STREAM
Scalable TRansdiagnostic Early Assessment of Mental Health
Completed project (2019-2024).
The Aim of the Study:
The STREAM study, led by Bhismadev Chakrabarti, with Dr. Gauri Divan and Dr. Supriya Bhavnani as site PIs in India and Prof. Melissa Gladstone and Dr. Emmie Mbale as site PIs in Malawi, and Dr. Vikram Patel (Harvard Medical School), aims to develop, and validate non-invasive, scalable tools for assessment of child development in preschool years. This early assessment supports identifying children (0 to 6 years of age) at risk of not reaching their full developmental potential, enabling their referral to timely interventions.
The Background:
STREAM has evolved from four pilot studies by Sangath and collaborating partners (2017-2019), which led to the development and validation of digital assessment tools for preschool children’s neurodevelopment. It aims to integrate these tools into a user-friendly platform for non-specialists to administer the tool in India and Malawi. The assessment is targeted towards evaluating cognitive, attention, social communication, and fine motor skills of approximately 2000 children in each country's diverse low-resource settings.
The Project Plan:
STREAM comprises of three main components:
Malawi Developmental Assessment Tool (MDAT):
An assessment tool combining observational and parent-reported developmental evaluations
Developmental Assessment on an E-Platform (DEEP):
A tablet-based, gamified cognitive assessment, honed since 2017
Screening Tools for Autism Risk using Technology (START):
A collaborative development incorporating camera-based eye tracking and gamified assessments
The tools are validated against the Griffiths-3 benchmark developmental assessment and supplemented with parent-reported questionnaires on home environment and psychosocial adversities.
Supporters and Project Duration:
This study received funding from the Medical Research Council, UK, and spanned from 2019 to 2025, with a commitment to delivering a scalable platform for assessing key neurodevelopmental domains.
For more information, visit: STREAM | Sangath; write to gauri.divan@sangath.in or supriya.bhavnani@sangath.in
Findings:
STREAM has completed assessments on over 2,000 children across two sites in Delhi, India, and Blantyre, Malawi, including 1,850 typically developing children from low-resource communities and 150 clinically diagnosed neurodivergent children. Six non-specialists have been trained to administer the STREAM app, and test-retest and longitudinal follow-up assessments have been conducted on a subset of participants. These results are being used to build a population-level dataset to validate the tool's ability to identify developmental delays and monitor change over time.
Next Steps:
Data collection for the project was completed in February 2024. The protocol paper promised two approaches to analyzing the data: a construct-driven approach and a machine-learning approach. Data analysis using the construct-driven approach has been completed (manuscript under review), and the machine learning approach is underway. Numerous secondary research questions are also being analyzed and prepared as manuscripts.
Publications:
- December 2025: Comprehensive neurodevelopmental assessment through non-specialists: Validation of the STREAM digital platform in India and Malawi
- June 2024: Scalable Transdiagnostic Early Assessment of Mental Health (STREAM): a study protocol
Team and Investigators:
Dr. Gauri Divan
Principal Investigator
Dr. Supriya Bhavnani
Co-Principle Investigator
Prof. Bhismadev Chakrabarti
Chief Investigator, University of Reading
Prof. Vikram Patel
Harvard Medical School
Prof. Melissa Gladstone
University of Liverpool
Prof. Mark Johnson
University of Cambridge
Prof. Sharat Chandran
Indian Institute of Technology, Bombay
Prof. Emily Jones
Birkbeck College, University of London
Prof. Gillian Lancaster
Keele University
Dr Emmie Mbale
University of Malawi