Implementation of a text-messaging intervention for adolescents who self-harm (TeenTEXT): a feasibility study using normalisation process theory.

BACKGROUND: There are few interventions that directly address self-harming behaviour among adolescents. At the request of clinicians in Child and Adolescent Mental Health Services (CAMHS) in England and working with them, we redeveloped an adult SMS text-messaging intervention to meet the needs of adolescents under the care of CAMHS who self-harm.

METHODS: We used normalisation process theory (NPT) to assess the feasibility of delivering it through CAMHS. We planned to recruit 27 young people who self-harm and their clinicians, working as dyads and using the intervention (TeenTEXT) for 6 months.

RESULTS: Despite strong engagement in principle from CAMHS teams, in practice we were able to recruit only three clinician/client dyads. Of these, two dropped out because the clients were too unwell. We identified a number of barriers to implementation. These included: a context of CAMHS in crisis, with heavy workloads and high stress levels; organisational gatekeeping practices, which limited the extent to which clinicians could engage with the intervention; perceived burdensomeness and technophobia on the part of clinicians, and a belief by many clinicians that CAMHS may be the wrong delivery setting and that the intervention may have better fit with schools and universal youth services.

CONCLUSIONS: User-centred design principles and the use of participatory methods in intervention development are no guarantee of implementability. Barriers to implementation cannot always be foreseen, and early clinical champions may overestimate the readiness of colleagues to embrace new ideas and technologies. NPT studies have an important role to play in identifying whether or not interventions are likely to receive widespread clinical support. This study of a text-messaging intervention to support adolescents who self-harm (TeenTEXT) showed that further work is needed to identify the right delivery setting, before testing the efficacy of the intervention.