ObjectivesTo evaluate the use of a risk stratification tool and explore the contributing factors to variation in practice by clinical pharmacists.MethodsThe quantitative phase was a prospective evaluation of adherence to the risk stratification tool. Patients were selected by convenience sampling from medical wards across two hospital sites. Researchers applied the risk stratification tool to each patient, documented the code, accessed health records in subsequent days, and recorded the code assigned by the pharmacist. These codes were compared. The kappa (κ) coefficient test was performed using SPSS software as a statistical measure of agreement. The qualitative phase was designed using focus groups with clinical pharmacists. One focus group was conducted at each of the two study sites. Participants were grouped to ensure a mix of experience levels. To augment the discussion, participants completed a short survey. Focus groups were recorded and a thematic analysis undertaken.ResultsThe final cohort for quantitative analysis was 73. Researchers and pharmacists allocated the same code to 19 (26%) patients. The highest match rate was observed between researchers and rotational pharmacists. The κ coefficient was 0.039 (slight agreement) with p value=0.52 (not significant). Ten pharmacists participated in the focus groups: three from site 1 and seven from site 2. All participants reported using the principles of the risk stratification tool every day, but they rarely accessed the tool. Pharmacists reported using the tool as a workload management and communication system.ConclusionsVariation in application of the risk stratification tool exists among pharmacists. Focus group participants described multiple scenarios where non-patient factors were considered in assigning a priority code for the patient. A schedule of regular review of the criteria; training and peer review; tool validation; and research identifying the relationship between structured professional judgement and risk stratification tools is recommended.