Total and excess bed occupancy by age, specialty and insulin use for nearly one million diabetes patients discharged from all English Acute Hospitals

Abstract

To investigate total diabetes bed occupancy and prolonged inpatient length of stay (LOS) in all English Acute Hospitals, we analysed hospital episode statistics (HES) discharge data for all English Acute Hospitals over 4 years for ICD10 discharge codes of E10 (‘insulin-dependent diabetes’) or E11 (‘non-insulin dependent diabetes’) by age-band (18–60, 61–75 and >75 years) and specialties. We matched these data to control discharges without these codes. There were 943,613 diabetes discharges (6,508,668 bed days) and 10,724,414 matched controls. Mean diabetes LOS increased with age for each specialty and both E10 and E11 codes, but excess diabetes LOS decreased with age. Excess diabetes LOS was <1.0 days in most groups and highest (1.2 days) in insulin-dependent surgical patients under 60 years old, where 19.7% of bed days were excess. A similar pattern was seen for 76,570 diabetes inpatients with key cardiac or surgical conditions. Excess bed occupancy due to prolonged mean LOS accounted for 325,033 bed days under general medical and surgical codes. There were 25,525 discharges with diabetic ketoacidosis (126,495 bed days) in these 4 years. Excess diabetes LOS is concentrated in younger age groups. Excess bed occupancy due to prolonged LOS in medical and surgical inpatients is three times greater than bed occupancy due to diabetic ketoacidosis. Strategies to reduce excess diabetes bed occupancy should emphasize reducing inpatient LOS in younger inpatients.

Introduction

Up to 10% of unselected inpatient populations have diabetes and most inpatients with diabetes are admitted because of medical or surgical conditions, rather than the acute or long term complications of diabetes [1], [2], [3], [4], [5], [6]. Populations with diabetes have higher hospital admission rates and inpatients with diabetes stay in hospital longer than age-matched controls without diabetes [1], [2], [3], [4], [5], [6]. The USA National discharge data described an excess mean inpatient length of stay (LOS) of 2.0 days in middle-aged diabetes inpatient populations [1], [2] and there are equivalent data from local populations in the UK and Europe [4], [5], [6], [7], [8], [9]. Although the causes for this excess length of stay are unknown, there is increasing evidence from the USA and UK that enhanced diabetes services for inpatients can reduce this excess diabetes bed occupancy [4], [10], [11], [12], [13]. The UK Diabetes National Service Framework (NSF) and the UK Long Term Conditions–Public Service Agreement (LTC–PSA) suggest ambitious targets for improving inpatient diabetes care and reducing chronic disease bed occupancy [14], [15]. There are no data on overall or excess diabetes related bed occupancy in a national UK or European population, certainly nothing to compare with the national hospital survey data in the USA [1], [2] and little on the contribution of excess LOS to overall national diabetes bed occupancy. The aim of this study was to provide an accurate descriptive analysis of total and excess diabetes bed occupancy due to prolonged length of stay in all English Acute Hospitals by age-band, specialty and insulin dependency in the main medical and surgical specialties and for key cardiac and surgical conditions.