Improving penicillin allergy documentation in UK Defence Primary Healthcare: a quality improvement project

Up to 10% of patients report a penicillin allergy, yet >90% of these are inaccurate when assessed. Unnecessary second-line antibiotics contribute to poorer outcomes, increased antimicrobial resistance and higher costs. Accurate documentation supports stewardship and operational medical capability by improving clinical risk decisions, particularly in deployed or austere settings. We aimed to improve the proportion of adequately documented penicillin allergy records in a UK Defence primary care setting.

In a baseline population (n=33/70 labelled patients), 21% (±10%, 95% CI) of records were adequately documented using national standards. A structured patient questionnaire, developed with NHS Lothian guidance, was piloted via email, refined and scaled across a broader primary care setting using SMS text messages plus follow-up email. Records were updated in the electronic health record. Using the Model for Improvement, three PDSA cycles tested usability, refined content and scaled implementation to the full cohort (n=504 coded penicillin allergy). Daily sampling (n=10/day) supported p-chart monitoring from a 21% baseline with a three-sigma upper control limit (UCL) of 60%.

270/504 (54%) patients responded. Daily p-chart monitoring demonstrated sustained non-random improvement following intervention (day 8): median shifted from 20% (day 7) to 55% (day 12) and 65% (day 18), reaching the UCL by day 10. 38% of patients were unclear about their index allergic reaction. 33% reported a mild rash, 18% gastrointestinal upset and 18% angioedema. Symptoms associated with potential type 1 hypersensitivity or severe cutaneous reactions were infrequently reported. Using NHS protocols, an estimated 79% of documented penicillin-allergic service personnel would be suitable for drug provocation testing.

A remote, patient-centred approach with standardised data entry improved documentation quality and is scalable across UK Defence Primary Healthcare. Next steps include embedding driver-diagram-informed actions (clinician awareness and digital templates), assessing prescribing outcomes and developing Defence standard operating procedures.