Background: In order to meet policy drivers on death in usual place of residence, it is key to understand how shared decision-making can be facilitated in practice. An integrated care pathway was implemented in primary care in the North East of England to facilitate death in usual place of residence. Aim: To understand how, for whom and in which circumstances death in usual place of residence is facilitated. Design: A mixed method realist evaluation was employed. Local primary care practice death audit data were analysed to identify outcomes using a mixed effects logistic regression model. Focus groups and interviews with staff of the integrated care pathway and bereaved relatives were analysed to identify the related contexts and mechanisms. Setting/participants: Death audit data of 4182 patients were readily available from 14 general practitioner practices. Three focus groups were conducted with primary and secondary care staff, voluntary sector organisations and care home representatives. Interviews with bereaved relatives were carried out in participants’ homes ( n = 5). Results: A mixed effects logistic regression model indicated a significant effect of year on death in usual place of residence when compared to a model without year using an analysis of deviance ( p = 0.016). Qualitative analysis suggested that this outcome was achieved when a triad of ‘experts’ (comprising patient, family members/family carers/formal carers and healthcare professionals) used open communication strategies. Conclusion: An empirically supported theory of how, for whom and in which circumstances death in usual place of residence happens is provided, which has important implications for both policy and practice.