Exploring Social Networks in the LatAmFINGERS Trial: The Importance of Pre‐Intervention Social Involvement

Abstract Background LatAm‐FINGERS is a non‐pharmacological multicenter randomized clinical trial aimed at preventing cognitive impairment. The intervention advocates for a lifestyle change based on diet, exercise, risk factor control, cognitive training, and socialization. However, the baseline assessment lacks a evaluation of the participants sociability before the intervention. Assessing social activity poses inherent challenges due to its elusive nature. An individual’s social networks encompass the entirety of their personal contacts and the interrelationships among them; it provides an objective way to measure sociability. Our aims are to test the feasibility of a method for assessing sociality based on social network features, to show the baseline state in the trial, and to suggest a standard for gauging intervention effectiveness in this domain. Method We designed a pilot study to assess the feasibility of evaluating social networks in 50 participants of LatAm‐FINGERS Argentina. Participants were invited to complete an online Spanish version of the International Social Network Questionnaire. It prompts the participant to identify social contacts (emotional support, companions in activities, and health‐related supporters). Participants are then inquired about the habits of these social contacts and their risk factors. With this information, applying graph theory, we reconstruct the each subject´s network (Figure 1), identify healthy and unhealthy nodes, and estimate structural measures of the network. Result The survey demonstrated good applicability, with 43/50 subjects able to independently complete it, while 7 subjects reported technology difficulties. Tables 1 and 2 summarize the structural and compositional characteristics of the subjects' social networks at baseline. A significant number of individuals (53.48%) had social networks where the proportion of family members exceeded 50%. There’s substantial age variability, with a mean deviation of 14.4 years among nodes in each network. 48% of social contacts are sedentary, and 31% have unhealthy diets. Men have a significantly higher proportion of sedentary individuals (p = 0.003) and non‐healthy diets (p = 0.009) in their networks. Conclusion Our study demonstrates the feasibility of assessing social networks, revealing significant disparities among individuals and genders. Previous studies highlight the influence of networks on individual risk factors, underscoring the need to incorporate this measure as a mediator of lifestyle changes' effectiveness.