Sampling and recruitment
The target Delphi sample size was based on the literature, which suggests that consensus from participants with homogenous backgrounds can be achieved with samples of 10-15 participants [22]. Where heterogeneity of participants, or complexity of subject is increased, greater numbers are needed but these rarely exceed 50 participants[22]. The sample was anticipated to be relatively homogenous for pharmacy services in each health sector represented in the study, so the target was therefore 10-15 participants from each hospital sector e.g. acute trust, community hospital, mental health unit. Expert Panel participants were hospital pharmacy managers with strategic responsibilities for delivering pharmaceutical care in their setting, with permission from the Chief Pharmacist to share data within a specified timeframe.
Invitations to participate were issued through regional and national professional forums, with a small number of personal invitations issued to individuals identified from professional networks. Where the inclusion criteria described above were met, no exclusion criteria were applied.