Health disparity is one of the most important forms of inequality. Although past research has shown that people’s social relationships are related to health and well-being, only a few studies have examined whether the patterns of social ties may increase or reduce health disparities. In other words, we know relatively little about under what conditions and for whom the forms of networks are more salubrious for health than others. By analyzing survey and digital trace data and focusing on obesity, I will examine (1) whether occupying marginalized network positions in earlier life stages, especially for those in minority groups, is associated with worse weight control and BMI trajectories over time? (2) whether having more interactions with peers who enrolled into a health intervention program is associated with healthier eating behavior? (3) whether in more socially cohesive environments people are more likely to be influenced by their peers, especially those who have a high BMI genetic risk? Given there are increasing endeavors applying network interventions to improve health, I will also use simulations to examine to what extent we can leverage social networks to counter adversities. This research will be of interest to health practitioners, education reformers, and organizations who are stakeholders in health promotion and health inequalities.