The UtiliZing Health Information for Meaningful Impact in East Africa through Data Science (UZIMA-DS) study is a five-year project funded by the National Institute of Health (NIH) that aims to address the analytical and computational barriers that impede the ability to use technological advances in data science to change health care at the community and individual level. This NIH U-54 grant led by Amina Abubakar, PhD, Professor and Director for the Institute for Human Development, Aga Khan University and Akbar Waljee, MD, MSc, Medicine in the Division of Gastroenterology, Director of the Michigan Integrated Center for Health Analytics and Medical Prediction (MiCHAMP), includes many partners, including the Brain and Mind Institute, which leads the mental health component. We are leveraging existing surveillance data as well as novel mobile technologies (e.g., mobile apps, wearables) for the development of AI/ML-based prediction models to identify adolescents, youth, and healthcare workers at risk of depression and suicide ideation in Kenya.
Our preliminary findings show that symptoms of depression (31.5%) and anxiety (25.5%) are relatively high among Kenyan healthcare workers, with female nurses afflicted disproportionately. Participants included doctors, nurses, nutritionists, psychologists, physiotherapists, patient porters, pharmacists, and radiographers. We plan to use predictive analytics to identify those at risk of common mental health disorders. Artificial Intelligence (AI) and Machine Learning (ML) predictive analyses will be used for this purpose.
Some of the challenges encountered in the implementation of this study included 1) limited access to the internet, 2) poor technical know-how of clients, and 3) minor glitches in wearables and apps. However, because wearables are well-received and capture reliable data in real time, UZIMA-DS has demonstrated the feasibility of using smart technologies to predict the near-term risk of common mental disorders in Kenya.
These preliminary findings emerge from activities rolled out under the first of three phases of the UZIMA-DS project which is rolling out a longitudinal study among 900 healthcare workers and collecting quarterly mental health indicators such as depression and personality as well as daily data on mood, sleep, and steps, using a mobile app synced with a Fitbit. The data collected will help us identify behavioral indicators for mood validated against AI/ML prediction models.
The findings in the mental health project will ultimately address the ongoing and evolving mental health needs of Africans by building a prediction infrastructure for mental health management and prevention.
