Impact of implementing a clinical decision rule on the use of empirical antibiotics and invasive procedures.

INTRODUCTION

We aim to evaluate the implementation of the modified Step-by-Step algorithm on rates of administration of empirical intravenous antibiotics and lumbar punctures performed among young infants at risk of serious bacterial infections (SBIs).

METHODOLOGY: A retrospective cohort study enrolled febrile infants aged ≤90 days old between December 2017 and June 2023 in KK Women's and Children's Hospital (KKH). The modified Step-by-Step algorithm was implemented as part of a quality improvement (QI) project. SBIs were defined as urinary tract infections, bacteraemia or bacterial meningitis. Propensity score matching was performed, and the rates of empirical antibiotic use and lumbar punctures were compared before and after QI.

RESULTS

Of the 3475 infants, 684 pairs (pre-QI and post-QI) were compared following propensity matching. The median age was 34 (IQR, 11-60) days pre-QI and 37 (IQR, 13-65) days post-QI (p=0.334). 175/684 (25.6%) infants pre-QI and 169/684 (24.7%) post-QI had SBIs (p=0.755). Overall, empirical antibiotic use reduced by 11.4% (513/684 (75.0%) to 435/684 (63.6%), p<0.001) and lumbar punctures performed reduced by 8.8% (378/684 (55.3%) to 318/684 (46.5%), p=0.001). Among infants without SBIs, empirical antibiotic use reduced by 13.6% (339/509 (66.6%) to 273/515 (53.0%), p<0.001) and lumbar punctures performed reduced by 11.0% (245/509 (48.1%) to 191/515 (37.1%), p<0.001).

CONCLUSION

We report a safe and effective reduction in exposure to empirical antibiotics and lumbar punctures performed on febrile young infants following the introduction of the modified Step-by-Step algorithm.