TY - JOUR T1 - Creation of the WHO Indicators of Infant and Young Child Development (IYCD): metadata synthesis across 10 countries JF - BMJ Global Health DO - 10.1136/bmjgh-2018-000747 VL - 3 IS - 5 SP - e000747 AU - Gillian A Lancaster AU - Gareth McCray AU - Patricia Kariger AU - Tarun Dua AU - Andrew Titman AU - Jaya Chandna AU - Dana McCoy AU - Amina Abubakar AU - Jena Derakhshani Hamadani AU - Günther Fink AU - Fahmida Tofail AU - Melissa Gladstone AU - Magdalena Janus Y1 - 2018/10/01 UR - http://gh.bmj.com/content/3/5/e000747.abstract N2 - Background Renewed global commitment to the improvement of early child development outcomes, as evidenced by the focus of the United Nations Sustainable Development Goal 4, highlights an increased need for reliable and valid measures to evaluate preventive and interventional efforts designed to affect change. Our objective was to create a new tool, applicable across multicultures, to measure development from 0 to 3 years through metadata synthesis.Methods Fourteen cross-sectional data sets were contributed on 21 083 children from 10 low/middle-income countries (LMIC), assessed using seven different tools (caregiver reported or directly assessed). Item groups, measuring similar developmental skills, were identified by item mapping across tools. Logistic regression curves displayed developmental trajectories for item groups across countries and age. Following expert consensus to identify well-performing items across developmental domains, a second mapping exercise was conducted to fill any gaps across the age range. The first version of the tool was constructed. Item response analysis validated our approach by putting all data sets onto a common scale.Results 789 individual items were identified across tools in the first mapping and 129 item groups selected for analysis. 70 item groups were then selected through consensus, based on statistical performance and perceived importance, with a further 50 items identified at second mapping. A tool comprising 120 items (23 fine motor, 23 gross motor, 20 receptive language, 24 expressive language, 30 socioemotional) was created. The linked data sets on a common scale showed a curvilinear trajectory of child development, highlighting the validity of our approach through excellent coverage by age and consistency of measurement across contributed tools, a novel finding in itself.Conclusions We have created the first version of a prototype tool for measuring children in the early years, developed using novel easy to apply methodology; now it needs to be feasibility tested and piloted across several LMICs. ER -