Background and Aim: Fecal microbiota transplantation (FMT) is a highly effective therapy for recurrent or refractory Clostridioides difficile infection (rCDI). Despite inclusion in society guidelines, the uptake of FMT therapy has been variable. Physician and patient attitudes may be a barrier to evidence-based uptake of therapies; however, data assessing attitudes regarding FMT for rCDI are limited. Methods: The South Australian FMT for CDI database prospectively recorded patient outcomes of FMT for CDI from August 2013 to January 2019. A total of 93 consecutive patients who underwent FMT for rCDI in South Australia were invited to participate in a 20-question survey regarding the patient experience of FMT. All gastroenterologists and infectious disease physicians practicing in South Australia were invited to participate in an online survey comprised of 22 questions that addressed referral experience, indications for referral, perceived risks, and regulation and funding. Results: Fifty-four patients (54/93, 58%) returned the survey, of whom 52 (96%) would recommend FMT to others, and 51 (94%) were satisfied with treatment outcome. Fifty physicians returned the online survey (50/100, 50%), of whom 23 (46%) were concerned about disease transmission risk, and 15 (30%) believed that the risk of FMT would outweigh the benefit. Infectious diseases physicians and advanced trainees had significantly greater concern regarding the potential alteration of the microbiome than gastroenterology physicians and advanced trainees (8/17 (47%) vs 6/ 33 (18%); P = 0.047). Conclusion: Despite high levels of patient-reported satisfaction following FMT, physician-reported reservations exist and may present a barrier to uptake of this therapy.

To utilise effectively tools that employ machine learning (ML) in clinical practice medical students and doctors will require a degree of understanding of ML models. To evaluate current levels of understanding, a formative examination and survey was conducted across three centres in Australia, New Zealand and the United States. Of the 245 individuals who participated in the study (response rate = 45.4%), the majority had difficulty with identifying weaknesses in model performance analysis. Further studies examining educational interventions addressing such ML topics are warranted.

Background: It is known that South Australia (SA) has the highest rate of knee arthroscopy use of any state in Australia; however, Level 1 evidence demonstrates that knee arthroscopy in patients with uncomplicated knee osteoarthritis confers no benefit. In SA, which patients are presenting with knee pain and what treatments are they receiving?Aims: To determine the prevalence, persistence and treatment modalities of knee pain in SA.Methods: This study analysed data from the North-West Adelaide Health Study (1999)(2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015), a longitudinal, population-based cohort study of people aged 18 years and over (n = 4060), initially randomly selected from the north-west region of Adelaide, SA. It incorporated clinic assessments, self-completed questionnaires and telephone interviews to collect demographic, anthropometric and biochemical data over four main stages (1, 2, 3 and North-West 15 (NW15)). Data were linked to Medical Benefits Scheme data. Results:In stages 3 and NW15 of the North-West Adelaide Health Study, 30-35% of participants reported knee pain (n = 803, 452). Demographic variables associated with knee pain included older age and lower educational level, while risk factors included obesity and high waist circumference. In the 12 months preceding NW15, 33% of participants with knee pain/stiffness consulted a general practitioner for their knee pain, 10.2% an orthopaedic surgeon, and 12.6% a physiotherapist. Between 2011 and 2015, 3.0% the cohort underwent a knee arthroscopy, and 3.1% underwent knee magnetic resonance imaging.Conclusions: Knee pain affects large proportions of the SA population. Knee pain was persistent with underuse of non-pharmacological treatments and high use of specialist referral. These data support the need for a national strategy to manage osteoarthritis effectively.Funding: None.

Machine learning may assist in medical student evaluation. This study involved scoring short answer questions administered at three centres. Bidirectional encoder representations from transformers were particularly effective for professionalism question scoring (accuracy ranging from 41.6% to 92.5%). In the scoring of 3‐mark professionalism questions, as compared with clinical questions, machine learning had a lower classification accuracy (P < 0.05). The role of machine learning in medical professionalism evaluation warrants further investigation.