Background Palliative care services have widened beyond cancer in recent years, yet frail older adults rarely receive such services. There is a need to understand the dynamic multidimensional end-of-life experiences of this group in order to assess how or if a palliative approach could be beneficial. Physical end-of-life trajectories for frail older people have been described but there remains little person centred research that describes changing experiences across physical, social, psychological and existential dimensions. Aims and objectives To explore the changing multidimensional experiences of frail older people towards the end of life and to reflect on the utility of a qualitative longitudinal multi-perspective design for this population. Methods Thirteen cognitively intact, community dwelling older adults considered to be moderately or severely frail, using a clinical frailty scale, and thirteen nominated informal carers participated in up to three narrative interviews over eighteen months. Eight nominated professionals were also interviewed. The interviews were participant led, audio-recorded and fully transcribed. The ‘voice centred relational’ narrative method, incorporating analyses of multidimensional experience, was used to analyse the data. The data were then analysed longitudinally to compile case studies for each older person. Findings The narratives of these frail older people approaching the end of life illustrate patterns of multidimensional experience that differ from the end-of-life trajectories of other groups. All participants experienced physical decline however three possible patterns of psychological, social and existential experience emerged. These were stable, regressive and tragic according to the capacity to hold on to core values, and maintain a sense of self and of belonging in the world and are illustrated using visual trajectories. When the sense of self was threatened these frail older people lacked valued alternative identities and struggled with the absence of clear causative factors to explain their circumstances. The participants frequently described fears of burdening others, of moving to a nursing home or of developing dementia more than fears of dying. Losses and sustaining factors are described for physical, social, psychological and existential dimensions revealing the importance of social and community networks for supporting frail older people. There were constraints and benefits to using a qualitative longitudinal multi-perspective method, however the method enabled a deep, contextualised and rich understanding of the dynamic experiences of frail older people. Conclusion Frail older people may be supported towards the end of life by considering ways to promote the integration of the self. This may involve promoting valued alternative identities, protecting personhood through social and health care practices and by investigating ways to alleviate or make tolerable greatest fears. Community health and social care structures and social and community networks appear essential for addressing the end-of-life needs of frail older people. A qualitative, longitudinal, multi-perspective design was beneficial for investigating the experiences of frail older people.