PT - JOURNAL ARTICLE AU - T Morris AU - S Mason AU - C O'Keeffe TI - ANALYSING VARIATION IN STAFFING LEVELS, ATTENDANCES AND PERFORMANCE AGAINST THE 4-HOUR TARGET USING DATA FROM THE ROYAL COLLEGE OF EMERGENCY MEDICINE'S SENTINEL SITE SURVEY AID - 10.1136/emermed-2015-205372.18 DP - 2015 Dec 01 TA - Emergency Medicine Journal PG - 984--984 VI - 32 IP - 12 4099 - http://emj.bmj.com/content/32/12/984.2.short 4100 - http://emj.bmj.com/content/32/12/984.2.full SO - Emerg Med J2015 Dec 01; 32 AB - Objectives & Background The achievement of safe and sustainable staffing levels has been highlighted as a major priority by the Royal College of Emergency Medicine's (RCEM) STEP campaign in the interests of reducing exit block and its associated problems. This study used data from the RCEM Sentinel Site Survey (SSS) to analyse variation and associations between staffing, numbers of attendees and performance against the 4-hour target in 12 emergency departments (EDs) in England.Methods A convenience sample of 12 EDs recorded the number of hours worked by all grades of staff on 20.03.14. The number attendances in each department was used to adjust for department size, and to calculate how long each staff grade could be expected to spend with the average patient in each department. SitRep data from NHS England was used to measure departmental performance against the 4-hour target in the week 17–23 March 2014.Results Having adjusted for attendances, variation was seen in all staffing grades between departments. The least variation between departments was observed in consultant hours per patient (mean=8.4 minutes, range=8.4 minutes). The most variation was observed in nurse hours per patient (mean=75 minutes, range=63 minutes). For doctors, core trainees (CT1-3) showed the most variation in hours per patient (mean=14.4 minutes, range=31.2 minutes).A significant association was observed between numbers of attendances and the number of hours worked by doctors (Spearman's Rank=0.643, P=0.024). This association was not seen between nursing hours and attendances (Spearman's Rank=0.071, P=0.077). No significant association was observed between staffing levels per patient and performance against the 4-hour target (P=0.829, P=0.186 for doctors and nurses respectively).Conclusion After adjusting for attendances variation in staffing hours can be seen in all EDs. Hours worked by doctors were seen to be associated with attendances, however hours worked by nurses were not. The number of staff hours recorded in departments was not significantly associated with departmental performance against the 4-hour target. Further analysis in a greater number of EDs across the UK would provide greater insight into variability in numbers of staff, and how this impacts patients attending the department.