Why real-world data will break the ‘one-size-fits-all’ model of wellbeing apps
Dr Valentin Tablan, chief artificial intelligence officer at Ieso explains the role real-world data will have in the the next wave of digital health care services, and break the ‘one-size-fits-all' model to many of today’s health and wellbeing apps.
We have reached a pivotal point in mental health care where, for the first time, we can really start to harness the power of artificial intelligence, real-world data insights, smart device development, as well as human interaction with devices, to develop ground-breaking digital services. Advancements in deep learning and natural language processing, for example, now mean that we can decode large-scale data sets on a level and at a speed which was never possible before, accelerating how we analyse data and providing truly valuable insight into treatments.
By using machines we can learn more about the behaviours and responses of people in therapy, as well as the resulting clinical outcomes. We can start to understand more about who responds well to what, and which wordings and phrases work for whom, across different demographics and conditions. For the first time, we can begin to apply this knowledge and context to future mental health therapy to build improvements, not only in the treatment, but in delivery, engagement and personalisation of online mental health therapy, which has been making more headlines recently in light of the COVID-19 pandemic.
One-size-fits-all apps Vs digital health care services based on real-world data
Technology is already being used today to enable a wide range of digital health care apps and services that are marketed as the new digital cure. This type of remote access to support and care has never been more critical than it is now, when more of us may need access to treatments and services while grounded at home. It is expected that a growing number of people will struggle with their wellness and mental health over the coming months as COVID-19 spreads across the world. This further emphasises the need for easy access to treatment through our devices, while at home.
However, a majority of these apps on the market today are designed to support general health and wellbeing, as well as milder conditions such as sleep deprivation, anxiety or stress. They are not always built to treat an individual’s needs as effectively as human or medicinal intervention, and are by no means a standalone treatment. Of the digital apps that are out there, few are supported by scientific evidence such as clinical trials and even those that are, can sometimes fail to work effectively in the real-world due to a lack of representation in the trial population, or mismatches between tightly controlled trial conditions and real life.
While clinical trials establish safety and efficacy, treatments should be reviewed and assessed on real-world populations to ensure they deliver true value to people that need them most, outside the context of a traditional tightly-controlled clinical trial. This will be ever more vital to ensure the success of treatments as reliance on digital therapeutics starts to increase.
In the current context, when mental health challenges are becoming more prevalent, people will try anything to get the support they need, or at least some respite, particularly while they are stuck at home. More people than ever will be downloading apps in the hope of getting some help. However, many people will be exposed to general health and wellbeing apps that largely fall short of expectations by delivering a ‘one size fits all’ program. These apps are lacking the core elements of human-delivered care that drive engagement, and users are unlikely to open them more than a few times, or to use them for the amount of time required to achieve positive results. This poor user experience will ultimately harm the credibility of all services in this category, including the ones that do have good clinical effectiveness.
Looking beyond apps, and at the wider field of digital health care, there are credible digital mental health services available which have been tried and clinically validated over many years, and which are actively treating those with anxiety and depression. One such service is online cognitive behavioural therapy (CBT) that provides people with access to anytime, anywhere therapy, all through a standard digital device and from their own home.
Online CBT can be delivered through a secure, real-time text messaging platform, much like WhatsApp, with patients assigned a high-intensity therapist or psychological wellbeing practitioner (PWP), according to their needs. Patients use this communication method to receive confidential one-on-one treatment for mild to severe mental health disorders, over a period of several weeks, similar to face-to-face care. This method means that therapy sessions can be monitored for quality and ensures that patient progress is captured and reported into a patient data system. For patients, this method of communication is arguably the most discreet, which is particularly beneficial during periods of lockdown where other family members are in close proximity, allowing the patient to chat to the practitioner in private.
With online CBT service the anonymised therapy data from patient/therapist sessions can also be gathered and analysed by computers, using deep learning and natural language processing methods, to get real insights into the mechanisms that make therapy work for different patients. This knowledge can be applied to treatments and allows software and tools to be developed that help clinicians continually improve interventions with patients.
The future of digital healthcare delivery and scalability
Online CBT works well when there are enough available and trained practitioners to meet demand. However, that is rarely the case, and is particularly challenging in parts of the world where there is a clear lack of mental health care support and a scarcity of therapists. To answer that challenge, there is a need for digital apps and services that not only provide easy access to effective evidence-based therapy, but can also scale, particularly in uncertain times like these.
These types of services are not far from reality. By applying AI and deep learning techniques to data that we are seeing with online CBT, we can use algorithms to see patterns in clinical outcomes and patients’ behaviour, and build a better picture of the active ingredients in therapy. This real-world insight can not only be used to inform improvements to today’s clinical interventions, but also to build new digital treatments and apps that are effective in treating mental health conditions.
We should also consider that people are now interacting with an array of different smart and virtual devices, including wearables, in a more frequent, seamless and natural way. They feel more comfortable interacting with virtual assistants, moving from traditional keyboards and touchscreens to voice triggered digital dialogue. This is paving the way for interaction with more intelligent systems and digital services, which are highly personalised and nuanced, combining human-centred design with leading edge technologies such as conversational voice interfaces. This has the potential to transform the way we engage with digital services. In the case of mental health care, this change of communication paradigm will evolve the therapist-patient relationship, with the arrival of virtual assistants that deliver effective treatment through personalised, and sticky interactions.
The confluence of data availability, new data science methods, new technology and ways of interacting with it, along with human-centred design, brings about an exciting opportunity to transform digital health care for good. We will move away from the ‘one-size-fits-all’ model and start building truly effective and personalised treatments, which are developed within a realistic setting and validated by real-world evidence.