Quantifying Community Resilience
June 2024
Sunbelt 2024: Communities and Resilience
Teri Garstka, PhD., Michaela Bonnett, MPH., Meaghan Kennedy, MPH., Jasmine Fernandez, MQM, Randi Harms, MA
Click image for video of full research talk. Go to KUScholarWorks for PDF of slides.
Recent research suggests that community resilience is a dynamic process capable of protective properties.
We presented results from a recent study on a large population of social care networks in in the context of community resilience during COVID. A social care networks is defined as: A formal or informal community or regional partnership of cross-sector organizations in medical care, public health, social services, and other systems designed to coordinate, collaborate, share resources, and/or exchange information to refer individuals to services.
We consider such networks as complex adaptive systems and as such, a key driver of community-level change. Using network analysis and linear regression, we test their role in enhancing resilience by measuring the dynamic process underlying the interconnectedness of social care networks. We believe this is a key factor in how communities respond to emerging needs, opportunities, and crises.
For this study, we conducted network analysis using service referral interaction data between cross-sector organizations in in 22 social care networks active in 44 US counties from Jan 1, 2020 - Dec 30, 2022. From these analyses, we derived network cohesion metrics. We then used a Non-equivalent Groups Design and utilized Difference-in-Difference (DiD) method to test the hypothesis that more cohesive social care networks positively affect health and well-being related factors compared to less cohesive networks.
We conducted four Generalized Estimating Equations linear regressions using the DiD interaction to test network group (High or Low Cohesion) as a predictor of changes on community-level Health Factors, Health Behaviors, Clinical Care, and Social/Economic Factors.
Results showed a strong COVID effect and as anticipated, every health factor significantly declined over time. However, network cohesion mattered. The Difference-in-Difference (Group X Time) test was significant for every dependent variable. Highly cohesive networks mitigated the negative effects of COVID. Communities with cohesive networks were protected against steep declines experienced by those with less cohesive social care networks.
We conclude that network analysis is better suited to quantifying resilience in a community ecosystem and allows for a better understanding of what influences outcomes at the community-level. This work supports the notion that community resilience is a dynamic process that describes a network of adaptive capacities that impact human society and allow communities to better respond after adversity and take advantage of opportunities (Garstka & Kennedy, 2023; Norris et al, 2007). From here, we can identify and test interventions to more effectively enhance cohesion, influence community resilience, and optimize impact at scale by using a structured ecosystem approach. This is Tech-Enabled Community Resilience.
