Abstract Introduction In the last few decades, hospital pharmacies have introduced new medication supply systems to increase safety and cost-efficiency1. Hospitals currently use a mixture of inpatient supply types, including ward stock, “one-stop” dispensing (OSD) and non-OSD patient labelled dispensing. A mixed supply model has become the dominant system in the UK1 despite some criticism2 and has been introduced in other countries3. There is no current guidance for deciding the type and duration of inpatient supply, and little known about how pharmacy staff make this decision. Aim To explore how pharmacy staff decide on type and duration of medication supply for hospital inpatients in an English teaching hospital organisation. Methods A list of 19 factors was developed through discussions with pharmacy staff. These were arranged into four categories: medication, patient, treatment, and miscellaneous. An online survey was designed using Qualtrics. This explored staff perceptions of the importance of each factor when making decisions, separately for type and duration of supply, using a 5 point Likert scale ranging from 1 (”not at all”) to 5 (“a great deal”). A free-text box was available to add any additional factors. All pharmacy staff responsible for ordering patient-specific medication were invited to take part. Demographics were collected, participation was voluntary and all response data anonymous. Responses were subjected to descriptive statistical analysis using Microsoft Excel. A pre-selected subset of factors were compared using Mann-Whitney U tests. The study was approved as a service evaluation in the organisation concerned; therefore ethical approval was not necessary. Results There were 47 pharmacist and 10 medicines management pharmacy technician (MMPT) responses (response rate 32% for each group). For decisions both about supply type and duration, 14 factors were considered important (median score 4 or 5) with “patient on dosette box” and “duration of treatment” rated 5. Others with a median of 4 included “high risk patient”, “medication route/formulation”, “day of the week” and “drug history completed”. Only one factor, “nurse skill set on ward”, was considered unimportant (median score 2). The remainder were considered moderately important (median score 3). Statistical analysis suggested that some factors were significantly more important in certain groups. The factor “medication screened by a pharmacist” appeared less important to pharmacists versus MMPTs (p=0.0002) and less important to staff employed at the trust for longer than 10 years versus those employed less than this (p=0.0011). Discussion/Conclusion Results of this study provide valuable insight into the factors pharmacy staff consider when deciding on the type and duration of inpatient medication supply. To our knowledge, no previous studies have explored the factors influencing this kind of decision-making. Limitations include a small sample size, and that we focussed on which factors were more or less important rather than how they were used to guide decisions. We conducted some sub-group analysis, but the project was not powered to demonstrate statistical significance. This could now be addressed with the results of this survey enabling a power calculation. Further work using semi-structured interviews is currently underway to further explore how the factors influence decision making.