Background: Much social network analysis in health related research has focused on the impact of social networks in the genesis of long term conditions and associated inequalities. However, there has been little research on the role of social networks in the management of such conditions. This is significant at a time when increasing policy emphasis is placed on individual self-care and assumptions are often made about social support, and familial support in particular. The management of chronic illness is complex involving the interplay of personal and contextual factors and comprised of a number of daily activities which include the ongoing negotiation of management into existing roles and the degree of individual engagement with self-management. Little is known about how these aspects interact to shape and influence management, what constitutes a social network for individuals with long term conditions and whether this changes over time.Methods: This study was embedded within the Whole Systems Informing Self-management Engagement (WISE) randomised control trial. It used a longitudinal qualitative design with initial face to face in-depth interviews, telephone follow-ups over a year and final face-to-face interviews where sociograms were also used to elicit network structure.Results: The findings suggest that who is in the social network, and the types of relationships which are present, influence how management practices are framed and the extent to which they are engaged with. Resources available to an individual through the network support, or undermine, engagement and changed over time. Networks included family, friends, GP, nurses and companion animals. The amalgamation of the different types of relationships that constitute the social networks are characterised by three typologies; the family focused network, the friend focused network and the health care professional focused network. These network types reflected where support was sought in times of crisis. In the absence of 'typical' sources accounts highlighted the substitutability of network members which was often narrowly sought from key individuals. Changes in illness management, either positive or negative, were framed around critical moments as they challenged existing norms of practice and involved significant network members. Further, co-morbid conditions placed an additional complexity to management and priorities were shaped by the recursive relationship with health services which seemingly influences patients' orientation in becoming either active assessors of health care or less engaged through being overwhelmed. Conclusions: This thesis challenges the notion of 'self'-management as an individual construct as many of the practices of illness management involved the support and/or negotiation of roles with others. Critical moments are a key point where normalised management practices are challenged and have particular relevance for future interventions which could foster these critical points to facilitate such changes in routines. Interventions and education need to reflect this wider setting in which chronic illness management occurs for the practices of management to become normalised into everyday routines.