Frailty-Adjusted Inpatient Glycaemic Targets for Preventing Hypoglycaemia: A Quality Improvement Project.

backgroundOlder adults with frailty are particularly vulnerable to harm from tight glycaemic control, with hypoglycaemia contributing to falls, cognitive decline, and increased mortality. National (JBDS: Inpatient Care of the Frail Older Adult with Diabetes) and local (Buckinghamshire, Oxfordshire & Berkshire-BOB) guidelines recommend individualising glucose targets according to frailty level, but adherence is inconsistent. This quality improvement project evaluated glycaemic management in inpatients with frailty, aiming to reduce hypoglycaemia and improve target setting in line with frailty status. METHODOLOGY: A two-cycle quality improvement project was conducted across four elderly-care wards at the Royal Berkshire Hospital. The inclusion criteria included patients aged ≥65 years with a diagnosis of either Type 1 or Type 2 diabetes who experienced one or more hypoglycaemic episodes (<4 mmol/L) during their admission. Ward-level hypoglycaemia incidence was calculated as the proportion of all elderly-care inpatients aged ≥65 with diabetes who experienced at least one hypoglycaemic episode during each cycle. Further analyses were performed only within the cohort of patients who experienced hypoglycaemia. Cycle 1 (January to March 2025) included 31 patients meeting these criteria out of 209 elderly-care inpatients aged ≥65 with diabetes (15%). Cycle 2 (May to July 2025) included 12 out of 202 patients (6%) following staff education and reinforcement of local hypoglycaemia management guidelines. Frailty was assessed using the Clinical Frailty Scale (CFS), and glucose targets were set according to frailty category. Data from Diabetes Specialist Nurse referrals and incident reports were analysed in both cycles to assess hypoglycaemia rates, documentation of glucose targets, and medication adjustments.

resultsBetween cycles, the incidence of hypoglycaemia in elderly inpatients with diabetes decreased by 60% (from 15% to 6%). Among those who experienced hypoglycaemia, the proportion with frequent episodes (≥3 episodes) fell by 10% (from 77% to 67%). The incidence of significant hypoglycaemia (≤3 mmol/L) remained unchanged at 58%. Among patients with a CFS of 7-9, correct glucose target documentation improved substantially, rising from 25% to 80%. In contrast, documentation fell from 56% to 20% in those with a CFS of four to six, showing variability in practice. Proactive medication optimisation, such as insulin de-escalation and adjustment of oral hypoglycaemic agents, was carried out more consistently in Cycle 2.

conclusionsFrailty-adjusted glycaemic targets and proactive medication review can markedly reduce hypoglycaemia in older adults with frailty during their admission. Sustaining improvements will require ongoing staff education, standardised pathways, and clear documentation to ensure safer diabetes management in this vulnerable population.