Decoding therapy to deliver the next wave of digital mental healthcare services

24 Jan 2020

The words we speak and hear create our experiences. They impact our thoughts, behavior and the way we feel.  Words are particularly important in psychotherapy, and what is said during treatment can help an individual heal.

In our society today, 1 in 4 adults experience at least one diagnosable mental health problem in any one year – but it often goes undetected and untreated, because people don’t necessarily deal with their symptoms or have easy access to care.   For those who do receive treatment in the developed world, there is a shortfall in the quality and rate of improvement in care. This is because the methods for measuring the treatments delivered, as well as understanding the impact of the words uttered during therapy, has been lacking… until now.

Further, many people globally do not have access to mental health treatment. In half the countries in the world there are fewer than 4 mental health workers for every 100,000 people and, in Africa alone, less than 1 per 100,000.  So what can be done to deliver accessible and personalized psychological therapy to everyone globally?

Addressing two BIG challenges: First make treatment accessible; second make sure treatment works

Over the past decade, the way the healthcare industry uses and analyses data has changed dramatically with the development of new techniques in areas of artificial intelligence (AI). For the first time, data scientists are now applying deep learning and natural language processing to large pools of data from the delivery of mental healthcare treatments to extract clinical insights and understand more about clinical outcomes in a way that’s never been done on this level before.

In 2017, The Economist stated that data was the world’s most valuable resource, but it is how that data is mined and analyzed that will unlock its real value. By studying the patterns of symptoms gathered from hundreds of thousands of hours of real world patient data, and understanding more about the words spoken in therapy itself, for the first time the healthcare industry can see the different ways depression affects people, and show how tailoring treatment in a way that recognizes these differences may lead to significantly better outcomes.  Not only could this type of data analysis fundamentally change our understanding of how treatments work, but naturel language processing enables us to process every word expressed by a therapist to find out what words REALLY matter to who, and how they help.

The new era of data-driven healthcare… using real world data to enable ‘new’ real world digital treatments

Ieso scientists are studying patient characteristics, interventions and dialogue to build a clearer picture on clinical outcomes based on real world evidence. You can watch a short video about how we use data here.

Ieso’s Eight Billion Minds research program is training deep neural networks to process and analyze the language used and patterns of symptoms in hundreds of thousands of people to discover what impacts people and where there are common experiences. This real-world data comes from Ieso’s internet-enabled CBT (ie-CBT) service, where we’ve captured the utterances from more than 40,000 online psychotherapy sessions.

By translating each therapist and patient ‘utterance’ into sequences Ieso is really learning which words need to be said when for each person, and examine how each word expressed helps or hinders recovery from illness. This is helping the healthcare industry form a much better picture of the effectiveness of treatments across multiple dimensions of patient and disease characteristics and, is not only bridge the gap between clinical claims and real world evidence, but driving the enhancement of effective treatments in mental health therapy – moving towards treatments that are for more accessible, scalable, and personal than ever before.

As the use of real world data grows, so does the ethical responsibility of those accessing and using it. We also believe that it is our responsibility to make purposeful use of de-identified data in the pursuit of new and better treatments in the future. We share our approach with all our patients, and you can read more about it here.

Decoding therapy in this data driven way is just the start of this revolution. For areas of the world where there are no therapists we may be able to ultimately use these datasets to train computer systems to perform some elements of personalized therapy without human intervention. Focused on digital therapeutics, Eight Billion Minds has been launched to remove barriers to mental healthcare for millions of people. With the technologies to understand what works for whom in the real world of mental healthcare, the program is building services that will connect people in need with quality treatment wherever they are.

The power behind words, data and deep learning! 

The power behind words is invaluable and how we express and capture them could change global mental healthcare services for the better. By applying deep learning to patient data, we now know what is uncovered will change the dynamics of digital mental healthcare forever where we can move away from a ‘one-size-fits-all’ approach, and deliver treatments anytime, anywhere that are fit for each patient.

We, at Ieso, believe that we will be able to use these words and utterance sequence data to train computer systems to deliver some of the basic ‘active ingredients’ of therapy without human intervention. Our vision is a world where no one is held back by mental illness and that whenever someone takes that brave step to ask for help, that help is available to them regardless of location or how much money they have. Where treatments work first time, every time.

Words matter but so does data and deep learning.  By combining the power of technology, data and learning more from the words we say in therapy, we can apply this to computers that learn to deliver the best clinical outcomes based on evidence from the real world. This will mean our vision will quickly becoming a reality.  By understanding what makes a difference, we can fuel a new era healthcare, where more people get better faster. This will change the world!