Anxiety and depression plague the U.S., and their impact grows by the year. Forty million adults throughout the country suffer from an anxiety disorder, and 16.2 million have gone through at least one major depressive episode. Not only do these mental health conditions wreak havoc on people’s well-being, personal lives, and careers, they also carry a significant economic toll.
Depression costs the U.S. $210 billion each year. That number includes costs directly linked to depressive cases, as well as related mental health issues and physical ailments and losses associated with workplace absenteeism due to depression. Globally, depression-related costs are $1 trillion, according to the World Health Organization. With mental health issues on the rise, the need for life-saving, cost-effective depression and anxiety treatments becomes ever more urgent.
But current treatment protocols are imperfect, for both patients and providers. Researchers found that although people suffering from depression will most often go to their primary care doctors for guidance, those doctors may not be best-equipped to help them. The predominant approach to disease management for depression doesn’t include ongoing monitoring and education. Without a plan for how to cope with and improve their conditions, people are more likely to continue suffering from depression and related health problems and therefore will need additional care. Expensive additional care.
Troublingly, researchers found that many medical practices lack the infrastructure and protocols to help people on an ongoing basis. Depression and anxiety require more time and potentially more involved patient visits than strictly physical diseases, and many providers are already overworked and strapped for time. They may also struggle to find the financial resources, especially as some insurance companies are still reluctant to cover mental health expenses.
Now technology may have a role to play in alleviating mental health suffering and reducing the cost burden on healthcare providers. Researchers and scientists are already exploring the use of artificial intelligence in diagnosing disease. But increasingly, AI appears to have applications in mental health treatment as well. All of which has the potential to finally bend the mental health cost curve significantly.
Deep learning for better treatment
Depression may emerge in a patient for a number of reasons, including genetic predisposition, biological factors, and hormone imbalances. There are also a number of risk factors that can trigger depressive episodes, such as anxiety and other mental health conditions, trauma, and some prescription medications.
With so many variables at play, there is no one-size-fits-all treatment for depression. Patients sometimes react adversely to one medication, forcing them to try out several prescriptions until they find one that helps. Some benefit from talk therapy while others require a combination of treatments. But finding the right fit isn’t a precise science and identifying an effective approach can be a costly and time-consuming process for everyone involved.
Deep learning may be able to help doctors create personalized depression treatment plans based on patients’ conditions and histories. By aggregating and analyzing a person’s medical records, current medications, and other factors, a deep learning algorithm could recommend potential treatments that doctors can incorporate into their assessments. There are no guarantees those plans would work, but they would save doctors a good deal of trial-and-error when trying to determine which course is most likely to help a patient.
Importantly, an AI program would be able to cross-check potential treatments to ensure that prescriptions won’t negatively interfere with one another or set off additional health crises. Doctors would of course still have final say in recommending a course of action. But AI could help them do so quickly and more effectively.
Chatbots for prevention and care
Ideally, access to mental health specialists would be ubiquitous for all patients. In reality, however, many people cannot afford to see a psychologist or psychiatrist. Even those who can afford therapy sessions may be reluctant to do so for fear of being stigmatized. Chatbots that use natural language processing (NLP) and other forms of artificial intelligence can bridge the gap for people in these circumstances. Free chatbot apps help people suffering from depression and anxiety by giving them someone to “talk” to and a place to gather resources and recommendations for coping with their conditions.
As with diagnosing and treating mental health conditions, an AI chatbot can’t replace a licensed therapist long-term. But daily check-ins via chatbot may help people stay motivated to take actions that will help them manage their depression and anxiety. Given that ongoing care is one of the challenges primary care physicians face with patients who have these diseases, a chatbot can help sustain important habits and lifestyle changes that lead to improved outcomes. To the extent that such chatbots contribute to people’s mental wellness, they also reduce the financial burdens the healthcare system and the workforce bear due to mental illnesses.
Biometrics: the next frontier in preventive medicine
Doctors may ultimately receive assistance in the form of biometric data gathered from patients’ smartphones. One company is exploring how gathering information about how patients use their phones, their pulse rates, and other physical indicators can be used to paint a more accurate picture for their doctors. When a doctor notices that a patient’s pulse rate is surprisingly high, she knows to discuss that issue and identify ways of addressing it. In mental health and other areas, AI may become a powerful tool for monitoring and managing health conditions.
Having quality data on people’s health also takes the conversation beyond a patient’s self-reported state of wellness. Someone might be embarrassed to admit that he’s felt anxious or depressed and so might avoid having that conversation with his doctor. But by looking at his biometric data, the doctor can detect patterns that might be cause for concern.
In theory, aggregating this data could help on a broader level as well. Collecting biometric data could enable deep learning algorithms to find trends in behavioral and emotional patterns and help doctors identify patients who have high suicide risks.
Mental health issues are complex and difficult to manage. Healthcare providers who are already overworked and underprepared to address the needs of depressed and anxious patients would be well-served by AI-powered technologies that make diagnosis and ongoing treatment easier and more cost-effective.