Background Changing population demographics and the call to meet the needs of patients with a non-malignant diagnosis (16% of all our hospice referrals 2013–14) led us to review our inpatient unit (IPU) service – a resource-intense (and costly) part of our service. We have 10 beds for a population of 550000. This specialist service provides for a small proportion of the total hospice patients who might need admission for symptom control, acute respite, planned respite, rehabilitation, and/or terminal care. We asked what measure (s) exist to demonstrate efficient use of existing resources, and how we utilise these to prepare for future demand.

Aim To collect additional information regarding bed occupancy and to compare this with 'throughput', benchmark against other units, and look at the effect of changes in demand upon these measures.

Findings for 2013–14 (2011–12 for comparison) will be reported.

The annual occupancy figure of 72% masked considerable unpredictable fluctuations from day to day. Holding a bed for an admission accounted for an additional 10% of available bed days. An 11% increase in admissions or an increased LOS of 1.1 days would have increased occupancy to 80%. Throughput had increased to 27.3 (23.4). Comparative data from other units will be reported.

Conclusion Demand has increased. No national recommendations exist for ideal 'occupancy rates' or 'throughput' and definitions are inconsistent. Data is only comparable in the context of unit size (and LOS) as neither reflects the greater impact on small units of time of death and holding beds. 'Bed turnover interval' is worth further exploration. We need to be clear whether it is a 'full' IPU that makes for efficient use or whether it is being able to provide a bed in a crisis?