Background
The contribution of community hospitals to NHS services is changing, particularly with the increased focus on intermediate care. Discussions in Dumfries and Galloway have identified a need to review local community hospital pharmacy services, which have historically focussed on drug supply.

Aim
To conduct a focussed needs assessment for pharmaceutical services operating in community hospitals in Dumfries and Galloway.

Method
A review of the published literature was undertaken to inform methodology and identify potential pharmaceutical services to be provided to community hospitals. Identified services were used to inform pre-piloted structured interviews with a local 'expert panel' to identify potential pharmacy services that may be offered to community hospitals in the area. The 'expert panel' was a multidisciplinary purposive sample of 25 stakeholders from across the health board accepted to have a knowledge and understanding of the research topic, selected by the researcher and validated by the Director of Health Services. Interviews were followed by three rounds of Delphi using a self-completion questionnaire, based on thematic analysis of the interview transcripts, to asses consensus (set at 70%) regarding the importance of each of the proposed services.

Key findings
Responses to the Delphi questionnaires were received from >90% of the panel for each round. A consensus level of 70% was achieved in 19 out of 33 potential pharmacy services; highest importance was placed on general housekeeping duties (e.g. stock control, kardex write-up legibility, checking dispensing technique), medication review, formulary advice, prescribing policy development, therapeutic drug monitoring, discharge summaries and planning, and pharmaceutical care planning. In addition there was consensus in the sample on the importance of clinical support, therapeutic substitution, patient counselling, professional education and training, admission screening, new products and use of patients' own drugs. No consensus was reached on the remaining 14 services which included formal ward rounds, health promotion and pharmacist led clinics. Perceived barriers to implementing pharmaceutical care included lack of resources, no protocols or guidance, and rural geography.

Conclusion
Whilst this study did not allow the direct prioritisation of proposed services against one another the results do indicate a trend in support for those services which focused on prescribing management at both patient and hospital levels. Those services that achieved consensus at a slightly lower level of importance tended to be those focused on the patient e.g. counselling or on supporting others in care of patients, e.g. education. Consideration was made of using a nominal group approach in this study but the geographical spread of stakeholders and recognition of the limitations of face to face group meetings favoured the use of Delphi. Whilst the Delphi technique does have limitations it has, in this study, provided a useful method of reaching a group consensus in a geographically spread multidisciplinary sample of stakeholders. Further work is planned to prioritise those identified services further and inform a strategy for service development.