Predictive Model of Community Disaster Resilience Across Serbia: A BRIC–DROP Composite Index and Spatial Patterns

Community disaster resilience is increasingly guiding risk-reduction investments, but in many Southeast European settings, comparable subnational data remain scarce. This study develops and tests a predictive model of perceived community disaster resilience across Serbia by combining BRIC–DROP dimensions into a single index and analyzing differences across hazard types and sociodemographic factors. A cross-sectional household survey was conducted using multistage random sampling and the “next birthday” method for respondent selection. The final sample included 1, 1,200 adults from 22 local government units across four regions: Belgrade, Vojvodina, Šumadija & Western Serbia, and Southern & Eastern Serbia. Participants evaluated preventive measures and societal resilience for ten hazard types and considered five social dimensions: social structure, social capital, social mechanisms, social equity/diversity, and social beliefs. Descriptive statistics, bivariate analyses (including Pearson correlations, t- tests, and ANOVA), and multiple linear regression identified key predictors of preventive behavior and perceived resilience. Composite scores highlighted spatial resilience differences. Overall perceptions mostly fell below the midpoint, with the highest ratings for pandemic/epidemic preparedness (M = 2. 32), storms/hail (M = 2. 24), and floods (M = 2. 15). The lowest ratings were for environmental pollution (M = 1. 81) and droughts (M = 1. 87). Perceived societal resilience was highest for snowstorms (M = 2. 30), storms/hail (M = 2. 28), and pandemics/epidemics (M = 2. 26), and lowest for environmental pollution (M = 1. 91) and droughts (M = 1.87. 98). Respondents reported strong family ties (M = 3. 05) and good communication and supply access (e. g., communication tools M = 3. 19; water/food access M = 3. 07), but weak institutional capacity, especially in areas like budget allocation (M = 1. 84), early warning/public notifications (M = 2. 11), rapid decision- making without bureaucracy (M = 2. 08), and evacuation/shelter capacity (M = 2. 12). Regression results were statistically significant but explained only a small portion of variance (perceived resilience R² = 0. 051; preventive measures R² = 0. 060). Age and employment in the public sector positively predicted perceived resilience; fear, income, and, to a lesser extent, education showed negative associations. These findings highlight the structural and psychosocial factors that shape perceptions of resilience. The BRIC–DROP composite indicates generally low perceived preparedness and resilience, especially in risk communication, evacuation and shelter readiness, and financing—the key bottlenecks in strengthening local resilience. The results recommend combining institutional reform with targeted risk communication to reduce fear and build trust, especially focusing on hazard areas with the lowest confidence, such as environmental pollution and drought.