Introduction Early involvement of specialists in emergency care has been shown to improve clinical outcomes. Delivery of high quality acute medical care by specialists has been highlighted as an aim by the Darzi review in 2008.

Aims (1) To develop a model of high quality, expedient, cost effective consultant-lead care of acute Gastroenterological emergencies. (2) To develop a model for training in acute Gastroenterology.

Methods In January 2007 consultant gastroenterologists at QMC campus of NUH set up a specialist consultant lead acute Gastroenterology on call service through job planning with dedicated allocation of programmed activity (PA) (and withdrawal from rota for unselected acute medicine). Consultants are on-call for 7 days/nights at a time supported by an SpR in training, during which time all elective commitments are cancelled. Rounds of medical high dependency unit, intensive care unit and of two acute medicine wards are performed at 8–9 am daily. Any referrals on speciality wards are also received, reviewed and managed as clinically appropriate. All inpatient endoscopy requests are vetted and procedures are then performed by the acute Gastroenterology team, in an endoscopy room dedicated to acute/inpatient service. The on-call team is also supported by a hospital-wide acute diarrhoea service with expedited flexible sigmoidoscopy to exclude C difficile infection. Data on clinical activity were collected prospectively including diagnosis on admission, interventions undertaken and admission arrangements.

Results Data were collected on 660 patients referred to the acute Gastroenterology team over a period of 76 days. Median number of referrals in a period of 24 h was 9.4 patients (range 5–17). This amounts to 13% of acute medical admissions to NUH. Indications for referral were GI bleeding 192 (29%), Hepatobiliary180 (27%), Diarrhoea 60 (9%), Abdominal pain 54 (8%), Dysphagia 50 (7%), Iron Deficiency Anaemia 40 (6%), IBD 20 (3%) and miscellaneous (11%). Patients were seen by a Consultant Gastroenterologist within 12 h of referral in 73% of cases and within 24 h in a further 18% of cases. Mean endoscopic workload was of five gastroscopies, one flexible sigmoidoscopy and 0.5 colonoscopy per day. Inpatient endoscopy requests were performed on the day of referral in 82% of cases and the following day in a further 12%.

Conclusion We present a new model of acute Gastroenterology service delivery which fulfils the targets of a high quality, expedient consultant lead, gastroenterology service and an excellent training opportunity. Data provide an estimate of demand for acute gastroenterology services.