Community-Level Vulnerability and Residents' Disaster Preparedness Levels: A Multilevel Analysis of the National Household Survey

Background and Purpose: Communities with certain characteristics are more vulnerable to disasters and experience disproportionate impact, so the preparedness efforts that aim at mitigating the impact of disasters are of vital importance for residents in those communities. Current research on disaster preparedness has primarily focused on individual-level vulnerability (e.g., individual poverty and minority status) without considering important community-level vulnerabilities that contextualize individual preparedness by influencing external assistance from public disaster management and internal social support networks. To address such gaps, this study aimed to investigate the impacts of different community vulnerability factors on individuals’ disaster preparedness levels.

Methods: Data from Federal Emergency Management Agency (FEMA)’s 2018 National Household Survey were matched with the 2018 county-level Social Vulnerability Index (SVI) developed by the Centers for Disease Control and Prevention (CDC). The working sample included 2,249 individuals nested within 352 U.S. counties. The dependent variables were cumulative preparedness and adequate preparedness. Cumulative preparedness was measured by a count of 13 preparation items such as household emergency plans and disaster supplies; higher scores indicated higher preparedness levels. Adequate preparedness was assessed by whether the respondents had prepared at least 7 of the 13 items (0 = no, 1 = yes). Community-level predictors included 15 SVI factors: socioeconomic status (income, poverty, unemployment, and high school dropout rates), household composition/disability (households with older persons, children, single parents, and people with disabilities), minority status/language (minority status, English language proficiency), and housing/transportation (vehicle access, multiunit housing, housing crowding, mobile homes, and persons in institutionalized group quarters). Control variables included individual age, gender, race, ethnicity, educational level, and homeownership. Multilevel linear regressions with random intercepts were conducted to examine the impacts of SVI factors on cumulative preparedness and multilevel logistic regressions with random intercepts were performed for adequate preparedness.

Results: Individuals living in communities with a higher percentage of population with disabilities (B = 0.11, p < .05), higher percentage of minority (B = 0.02, p < .05), and higher percentage of multiunit housing (B = 0.04, p < .001), reported better cumulative preparedness. In contrast, residents in communities with a higher percentage of household with no vehicle available tended to prepare fewer items (B = -0.04, p < .01). People who lived in communities with a higher percentage of older adults (OR = 1.05, p < .05), children (OR = 1.09, p < .05), population with disabilities (OR = 1.11, p < .05), minority (OR = 1.01, p < .05), and multiunit housing (OR = 1.04, p < .001), were more likely to be adequately prepared. Those residing in communities with a higher percentage of households with no vehicles available were less likely to be adequately prepared (OR = 0.97, p < .01).

Conclusions and Implications: These findings suggested that neighborhoods with a general lack of household vehicles would hinder residents’ preparedness; surprisingly, some other community vulnerability factors may contribute to individual preparedness. Pre-disaster social services and workers should emphasize community-level vehicle access and strengthen preparedness education in vulnerable communities.