2.1 Data analysis
The focus groups were audio-recorded and verbatim transcripts underwent thematic analysis by the authors based upon a pragmatic process where components of experiences were pieced together to form a picture of collective experience of the participants, based upon a method described by Aronson15. Coding of the transcripts was initially completed independently and finalised after meetings. The subthemes derived from analysis were matched, where appropriate, to a proposed new model of physician prescribing decisions published by Murshid and Mohaidin4.
The initial coding of transcripts followed by aggregation into thematic dimensions and subthemes was informed by our philosophical approach. The study was originally conceived as a descriptive quantitative survey and the focus groups that followed were intended as a means to critique the appropriateness of prescribing decisions as audited against quantitative data derived from the first phase of the study. However, it became apparent, after transcription of the three focus groups, that there was greater potential to apply interpretivism in our approach to analysis. We realised that there were further insights to be gained by comprehending the thought processes that provide context and foundation to prescribing decisions. Philosophically, our approach therefore evolved more into alignment with verstehen , a term that was originally introduced by Max Weber 16 which essentially refers to an understanding of the world as others see it. It is recognised that the term ‘interpretivism’ embraces a variety of different philosophical approaches.17 This study essentially involves human interpretation and we believe that it may be more accurately defined as one that is phenomenologically orientated. Thus, we sought to understand what it is like to be a prescriber and pharmacist on the AMU (their ‘lived experience’) and to appreciate the conscientiousness of prescribers and prescribing advisors as they embark upon making prescribing decisions. It is important to acknowledge that the medical and pharmacological information that was presented within the case studies, to provide context for participants, was not relevant to this phenomenological analysis because there was no intention to pass judgement over the choice of medicines or to question the appropriateness of decisions. Thus, we felt that it was important to disassociate, from the analysis, clinical details relating to care of patients. For these reasons, the case studies are not included in this paper.
FINDINGS AND DISCUSSION
The findings highlighted some well-known driving forces that influence prescribing decisions. These include ‘patient characteristics’, ‘drug characteristics’, ‘drug characteristics, ‘pharmacist factors’ and ‘trustworthiness’. Interestingly, the influence of ‘marketing effects’ by the pharmaceutical industry, a variable derived from persuasion theory 4 that has been advocated as a main driver for prescribing decisions in connection with, for example, brands of drug, was not overtly observed in the present study. The reason for this finding is unknown but we postulate that, while complacency must be avoided, there may today be a greater awareness of potential conflict in interest that can arise between pharmaceutical representatives and hospital prescribers.18 The findings uncovered three new attitudinal factors that influence prescribing decisions that have not, to date, been described in the literature:1. Reliability of medication history, 2. Competing pressures and priorities and 3. Perceived responsibilities of prescribers. A summary of dimensions and subthemes identified within the findings and presented in relation to existing prescribing theory and models is shown in Table 1.