Provide a structured summary including, as applicable: background; objectives; data sources; study eligibility criteria, participants, and interventions; study appraisal and synthesis methods; results; limitations; conclusions and implications of key findings; systematic review registration number.

BACKGROUND There is growing interest internationally in using remote consultations in primary care, particularly amidst the current COVID-19 pandemic. Despite this, the evidence surrounding safety of remote consultations is inconclusive. Appropriateness of antibiotic prescribing in remote consultations is an important aspect of patient safety that needs to be addressed. OBJECTIVE To summarise evidence on the impact of remote consultation in primary care on antibiotic prescribing. METHODS Searches were conducted in MEDLINE, Embase, HMIC, PSYCINFO and CINAHL from their inception to February 2020. Peer-reviewed publications conducted in primary healthcare settings were included. All remote consultation types were considered, and studies were required to report any quantitative measure of antibiotic prescribing. Studies were excluded if there were no comparison group (face-to-face consultations). RESULTS Thirteen studies were identified. Five studies demonstrated higher antibiotic prescribing rates in remote consultations compared to face-to-face consultations, three studies demonstrated lower antibiotic prescribing rate and two studies found no significant difference. Guideline-concordant prescribing was not significantly different between remote and face-to-face consultations for sinusitis patients, but conflicting results were found for patients with acute respiratory infections. CONCLUSIONS There is insufficient evidence to conclude confidently that remote consulting has a significant impact on antibiotic prescribing in primary care. However, studies indicating higher prescribing in comparison with face-to-face consulting are a concern. Further well-conducted studies are needed to inform safe and appropriate implementation of remote consulting, to ensure that there is no unintended impact on antimicrobial resistance.

We read with interest Shields's article 1 that evaluated the impact of teaching attendings how to ask their colleagues questions during bedside rounds. The authors stated that this had the potential to increase the engagement of multidisciplinary team members and learning opportunities. These are qualities that can be lacking in rounds, as shown in research and our personal experiences. 2 However, we would like to discuss how overly broad aims made it unclear who would benefit from the intervention and how they would benefit, thus reducing the study's clinical relevance.The study objectives seemingly focused on improving clinical teaching during bedside rounds by teaching attendings to ask more questions. However, we would argue that each group of participants gains different information from rounds. For example, residents may desire teaching on diagnostic strategies, while nurses aim to advocate for patients, and patients might want their condition explained in layman's terms. 3 The article assessed the number and type of questions asked by attendings and to whom they were directed, but it was not clear whether attendings were taught to adapt their questioning style to each participant group. The study's "question, listen, respond" conceptual framework encourages teachers to assess students' responses (listen) and tailor their next question to improve learning (respond). 4 The findings of the study potentially support our argument. Nurses and patients rated how "engaging" and "worthwhile" rounds were and no significant differences between the experimental and control groups were reported. Meanwhile, residents in the experimental group rated the rounds significantly higher. 1 In future research, the authors could consider using the audio-visual recordings of rounds to assess the appropriateness of attendings' questions to the role of each team member.We also believe that the authors could have evaluated participant's perception of how the round was "worthwhile" or "engaging", to offer insight into which part of the intervention was efficacious and identify confounding factors. Participants' satisfaction may not be due to the increased number or type of questions alone, but how the questions were asked, such as the attendings' use of body language or appropriateness of the topic.