How should we implement collaborative care for older people with depression? : A qualitative study using normalisation process theory within the CASPER plus trial

BACKGROUND: Depression in older people may have a prevalence as high as 20%, and is associated with physical co-morbidities, loss, and loneliness. It is associated with poorer health outcomes and reduced quality of life, and is under-diagnosed and under-treated. Older people may find it difficult to speak to their GPs about low mood, and GPs may avoid identifying depression due to limited consultation time and referral options for older patients. METHODS: A qualitative study nested within a randomised controlled trial for older people with moderate to severe depression: the CASPER plus Trial (Care for Screen Positive Elders). We interviewed patient participants, GPs, and case managers (CM) to explore patients' and professionals' views on collaborative care developed for older people, and how this model could be implemented at scale. Transcripts were analysed thematically using normalization process theory. RESULTS: Thirty-three interviews were conducted. Across the three data-sets, four main themes were identified based on the main principles of the Normalization Process Theory: understanding of collaborative care, interaction between patients and professionals, liaison between GPs and case managers, and the potential for implementation. CONCLUSIONS: A telephone-delivered intervention, incorporating behavioural activation, is acceptable to older people with depression, and is deliverable by case managers. The collaborative care framework makes sense to case managers and has the potential to optimize patient outcomes, but implementation requires integration in day to day general practice. Increasing GPs' understanding of collaborative care might improve liaison and collaboration with case managers, and facilitate the intervention through better support of patients. The CASPER plus model, delivering therapy to older adults with depression by telephone, offers the potential for implementation in a resource-poor health service