In many parts of the world, mental health is one of the most neglected areas of public health. By some estimates, close to 1 billion people are living with a mental disorder, 3 million people die annually from alcohol misuse, and one person dies every 40 seconds by suicide.
Few people have access to quality mental health services. In low- and middle-income countries, more than 75% of people with mental, neurological and substance use disorders receive no treatment. Stigma, discrimination, punitive legislation and human rights abuses are widespread.
Mental health conditions can have a substantial effect on all areas of life, such as school or work performance, relationships with family and friends and ability to participate in the community. Two of the most common mental health conditions, depression and anxiety, cost the global economy US$1 trillion each year. Despite these figures, the global median of government health expenditure that goes to mental health is less than 2%.
The evidence is overwhelming; there is work to be done. But how can scarce expertise and resources be most effectively matched with needs that are so widespread and yet so difficult to pinpoint?
In an innovative pilot project called “Mapping Mental Health in Kenya”, the Brain and Mind Institute is exploring the possibility of leveraging big data and data science to help identify areas of greatest need.
While the aspirations for this research involve ambitious goals, the project is underway within a confined time and place. The long-term goal is to produce a global atlas mapping mental health. Similar to the Johns Hopkins global COVID-19 Dashboard run by the Center for Systems Science and Engineering at Johns Hopkins University, this proposed global mental health dashboard would show how mental health is manifest across the world and with a level of granularity that is unparalleled in mental illness disease surveillance.
The world is now replete with valuable data. Access is challenging, but new avenues are emerging with the development of sophistication in big data access and analytics. Global participation in the web, social media, and the quality of satellite imagery has generated a wealth of useable data and insights – when the right technology is available to discover, organize, and analyze that data. Machine learning can harness, handle, and harmonize data technology for specific purposes. Machine learning can function as a non-bias search engine to understand previously unseen patterns in vast amounts of data, allowing for analysis of quality-of-life inputs, outputs, and outcomes. In other words, this data science-driven approach to seeing a population of interest (which can be defined broadly, as in an entire country or city, or narrowly, as in a specific cohort of first-year students on a specific college campus) does not depend on hypotheses, assumptions, predictions, sampling, or other research techniques. Rather, these machine learning and data intensive technologies involve gathering massive amounts of information and relying on complex computing to determine what factors and patterns emerge as relevant and impactful on outcomes of interest.
This pilot project is showcasing the technology in a limited context. Our intention is to demonstrate the potential for this approach at a global scale. Based on an assessment of the pilot’s results, we see scalability of the pilot to the global level as readily facilitated by the transferability of the models materializing from the Kenyan pilot to other countries, albeit with careful validation of methodology, data sources, and cultural, geographic, historical and other contextual elements. The pilot project should demonstrate the feasibility of the approach, such that funders of a larger project will have evidence on which to base investments.
In particular, the project:
Explores how data in the public domain can help stakeholders understand shifts in mental health through the utility of new AI approaches
Adds visibility to the prevalence of poor mental health in Kenya and the factors that lead to deterioration or improvement of wellbeing
Explores opportunities to improve supply chain efficiencies in the delivery of mental health support and interventions.
The goal is far larger than the production of a global map of mental health. This project envisions informing policy and intervention advancements based on access to and use of data. We are undertaking this work not only to gather insights, but also for the application of the results in public policy and by all those seeking to improve the well-being of people in Kenya, and eventually, many more people around the world.
