Social vulnerability approaches seek to identify social, economic, and political drivers that exacerbate environmental risks, and inform adaptation strategies that redress uneven vulnerabilities. Social Vulnerability Indices (SVIs), one such approach, have exponentially increased in use since their inception in 2003. This paper contributes the most comprehensive and rigorous systematic assessment of SVIs to-date, as applied in hazard and disaster contexts. We evaluate how 246 peer-reviewed articles, published online between 2003–2021, conceptualized, constructed, and applied SVIs across 20 distinct hazard and disaster contexts in 91 countries. Our review extends previous assessments, not only by analyzing a larger diversity and volume of burgeoning scholarship, but by linking the content, method, and objectives of SVIs to synthesize their strengths and limitations for addressing social vulnerability. Three overarching results are reported. First, we find indicators used to assess social vulnerability across hazards, spatial scales, and geographical contexts, were relatively homogenous. Most articles (81%) drew indicators and theories from already existing SVIs. While such replication is not inherently problematic, and to some extent, reflects established risk factors associated with hazards globally, the epistemological and methodological processes through which SVIs are readily reproduced and replicated warrant serious deliberation. Second, and relatedly, articles commonly used deductive, a priori approaches to identify indicators from secondary datasets, often at the expense of inductively-derived representations of vulnerability. Most articles exclusively relied upon quantitative and/or spatial methods (94%) and used secondary or tertiary data alone (80%), without validation and groundtruthing processes (76%). Together, the replication of previous SVIs through deductive research approaches, and their wide application across diverse hazards and geographies, undermines the ability to capture vulnerability as a place-based and context-specific phenomena. Third, and compounding potential ineffectiveness, SVIs appear most often as reactive and post-hazard riskmitigation instruments, with data and findings rarely applied for policy change to address socioeconomic and political causes of vulnerability. Overall, SVIs are an increasingly used instrument; however, their replication without evolving epistemic and methodological approaches, combined with their constrained focus on reactive policy measures, hamper novel and necessary research for countering social vulnerability to increasingly severe socio-environmental and public health risks.