The utility of psychometric network analysis for the study of social isolation, loneliness, and social support in persons with cancer: methodology and future directions

It has been well-established that social isolation, loneliness, and inadequate social support are critical determinants of psychological and physical outcomes in persons with cancer. In contrast to latent variable analysis, upon which most current research is based, psychometric network analysis offers a framework for understanding how these social factors relate to specific symptoms, coping resources, and wellbeing in the context of cancer. Although very few studies in this area of research have used network analysis methods, this presentation focuses on a small number of recent network-analytic studies in oncology and other related examples, emphasizing methodological choices, stability procedures, interpretation of centrality indicators, and global network analyses. In addition, recommendations are presented for future research, such as longitudinal models, network analysis of interventions, and multifaceted data integration that might inform translational work in psycho-oncology. Psychometric network analysis holds strong potential to advance psychosocial oncology through the identification of specific modifiable targets that might help mitigate social isolation, loneliness, and inadequate support.