Statistical Invisibility of Disability in Humanitarian Contexts: A Scoping Review and Case Lessons from Syria

Abstract Background An estimated 1.3 billion people live with disabilities globally, yet they remain systematically underrepresented in public health research and humanitarian policy. This "statistical invisibility" is driven by traditional quantitative methods that treat disability as a monolithic, static condition and fail to capture complex, intersectional realities. In conflict-affected settings like Syria, health information fragmentation severely amplifies these data gaps, impairing targeted relief and service delivery. Methods Guided by the Arksey and O’Malley framework, a scoping review was conducted across five electronic databases (PubMed, Scopus, Web of Science, EMBASE, and CINAHL) for studies published between 2010 and 2024 evaluating traditional statistical practices in public health for people with disabilities (PWDs). Findings were analyzed thematically and contextualized through a real-world humanitarian case example of disability data reporting in Syria. Results From 41 included studies, four core methodological limitations emerged: reliance on rigid cross-sectional modeling (e.g., standard logistic regression), pervasive absence of data disaggregation by impairment type or severity, systematic sampling exclusion of non-physical and invisible conditions (psychosocial and neurodevelopmental), and methodological inflexibility in capturing dynamic environmental barriers. In the Syrian context, a single methodological shift in data collection instruments between consecutive Humanitarian Needs Overviews caused reported disability prevalence to drop precipitously from 29% to 17%, statistically excluding over 1.5 million individuals from formal response plans. Conclusions Conventional statistical tools fail to reflect the dynamic reality of disability in crisis settings. A critical paradigm shift toward disaggregated, intersectional, and flexible data collection mechanisms is required to prevent data-driven exclusion in humanitarian aid.