Prevalence of antibiotic use in a tertiary care hospital in Italy, 2008-2016.

Background: Few data are available about temporal trends of antibiotic use in hospitalized children. The aim of the current study was to investigate the pattern and trends of antibiotic use over the years 2008-2016 in the largest children's hospital in Italy.

Methods: Annual point prevalence surveys of antibiotic use were conducted by reviewing medical charts of 0-17 year-old children hospitalized for ≥48 h. Prevalence of antibiotic use was computed by year, type of ward and indication. Trends in prevalence over time were evaluated using the Cochrane-Armitage test. Possibile determinants of antibiotic use were assessed at univariate analysis and through a logistic regression model.

Results: Out of 3015 children, 1516 (50.3%) received antibiotics, 58.1% of which for medical/surgical prophylaxis. Prevalence of antibiotic use increased from 42.0% in 2008 to 56.2% in 2016 (p = 0.001). The prevalence of patients receiving antibiotics for medical prophylaxis increased from 6.1% in 2008 to 24.2% in 2016 (p < 0.001), whereas the prevalence of patients receiving antibiotics for surgical prophylaxis significantly decreased (from 13.7 to 11.8%; p = 0.04); no significant temporal trends were found in antibiotic use for treating infections. The administration of third-generation cephalosporins for surgical and medical prophylaxis significantly decreased over time, while the proportion of antibiotics prescribed to treat infections after microbiological investigations significantly increased. Year (ORadj: 1.8 in 2016 compared to 2008, p < 0.001), age (ORadj ≥1.5 in children ≥1 year, compared to infants ≤2 months, p < 0.001), length of stay (LOS) (OR: 1.4 in case of LOS between 8 and 30 days compared to LOS ≤ 7 days, p < 0.001), and type of ward (ORadj: ≥1.3 in intensive-care, surgical and medical-subspecialty units compared to medical units, p < 0.001) were significantly and independently associated with antibiotic use.

Conclusions: Comparing prevalence rates of antibiotic use among hospitals and over time should consider differences in patient characteristics, such as age, ward of hospitalization and length of stay. Over the years, we documented an improvement in the choice of antibiotics prescribed for medical and surgical prophylaxis. However, further efforts are needed to avoid antibiotic misuse for medical prophylaxis, and to reduce the empirical use of broad spectrum antibiotics.