Objectives The aim of this study was to determine substance use among older adults based upon their social isolation and loneliness profiles. Methods Data were derived from the New Jersey Older Adult Survey on Drug Use and Health (OASDUH). Latent profile Analysis (LPA) was used to determine the profiles of older adults (N = 801) based on five dimensions of social isolation and loneliness. Logistic and generalized ordered logistic regressions were conducted to assess the relationship between the latent profiles and substance use. Results LPA identified five social isolation/loneliness profiles. The “connected and active” group had the lowest odds of cigarette use. The “alone but not lonely” group had the highest odds of cigarette use, alcohol use, and high-risk drinking. The “alone and lonely” group had the highest odds of non-medical drug use. Discussion In working with older adults who are using substances, it is important to inquire about their social isolation and loneliness. Cognitive behavioral therapy for the “alone and lonely” group may be beneficial, as it has been deemed effective in reducing loneliness and enhancing social network.

Although data about COVID-19 cases and deaths in the United States are readily available at the county-level, datasets on smaller geographic areas are limited. County-level data have been used to identify geospatial patterns of COVID-19 spread and, in conjunction with sociodemographic variables, have helped identify population health disparities concerning COVID-19 in the US. Municipality-level data are essential for advancing more targeted and nuanced understanding of geographic-based risk and resilience associated with COVID-19. We created a dataset that tracks COVID-19 cases and deaths by municipalities in the state of New Jersey (NJ), US, from April 22, 2020 to December 31, 2020. Data were drawn primarily from official county and municipality websites. The dataset is a spreadsheet containing cumulative case counts and case rates in each municipaly on three target dates, representing the peak of the first wave, the summer trough after the first wave, and the outbreak of the second wave in NJ. This dataset is valuable for four main reasons. First, the dataset is unique, because New Jersey's Health Department does not release COVID-19 data for the 77% (433/565) of municipalities with populations smaller than 20,000 individuals. Second, especially when combined with other data sources, such as publicly available sociodemographic data, this dataset can be used to advance epidemiological research on geographic differences in COVID-19, as well as to inform decision-making concerning the allocation of resources in response to the pandemic (e.g., strategies for targeted vaccine outreach campaigns). Third, county-level data mask important variations across municipalities, so municipality-level data permit a more nuanced exploration of health disparities related to local demographics, socioeconomic conditions, and access to resources and services. New Jersey is a good state to explore these patterns, because it is the most densely-populated and racially/ethnically diverse state in the US. Fourth, New Jersey was one of the few locations in the US with a high prevalence of COVID-19 during the first wave of the pandemic in the US. Thus, this dataset permits exploration of whether sociodemographic variables predicted COVID-19 differently as time progressed. To summarize, this unique municipality-level dataset in a diverse state with high COVID-19 cases is valuable for scholars and policy analysts to explore social and environmental factors related to the prevalence and transmission of COVID-19 in the US.