Machine learning (ML) has spread rapidly from computer science to several disciplines. Given the predictive capacity of ML, it offers new opportunities for health, behavioral, and social scientists. However, it remains unclear how and to what extent ML is being used in studies of social determinants of health (SDH). Using four search engines, we conducted a scoping review of studies that used ML to study SDH (published before May 1, 2020). Two independent reviewers analyzed the relevant studies. For each study, we identified the research questions, Results, data, and algorithms. We synthesized our findings in a narrative report. Of the initial 8097 hits, we identified 82 relevant studies. The number of publications has risen during the past decade. More than half of the studies (n = 46) used US data. About 80% (n = 66) utilized surveys, and 70% (n = 57) employed ML for common prediction tasks. Although the number of studies in ML and SDH is growing rapidly, only a few studies used ML to improve causal inference, curate data, or identify social bias in predictions (i.e., algorithmic fairness). While ML equips researchers with new ways to measure health outcomes and their determinants from non-conventional sources such as text, audio, and image data, most studies still rely on traditional surveys. Although there are no guarantees that ML will lead to better social epidemiological research, the potential for innovation in SDH research is evident as a result of harnessing the predictive power of ML for causality, data curation, or algorithmic fairness.

It is well established that disasters are associated with an increase in many psychological problems, including depressive symptoms and posttraumatic stress symptoms (PTSS).¹ However, the long-term prognosis of mental disorders following disaster remains sparsely documented. Few studies have followed survivors for more than 3 years.^2,3 Even fewer studies have managed to capture the mental health status of survivors predating their disaster experiences,^2,3 so we cannot determine whether the disaster caused worsening of mental health or whether poor mental health already predated disaster exposure in affected populations.

Background: Loneliness and social isolation are elevated in older adults and associated with a range of detrimental outcomes. Despite this, there has been little research on these phenomena or on similarities and differences in their occurrence or combination in older Japanese adults. The current study aims to (i) determine what factors are associated with social isolation and loneliness among older adults in Japan; and (ii) describe the characteristics of individuals who are socially isolated but not lonely, as well as those who feel lonely but are not socially isolated.Methods: Data were analysed from 13 766 adults aged ≥65 years who participated in the 2019 wave of the Japan Gerontological Evaluation Study. Poisson regression analysis was used to examine associations.Results: Among older Japanese adults, the attributes of higher age, male gender, lower socioeconomic status, being a welfare recipient, and having depressive symptoms were associated with social isolation, while lower socioeconomic status, unemployment, welfare receipt, and poor physical and mental health were associated with loneliness. In addition, better educated, and mentally and physically healthy people were less likely to feel lonely even when socially isolated, while people who were not working and who had mental or physical health problems were more likely to feel lonely even if they were not socially isolated.Discussion: Our results indicate that in order to reduce unwanted social isolation and loneliness among older Japanese adults, in the first instance the focus should be on those individuals who are socioeconomically disadvantaged and unhealthy.