Many readers will be familiar with the challenges associated with the measurement of quality in acute medicine. The annual Society for Acute Medicine Benchmarking Audit (SAMBA) has attempted to map performance of acute medical units (AMUs) against quality standards established by SAM in 2011. The data presented in this autumn’s edition need to be interpreted with some caution – a single weekday in June is not necessarily representative of practice at other times of the year, and the figures are more than one year old, with the 2015 audit data currently being analysed. However the paper contains some interesting points which are worthy of comment. Overall, around 20% of patients waited more than 4 hours from arrival on the AMU before being seen by a ‘competent decision maker’, while more than a third of patients waited longer than the defined standard (8 hours during daytime and 14 hours after 5pm) to see a consultant. It should be noted, however that almost 70% of these patients passed through the Emergency Department (ED) prior to their arrival on the AMU, and many will therefore have already seen a senior ED clinician with a management plan being initiated. It is clearly important that Units who admit significant numbers of patients directly from General Practitioners are recording time delays for this group of patients, who should be subject to the same degree of rigorous prioritisation afforded to patients in the ED. It is also of interest that those patients least likely to fulfil all three of the defined standards were those who arrived on the AMU in the early evening period (5pm-10pm), while patients who arrive after 10pm were most likely to be seen within the appropriate time period. This is not surprising when one considers the way in which arrangements for consultant review are often designed, with morning ‘post take’ ward rounds enabling review of those patients who were admitted overnight. This needs to be addressed – 25% of patients arrived on the AMU during the evening period, and local experience suggests that many of these are patients referred directly by general practitioners. Many units now have a consultant presence on AMU until 8pm and some have extended this further to enable real-time consultant review of this cohort of patients later into the evening. Of the units which participated in SAMBA14, 97% were visited daily by a pharmacist. Pharmacists have clearly become an integral part of the AMU team over the past decade, and their value in performing medicines reconciliation is highlighted in Maria Richards’ article. While it is disappointing to read that the error rate in the medical drug history was so high for patients admitted to the AMU, accuracy was greater for patients sent in by general practitioners, suggesting that the availability of information at the time of clerking was a key factor. Widening access to, and use of the Summary Care Record, as recommended by the authors, would be a major step forward in reducing the risk associated with drug errors. Our ‘viewpoint’ section features two articles promoting different approaches to the interface between primary and secondary care. Stephen Gulliford has presented a year’s worth of data from his Ambulatory Emergency Care service in Wigan. The importance of a protected space in close proximity to the ED is demonstrated by the dramatic reduction in numbers of patients utilising the service when it was relocated due to local bed pressures. Overnight conversion of ambulatory care areas into bedded facilities may provide a temporary solution to operational pressures, but the knock-on impact can be significant. Ben Jamieson’s ambulatory service in Plymouth is staffed by General Practitioners, who also take calls from their colleagues working in the community. Utilising this model has enabled 50% of GP referrals to be managed without overnight admission, with a combination of clinical advice, ambulatory care and alternative pathways. Dr Jamieson highlights the complimentary skills which GPs can bring to an acute medicine team, although widespread adoption of this model might be limited by current recruitment challenges – which are apparently as great in general practice as they are in acute medicine.