The purpose of this proposal is to identify individual, community (population) and structural factors associated with lower rates of Covid-19 testing in Northern New England (NNE), with a focus on underserved and vulnerable populations. Our study includes several COVID-19 medically and/or socially vulnerable populations: communities with high levels of social vulnerability; community-dwelling older adults; individuals with medical comorbidities known to increase risk of severe COVID-19 and, particularly, rural and remote communities. Analytically, we will first qualitatively estimate individual, population and structural factors associated with higher or lower probability of having been tested for COVID-19 by combine comprehensive all- payer claims data across two states with state-level COVID-19 testing data and the CDC vulnerable community index. We will also assess the geospatial distribution of disparities in COVID-19 testing in NNE using geographic information system methods to examine factors like testing center density and distance on testing rates. We will exploit differences in structure between Vermont and Maine to identify system level factors, including provider accessibility, testing availability and provider payment rules. Our key outcomes will be COVID-19 testing, hospitalizations and excess mortality among underserved and vulnerable populations in NNE. We will augment the quantitative analysis with focus groups to identify additional barriers to testing. We will conduct multiple focus groups with individuals from vulnerable populations to identify barriers to COVID-19 testing. Once we have identified the individual, community (population) and structural factors that create barriers to COVID-19 testing and excess mortality, we will test potential interventions in two different ways, First, we will develop and deploy a Discrete Choice Experiment (DCE) both in vulnerable communities in NNE and in a nationally representative sample of rural adults to test optimal strategies to increase testing using hypothetical scenarios. Second, we will test the effect of targeted communication using a rural communication network using optimal communication strategies to facilitate increased testing and test if the communications change individual behavior and reduce health disparities. This study will be conducted in partnership with the Department of Health in both Vermont and Maine and numerous community partners.