Improving fluoroquinolone safety practices on medical wards through a quality improvement (QI) initiative

Abstract

Introduction Fluoroquinolones are the fifth most frequently prescribed antibiotic class in the United Kingdom (after penicillins, tetracyclines, macrolides and sulphonamides), with a defined daily dose of around 0.5 per 1000 inhabitants in England [1]. While they provide oral antibiotic coverage for a broad range of pathogens, there are serious complications associated, including peripheral and central neuropathy, tendon rupture and suicidal ideation. In January 2024, the Medicines and Healthcare products Regulatory Agency (MHRA) issued guidance advising that fluoroquinolone prescriptions should be restricted to situations where other antibiotics are deemed inappropriate due to side effects, antimicrobial resistance, treatment failure or contraindication [2]. Additionally, patients should be adequately counselled for adverse effects. At Barts Health NHS Trust, to prevent inappropriate use, clinicians are advised to only prescribe a fluoroquinolone if in accordance with trust microbiology guidelines or on direct specialist microbiology/infectious diseases (ID) advice. An audit from December 2024-March 2025 identified inappropriate prescribing along with inconsistent drug counselling and documentation of safety-netting advice to patients. Inappropriate use of fluoroquinolones, particularly in high-risk patients (age > 60, on corticosteroids, renal impairment or solid organ transplant), increases the risk of adverse effects compromising patient safety. We aimed to address this issue by increasing the number of appropriate fluoroquinolone prescriptions, i.e., those that are either in accordance with trust microbiology guidelines or directly advised by ID/microbiology to more than 80%. Additionally, we aimed to ensure at least 60% of patients discharged on a fluoroquinolone receive counselling in the form of documented safety-netting advice on their discharge summary. Method This QI initiative was implemented across eight medical wards including acute medicine, geriatrics and specialty medicine from December 2024 and is currently ongoing. Using Plan-Do-Study-Act (PDSA) methodology [3], the first PDSA cycle was launched on 29 April 2025. This involved the introduction of infographic posters summarizing trust guidance, alongside meetings with key stakeholders from microbiology and pharmacy and targeted teaching sessions on wards with a higher incidence of fluoroquinolone prescribing. Data were collected prospectively using electronic patient records to assess appropriateness of prescriptions and documentation of patient counselling. Appropriateness of prescriptions was defined by whether they were in accordance with trust guidelines or following direct microbiology/ ID advice. Fishers exact test was used to assess for differences in the proportion of patients prescribed a fluoroquinolone appropriately and safety-netting at discharge at baseline and postintervention. Additionally, weekly percentage rates were plotted on a run chart to track changes over time and assess for non-random variation. Results Baseline data revealed n = 32 patients were prescribed a fluoroquinolone antibiotic. Of these, 66% (n = 21) were deemed to have received appropriate prescriptions. However, only 22% (n = 7) of patients discharged on a fluoroquinolone received documented safety-netting advice; 72% (n = 23) of patients prescribed a fluoroquinolone were identified as high risk of adverse effects. Using weekly percentages, the baseline median was calculated as 70.8% for appropriate prescribing and 0% for safety-netting. Post-PDSA Cycle 1 data identified n = 19 patients; 79% (n = 15) received appropriate prescriptions, 16% (n = 3) of patients received documented safety-netting advice at discharge. Baseline and postintervention cohorts did not significantly differ in rates of appropriate prescriptions (p = 0.36) or documented safety-netting advice at discharge (p = 0.73). However, run chart analysis (Figure 1) did demonstrate a shift in weekly appropriate prescribing with six consecutive data points above the baseline median indicating non-random variation. This was not demonstrated for safety-netting (Figure 2) with no trends or shifts observed. Conclusions This QI initiative has shown early progress towards meeting the goal of improving appropriate fluoroquinolone prescribing, with further PDSA cycles ongoing to assess for sustained changes. In contrast, no significant improvement was observed in the provision of safety-netting advice, which may reflect limited clinician awareness and service constraints. Ongoing efforts within the second PDSA cycle are aiming to address this gap. To ensure the long-term sustainability of the initiative, future interventions will need to be multifaceted, engaging not only prescribers but also pharmacists, dispensary technicians and microbiology clinical leads. (Figure present).