A multicentre observational study to investigate feasibility of a direct oral penicillin challenge in de-labelling 'low risk' patients with penicillin allergy by non-allergy healthcare professionals (SPACE study): Implications for healthcare systems

OBJECTIVE

The huge burden of inaccurate penicillin allergy labels (PALs) is an important driver of antimicrobial resistance. This is magnified by insufficient allergy specialists and lack of 'point-of-care' tests. We investigated the feasibility of non-allergy healthcare professionals (HCPs) delivering direct oral penicillin challenges (DPCs) for penicillin allergy de-labelling.

METHODS

This prospective observational study was conducted in three hospitals in England across three settings (acute medical, pre-surgical and haematology-oncology). Patients with a PAL were screened and stratified as low risk/high risk. Low risk patients (non-immune mediated symptoms, benign rash, tolerated amoxicillin since and family history) underwent a DPC.

RESULTS

N = 2257 PALs were screened, 1054 were eligible; 643 were approached, 373 declined, 270 consented and 259 risk stratified (low risk = 155; high risk = 104). One hundred and twenty-six low risk patients underwent DPC, 122 (96.8%) were de-labelled with no serious allergic reactions. Conversion rate from screening-to-consent was 12% [3.3% and 17.9% in acute and elective settings respectively; odds ratios for consent were 3.42 (p < 0.001) and 5.53 (p < 0.001) in haematology-oncology and pre-surgical setting respectively. Common reasons for failure to progress in the study included difficulty in reaching patients, clinical instability/medical reasons, lacking capacity to consent and psychological factors.

INTERPRETATION

DPCs can be delivered by non-allergy HCPs. A high proportion of patients with PALs did not progress in the study pathway. Strategies to deliver DPC at optimal points of the care pathway are needed to enhance uptake. Elective settings offer greater opportunities than acute settings for DPC. The safety and simplicity of DPCs lends itself to adoption by healthcare systems beyond the UK, including in resource-limited settings.

A service evaluation of patient care relating toClostridioides difficile infection (CDI) in Oxford University Hospital NHS Foundation

Introduction

The bacterium Clostridioides difficile (C. difficile) frequently colonises the human intestine. In an adult who is healthy, this organism is relatively harmless. However, disruption of the normal gut flora, which is frequently brought on by the use of antibiotics, can result in Clostridioides difficile (C. difficile) infection (CDI). As a result, C. difficile can grow and produce a toxin that causes diarrhoea. Due to the ease with which C. difficile can be spread in healthcare environments, it is essential to recognise and treat CDI patients as soon as possible and in accordance with best practises.

Aim

This service evaluation aimed to explore and understand how adult inpatients presenting with diarrhoea where (CDI is suspected are managed according to Trust Clostridioides difficile guidelines.

Methods

Purposive sampling of up to 100 adult inpatients presenting with diarrhoea where C. difficile is suspected and the patient has been an in-patient for least 48 hours post sample request. A data collection tool was developed, pilot tested and used to collect patient data from the Oxford University Hospital's Electronic Prescribing Record. Data was analysed using descriptive statistics and the Functional Resonance Analysis Method (FRAM) that involved 1) Identifying and describing key functions relating to the management of patients 2) Identifying variability in performing key functions and 3) Determine the possibility of function resonance. Ethical approval was obtained from University of Reading.

Results

Fifty patient records were accessed. The results showed that many patients had a stool sample sent for review as soon as possible, a clinical review of their condition and where there was the absence of a strong suspicion or strong evidence of C. Difficile a delay in the initiation of empirical therapy. Documentation was an area with poor compliance to guidelines. 50 % had a Bristol stool chart completed before sample sent and after CDI suspected, 60 % had isolation of the patient documented, 47 % had placement of contact precautions documented, and 20 % had documentation of discontinuation of empirical therapy. A FRAM model was developed and a total of 14 functions were identified with three functions “Record Bristol stool charts”, “clinical review of the patient”, and “placement of contact precautions” being most connected to the other functions, demonstrating the significance of these functions in the management of patients who are suspected of having CDI.

Discussion/Conclusion

The three key tasks outlined by FRAM are important to ensure that provide patients with the best care. Understanding variability in how the functions identified in FRAM are performed is important to understand how work is performed in order to then design ways of working that benefits both patients and HCPs. Further work is required to explore this area.