question, a huge question now when AI is treated as a "shining little thing" that sells and draws the attention of VCs who - even though they don't understand how it works - know that it can turn code into gold.

But is AI truly capable of resolving years of neglect in mental health care? Well, that's not so evident, and - given the sensitive nature of the problem - all AI limitations must be taken into account before jumping to any conclusions.

Let's delve into the topic.

When ChatGPT entered the room, AI immediately began to be identified with Generative AI, i.e., - among others - LLM models that form the basis of ChatGPT. And that is a simplification.

Generative AI and - as we call it - Predictive AI are two distinct branches of artificial intelligence, and the latter has been here for a long, long time. It is designed to make predictions based on existing datasets, forecast future outcomes, or support decision-making processes. Generative AI - on the other hand - is designed to produce data that wasn't in the original training set, such as images, music, text, or even videos.

Predictive AI is used in multiple applications, mainly to improve the overall productivity of companies by reducing manual labor, minimizing risk, and/or improving sales, but there is a strong representation of the Healthcare sector that successfully explores AI usage as well. AI that treats data equally can work exactly the same when monitoring health as it does when monitoring machinery.

Early Detection

AI has the capability to assess a variety of patient information, including medical records, behavioral indicators, and voice recordings from calls to intervention services, among other data points. It can identify early indicators of mental health issues before they become severe. Additionally, it can be employed in "personal sensing" or "digital phenotyping," where it gathers data from smartphones and wearable devices to detect early signs of mental health issues.

Recommendation Treatment 

Another way AI has improved mental health treatment is by identifying the “active” ingredients in personal therapy and building an individual "engine" to improve the overall effectiveness of therapy and sometimes reduce the very extended process of seeking the right clinicians or treatments.

Remote Monitoring

AI-driven platforms can facilitate teletherapy sessions, ensuring that individuals in remote areas or those unable to attend in-person sessions can still receive care. Remote monitoring tools can track a patient's progress and adherence to treatment, ensuring timely interventions if needed.

Training for Clinicians

AI can assist in training mental health professionals, offering simulations, case studies, and feedback to enhance their skills. It can also provide clinicians with decision-support tools, suggesting interventions based on the patient's data.

Generative AI chatbots

While not a replacement for human therapists, LLM-based chatbots can serve as a supplementary tool, providing exercises, coping strategies, and information between therapy sessions. Chatbots can be fine-tuned to recognize signs of severe distress or suicidal ideation and direct users to appropriate crisis helplines or emergency services. However, it is essential to note that chatbots - even the most advanced ones - should be used as a complementary tool alongside traditional therapeutic methods. This approach goes beyond human-like speaking, including empathy and professional judgment.

Summary

While AI has vast potential and capabilities surpassing human skills, it's vital to integrate it cautiously, especially in sensitive sectors like health care, to prevent unintended consequences. For sure, it can play a vital role in addressing issues related to global mental health, but now when we are often unaware of how it exactly works, it would be too risky to consider it a complete solution.