Background
The influence of patient pressure on the decision to prescribe has been previously investigated in primary care1. Perception of such pressure can lead to a prescription that is judged to be inappropriate according to evidence based medicine and this can often evoke a strong feeling of discomfort in the prescriber regardless of whether a prescription is issued2. To date, no studies have explored sources of discomfort in the secondary care prescriber and the issue of perceived patient pressure to prescribe.

Method
Thirty-two doctors working in various specialties and of varying grades were purposively selected for interview from four hospitals (two district general and two teaching hospitals). A further ten doctors were then theoretically sampled to explore emerging themes from the data. Prior to interview, participants were requested to remember any instances where they felt uncomfortable when making a prescribing decision, regardless of whether a prescription was issued. Focusing on 'uncomfortable decisions', using a technique known as the 'critical incident technique', made it possible to avoid general perceptions about what doctors believed influenced their prescribing. During the interview, the incident(s) were discussed in-depth, and more general themes emerging from the interviews were also explored. All interviews were tape-recorded and transcribed verbatim. Data analysis was approached with a modified grounded theory technique.

Findings
Over half of interviewees discussed an uncomfortable prescribing decision caused by a perceived pressure to prescribe. Doctors working in paediatrics, intensive care and care of the elderly also reported pressure to prescribe from patients' relatives. All grades of doctor were affected by this phenomenon. Patients involved in the incidents were often described as 'well informed' and 'manipulative'. Interviewees resisted conceding to patient pressure in approximately half of all incidents. Referral was often the method of resistance, passing on the problem as opposed to reaching an immediate resolution. The extent of the discomfort seemed dependant on the grade of the doctor. This could be attributed to the ease and feasibility for referring patients, e.g. prescribing decisions made whilst on call. Reasons given for capitulating to patient's requests included an uncertainty in diagnosis, maintenance of a good doctor-patient relationship, and to get rid of the problem. The avoidance of conflict with patients, patient's relatives and the multi-disciplinary team was also cited as a reason for capitulating.

Discussion
Patient and relative pressure to prescribe is a very real phenomenon within secondary care. Similar to primary care, the doctor's relationship with the patient was an important factor in whether the doctor capitulated to the request. However, a theme not reported in primary care was agreeing to the patient's request in order to minimize disruption within the overall health care team. The frequent use of emotive language by participants suggested that opinions of a patient's personality and the formulation of stereotypes were important in such decisions to prescribe. This study has highlighted a much underdeveloped area of prescribing influences in secondary care and raised areas of further study, along with potential implications for training of secondary care prescribers.