Brian L. Levy’s research while affiliated with University of South Carolina and other places

Despite a large body of work on neighborhood effects on health, past studies are limited in their treatment of neighborhoods as largely static spaces, defined primarily by the average characteristics of their residents. In this study, we draw on the triple neighborhood disadvantage perspective to explore how socioeconomic disadvantage in a neighborhood's mobility network uniquely relates to children’s overall health levels, independent of residential disadvantage. We investigate this by combining 2019 SafeGraph data on mobility patterns from roughly 40 million U.S. mobile devices with information on children, families, and neighborhoods from the 2015-19 American Community Survey and 2019 Panel Study of Income Dynamics Child Development supplement. We find that mobility-based neighborhood disadvantage (MND) generally predicts child health independently of residential neighborhood disadvantage (RND), but associations vary by race and family income, and are contingent on the broader metropolitan context. Our study advances existing research on the effects of mobility networks by shifting from analyzing aggregate-level outcomes to exploring how mobility-based disadvantage affects individual outcomes. Overall, our results indicate that the relationship between neighborhood disadvantage and child health is nuanced and complex. Findings from our study suggest that researchers aiming to understand the influence of neighborhood contexts should examine individuals’ residential environments as well as the environments of neighborhoods connected through individuals’ everyday mobility.

Background: Amidst the COVID-19 pandemic, there has been growing concern about the declining mental health and healthy behaviors compared to pre-pandemic levels. Despite this, there is a lack of longitudinal studies that have examined the relationship between health behaviors and mental health during the pandemic. In response, the statewide COVIDsmart longitudinal study was launched. The study's main objective is to better understand the effects of the pandemic on mental health. Findings may provide a foundation for the identification of public health strategies to mitigate future negative impacts of the pandemic. Methods: Following online recruitment in spring of 2021, adults, ages 18 to 87, filled out social, mental, economic, occupational, and physical health questionnaires on the digital COVIDsmart platform at baseline and through six monthly follow-ups. Changes in the participant's four health behaviors (e.g., tobacco and alcohol consumption, physical activity, and social media use), along with sex, age, loneliness score, and reported social and economic (SE) hardships, were analyzed for within-between group associations with depression and anxiety scores using Mixed Models Repeated Measures. Results: In this study, of the 669 individuals who reported, the within-between group analysis indicated that younger adults (F = 23.81, p < 0.0001), loneliness (F = 234.60, p < 0.0001), SE hardships (F = 31.25, p < 0.0001), increased tobacco use (F = 3.05, p = 0.036), decreased physical activity (F = 6.88, p = 0.0002), and both positive and negative changes in social media use (F = 7.22, p = 0.0001) were significantly associated with worse depression scores. Additionally, females (F = 6.01, p = 0.015), younger adults (F = 32.30, p < 0.0001), loneliness (F = 154.59, p < 0.0001), SE hardships (F = 22.13, p < 0.0001), increased tobacco use (F = 4.87, p = 0.004), and both positive and negative changes in social media use (F = 3.51, p = 0.016) were significantly associated with worse anxiety scores. However, no significant changes were observed in the within-between group measurements of depression and anxiety scores over time (p > 0.05). Physical activity was not associated with anxiety nor was alcohol consumption with both depression and anxiety (p > 0.05). Conclusions: This study demonstrates the longitudinal changes in behaviors within the context of the COVID-19 pandemic. These findings may facilitate the design of preventative population-based health approaches during the COVID-19 pandemic or future pandemics.

Although there is an abundance of research on the association of school poverty (or socioeconomic status) and test score level, there is very little rigorous longitudinal evidence on the cumulative effects of exposure to differing school contexts. Drawing from methods used first in epidemiology and then in neighborhood effects research, we use population-level longitudinal data from North Carolina to estimate a structural nested mean model that permits proper adjustment for time-varying confounding. Unlike panel data studies using student fixed effects, which often report close to null findings, we find evidence of modest but significant negative effects of school poverty composition on eighth-grade reading and math test scores in models that control for third-grade test scores and baseline treatment status.

Personal religion and spirituality can be sources of strength during a crisis like the COVID-19 pandemic. Drawing from a larger study of LGBTQ+ wellbeing during the COVID-19 pandemic, this article reports on the faith and spirituality of LGBTQ+ adults in the United States. Nearly 700 diverse LGBTQ+ individuals were recruited via Qualtrics Panel. Analysis showed that those who prayed more and who were religious or spiritual prior to the pandemic were more likely to rely on their religion or spirituality to understand and cope with the pandemic. This study has implications for social workers, helping professionals, and faith leaders, including avoiding assumptions, utilizing a person-centered approach, and being familiar with community resources.

