Purpose The purpose of this paper is to describe research into attachment styles of rough sleepers and considersthe implications for practice. Design/methodology/approach The research was structured interviews with a cohort of rough sleepers analysed through evidence-based techniques, and the implications were drawn out with reference to current best practice. Findings The rough sleepers in the cohort had a very different pattern of attachment styles to the housed population, with 100% insecure vs c35%, and 50% insecure disorganised vs >15%. Research limitations/implications The limitation is that the cohort was relatively small, n = 22 and was a sample of convenience. The implications are that homelessness services working with rough sleepers need to be attachment-informed as much as trauma-informed. Practical implications Practical implications are that homelessness services need to have a more rounded psychological perspective such as psychologically informed environments rather than just a trauma-informed approach. Social implications Rough sleepers suffer from deeply pervasive and severe attachment disorders, and this may be causal to their becoming rough sleepers and is certainly a factor in whether or not they are successfully rehoused. Originality/value There is almost no other original research published into the attachment styles of rough sleepers or homeless people. The current trend is for trauma-informed services: the call for attachment-informed ones is original.

N/a.