The burden of β-lactam allergy labels in health care: a systematic review and meta-analysis

Abstract

Background

Unverified β-lactam allergy labels (BALs) pose a considerable barrier to optimal antimicrobial treatment and represent a growing public health concern. However, no comprehensive meta-analysis has been conducted to explore the associations between BALs and clinical outcomes. We aimed to evaluate existing evidence on the clinical outcomes associated with BALs to determine their global burden.

Methods

In this systematic review and meta-analysis, we searched PubMed, the Cochrane Central Register of Controlled Trials (CENTRAL), and Embase from Jan 1, 2000, to Nov 30, 2024. We included observational and interventional studies that compared clinical outcomes related to the presence or absence of a BAL (as reported or documented in any clinical record), irrespective of patient age or clinical setting. The outcomes assessed included the incidence of surgical site infections, the incidence of infections or colonisation due to multidrug-resistant organisms (MDROs) or Clostridioides difficile, mortality, and length of hospital stay. Pooled estimates were calculated using random-effects models, with subgroup analyses conducted by region, country income level, type of BAL, hospital setting, sample size, age group, and quality of evidence. Publication bias was assessed using Begg's funnel plots and Egger's regression test. This study is registered with PROSPERO (CRD42023484030).

Findings

63 studies were included in this systematic review, of which 60 (95%) were from high-income countries. Studies were done in the Americas (41 [65%]), Europe (15 [24%]), and the Western Pacific region (seven [11%]). Seven studies were of moderate quality and none were classified as low quality. No significant publication bias was detected for most outcomes, except for length of hospital stay (p=0·0062). Overall, BALs were associated with increased rates of surgical site infection (OR 1·60, 95% CI 1·27–2·01; p<0·0001; I²=70·3%), rates of infection or colonisation with both MDROs (1·42, 1·22–1·64; p<0·0001; I²=84·4%) and C difficile (1·26, 1·16–1·37; p<0·0001; I²=56·4%), and length of hospital stay (standardised mean difference 0·06 days, 95% CI 0·05–0·08; p<0·0001; I²=86·1%). BALs were also associated with death at or after 180 days but not with overall, in-hospital, or 30-day mortality.

Interpretation

BALs are associated with an array of adverse health outcomes, especially surgical site infection and infection or colonisation with MDROs and C difficile. Although BALs were associated with longer hospital stays, the observed difference was unlikely to be clinically relevant. The heterogeneity and methodological limitations of the included studies could limit the robustness of some of our conclusions. However, these findings underscore the need to develop and evaluate public health initiatives to curb inaccurate allergy labelling, thereby reducing unnecessary avoidance of first-line β-lactam antibiotics.

Funding

National Natural Science Foundation of China.