Background: The COVID-19 pandemic has affected people's lives beyond severe and long-term physical health symptoms. Social distancing and quarantine have led to adverse mental health outcomes. In addition, COVID-induced economic setbacks have likely exacerbated the psychological distress affecting broader aspects of physical and mental well-being. Remote digital health studies can provide information about the pandemic's socioeconomic, mental, and physical impact. COVIDsmart was a collaborative effort to deploy a complex digital health research study to understand the impact of the pandemic on diverse populations. Here we describe how digital tools were used to capture the effects of the pandemic on the overall wellbeing of diverse communities across large geographical areas within the state of Virginia. Objective: The objective of this paper is to describe the digital recruitment strategies and data collection tools applied in the COVIDsmart study and share preliminary study results. Methods: COVIDsmart conducted digital recruitment, eConsent, and survey collection through a HIPAA compliant digital health platform. This is alternative to traditional in-person recruitment and onboarding method used for studies. Participants in Virginia were actively recruited over three months utilizing widespread digital marketing strategies. Six months of data were collected remotely on participant demographics, COVID-19 clinical parameters, health perceptions, mental and physical health, resilience, vaccination status, education/work functioning, social/family functioning, and economic impact. Data were collected using validated questionnaires or surveys reviewed by an expert panel that were completed in a cyclical fashion. To retain high level of engagement throughout the study, participants were incentivized to stay enrolled and complete more surveys to further their chances of receiving monthly gift card drawing and/or one of multiple grand prizes. Results: Virtual recruitment demonstrated relatively high rates of interest in Virginia (N=3,737), and 782 have consented to participate in the study (21.1%). The most successful recruitment technique was the effective use of newsletters/emails (41.4%). The primary reason for contributing as a study participant was advancing research (79.9%), followed by the need to give back to their community (64.8%). Incentives only reported as a reason among 21% of the consented participants. Overall, the primary reason for contributing as a study participant was attributed to altruism at 88.6%. Conclusions: The COVID-19 pandemic has accelerated the need for digital transformation in research. COVIDsmart is a statewide prospective cohort to study the impact of COVID-19 on Virginians' social, physical, and mental health. The study design, project management, and collaborative efforts across multi-disciplinary groups led to the development of effective digital recruitment, enrollment, and data collection strategies to evaluate the pandemic effects on a large, diverse population. These findings may inform effective recruitment techniques across diverse communities and participants' interest in remote digital health studies.

BACKGROUND The COVID-19 pandemic swiftly and significantly impacted lives in the United States. A “Wisdom in Crowds” model posits that a team of diverse, independently thinking individuals works better in predicting outcomes. Strong relationships are fundamentally important for working collaboratively with diverse stakeholders and co-producing research projects on digital health platforms during the COVID-19 pandemic. We present here an effort of three diverse organizations in designing and deploying a complex digital health research study called COVIDsmart to understand the impact of the pandemic on study participants, their wellbeing, and their communities across large geographical areas with a low-financial and low-personnel cost design. OBJECTIVE The objective of this paper is to describe the digital solutions adopted in the implementation of the COVIDsmart study and share the preliminary study results. METHODS In a multi-institutional effort, COVIDsmart used a combination of digital recruitment, eConsent, survey design through a precision medicine digital health platform. This HIPAA compliant digital platform is alternative to the traditional in-person recruitment and onboarding method for studies. Participants in Virginia were recruited using over three months of active recruitment utilizing widespread digital marketing strategies. Six months of data were collected on the participants of the COVIDsmart study on demographics, COVID-19 clinical parameters, health perceptions, mental and physical health, resilience, vaccination status, education/work functioning, social/family functioning, and economic impact. Data were collected using validated questionnaires or surveys reviewed by an expert panel that were completed in a cyclical fashion. To retain high level of engagement throughout the study, participants were incentivized to stay enrolled and complete more surveys to further their chances of receiving monthly gift card drawing and/or one of multiple grand prizes. RESULTS Virtual recruitment demonstrated relatively high rates of interest in Virginia (N=3,737), and 782 have consented to participate in the study (21.1%). The most successful recruitment technique was the effective use of newsletters/emails (41.4%). The primary reason for contributing as a study participant was advancing research (79.9%), followed by the need to give back to their community (64.8%). Incentives only reported as a reason among 21% of the consented participants. Overall, the primary reason for contributing as a study participant was attributed to altruism at 88.6%. CONCLUSIONS From its start-up, to its careful planning and structured process, and to its successful completion, the development of this large-scale digital COVID-19 study is attributed to the “Wisdom in Crowds” approach. This method has helped forge a partnership among multiple diverse institutions and disciplines where the collaborative decision-making process is the essential element. With the help of the “Wisdom in Crowds” approach, more large-scale digital study designs can be co-developed to research health impacts on diverse populations over large geographic areas.

Race and class disparities in COVID-19 cases are well documented, but pathways of possible transmission by neighborhood inequality are not. This study uses administrative data on COVID-19 cases for roughly 2000 census tracts in Wisconsin, Seattle/King County, and San Francisco to analyze how neighborhood socioeconomic (dis)advantage predicts cumulative caseloads through February 2021. Unlike past research, we measure a neighborhood's disadvantage level using both its residents' demographics and the demographics of neighborhoods its residents visit and are visited by, leveraging daily mobility data from 45 million mobile devices. In all three jurisdictions, we find sizable disparities in COVID-19 caseloads. Disadvantage in a neighborhood's mobility network has greater impact than its residents' socioeconomic characteristics. We also find disparities by neighborhood racial/ethnic composition, which can be explained, in part, by residential and mobility-based disadvantage. Neighborhood conditions measured before a pandemic offer substantial predictive power for subsequent incidence, with mobility-based disadvantage playing an important role.