Back

Creation of a tool for the Identification of Neurodevelopmental Disabilities to Improve Global Outcomes (INDIGO) in children aged 0-3 years in Malawi, Pakistan and Uganda, feasibility of implementation and diagnostic accuracy.

Gladstone, M.; McCray, G.; Babikako, H.; Rasheed, M. A.; Stockdale, E.; Mwesige, A. K.; Chand, P.; Mbale, E.; Maliwichi, L.; Lynch, P.; Kieran, B.; Abubakar, A.; Kitt, H.; Donald, K. A.; van den Heuvel, M.; Nabwera, H.; Lancaster, G.

2025-11-15 pediatrics
10.1101/2025.11.13.25340081 medRxiv
Show abstract

BackgroundOver 50 million children under five have disabilities, the majority residing in resource-limited settings. Many children with neurodevelopmental disabilities (NDDs) remain undetected due to lack of culturally appropriate tools for identification. This study aimed to develop and evaluate the feasibility of the Identification of Neurodevelopmental Disabilities to Improve Global Outcomes (INDIGO) tool to identify NDDs in children aged 0 -3 years. MethodsA systematic approach was used to construct the INDIGO tool. First, 114 items of the Malawi Developmental Assessment Tool (MDAT) were mapped onto the International Classification of Functioning, Disability and Health for Children and Youth (ICF-CY). Gaps were identified leading to review of 2,158 items from existing tools. After consensus, 59 additional items (28 existing, 31 newly created) were incorporated, ensuring comprehensive coverage of ICF-CY domains. The preliminary INDIGO tool (203 items) was pilot tested in Uganda, Malawi and Pakistan with a gold standard assessment protocol developed to validate diagnostic accuracy. Feasibility outcomes including recruitment, diagnostic accuracy, cultural applicability and practical implementation were assessed. Items with over 80% diagnostic accuracy were retained for the final INDIGO prototype. ResultsA total of 425 children (151 Malawi, 145 Pakistan, 129 Uganda) were assessed using both INDIGO and gold standard evaluation. The feasibility study highlighted challenges in recruiting younger children with behavioural or socioemotional conditions and difficulties assessing sensory impairment due to resource constraints. Despite challenges, 99 high performing items were selected for the final INDIGO prototype emphasizing diagnostic accuracy, feasibility and cultural neutrality. ConclusionsThe INDIGO tool is a novel, rigorously developed instrument designed for early detection of NDDs in children 0-3 years in LMICs. By integrating a broad range of developmental domains, INDIGO addressed key limitations of existing tools. Future large-scale validation and implementation studies are needed to assess its effectiveness in routine child health surveillance programmes. Key MessagesO_ST_ABSWhat is already known on this topicC_ST_ABSO_LIMost existing developmental screening tools used globally are not culturally adapted for LMICs, require specialized training, and fail to comprehensively assess structural, functional, and environmental factors important for detecting NDDs in early childhood. C_LI What this study addsO_LIThis study presents the development and feasibility testing of the INDIGO tool, a novel screening instrument for identifying moderate to severe NDDs in children under 3 years in resource-limited settings, incorporating items to assess structural anomalies, sensory impairments, and participation, offering a more holistic and contextually relevant approach to NDD detection. C_LIO_LIThe study provides strong initial evidence of diagnostic accuracy across multiple LMIC settings, with high feasibility and acceptability among caregivers and health workers. C_LI How this study might affect research, practice or policyO_LIFuture studies could enable scaled up implementation of tools such as INDIGO integrated into national child health systems for early NDD detection, enhancing referral and intervention in LMICs. C_LI

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.