Development of evidence-based UK-wide antimicrobial intravenous-to-oral switch criteria: a DELPHI consensus process

Abstract

Abstract Introduction Antimicrobial intravenous-to-oral switch (IVOS) is an antimicrobial stewardship (AMS) strategy with patient safety and cost-effectiveness benefits. Studies show that IVOS decisions had no negative impact on patient outcomes,1 and pharmacist involvement brought about more appropriate antibiotic prescribing and 34% reduction in medication costs.2 Healthcare settings have developed individualised local IVOS checklists. Aim To collate IVOS criteria from local checklists, complete a rapid literature review to evaluate IVOS criteria and undertake a 3-step Delphi consensus-gathering process to agree evidence-based, UK-wide IVOS criteria for hospitalised adults. Methods Local IVOS checklists from the UK nations were sought through stratified sampling. The rapid review followed PRISMA guidance and is registered on PROSPERO [CRD42022320343]. Criteria with highest, or over 50%, appearance in checklists and literature were combined for Step 1 of the 3-step Delphi process. Step 1: Pilot/1st round questionnaire: An expert panel appraised each criterion via a 5-point Likert scale for relevance to safe and effective IVOS and ease of assessment in the clinical setting. Step 2: Virtual meeting: Step 1 respondents discussed questionnaire results and attained criteria consensus for Step 3. Step 3: 2nd round questionnaire: Healthcare professionals UK-wide stated level of agreement via a 5-point Likert scale regarding the need of each criterion for safe and effective IVOS. Results Forty-five local IVOS checklists were selected: 42 from English Acute Trusts, two from Scottish Health Boards and the All-Wales checklist. Sixteen of 477 papers were included in the rapid review. Criteria were synthesised into a 5-section framework: 1-Timing of IV antimicrobial review; 2-Clinical signs and symptoms; 3-Infection markers; 4-Enteral route; 5-Infection exclusions. Combination of IVOS criteria resulted in 41 criteria for Step 1. Step 1 had 24 respondents, 15 of those participated in Step 2. Thirty-five criteria went into Step 3, which had 242 respondents (England n=195, Northern Ireland n=18, Scotland n=18, Wales n=11). The majority were female (n=154; n=7 preferred not to say) and based in NHS Acute Teaching Trusts (n=121). Antimicrobial or infection specialist pharmacists were the highest respondents (n=65) followed by general physicians (n=55). Sixty-seven per cent of respondents (n=161) considered IVOS to have a positive impact on outcomes of clinically stable patients. Consensus was achieved for 27 IVOS criteria (66% of initial criteria). Examples of agreed criteria were: ‘IVOS should be considered within 48 hours of first dose of IV antimicrobial being administered’ and ‘Temperature between 36-38°C for past 24 hours’. Discussion/Conclusion Evidence-based consensus was achieved for IVOS criteria from a UK-wide multidisciplinary group and will be used to develop an IVOS tool for hospitalised adults. A study strength was use of the Delphi process to capture consensus; a limitation was no checklist from Northern Ireland was identified at time of the study, however Steps 1-3 had participation from Northern Ireland. Early IVOS initiatives led by ward pharmacists have previously been shown to reduce use of hospital intravenous broad-spectrum antibiotics: co-amoxiclav (-27.9%) and cefuroxime (-46.5%).3 Further research is needed to operationalise the nationally agreed criteria in UK hospitals, expand AMS roles of pharmacy and nursing teams and assess impact.