First Point-Prevalence Study of Inpatient Antimicrobial Use in Five Public Hospitals in Southern Sri Lanka

Abstract

Objectives Inappropriate antimicrobial use is associated with adverse drug effects and antimicrobial resistance. Understanding prescribing practices is important for optimizing antimicrobial use. We investigated the prevalence, types, and indications for antimicrobials in Sri Lanka.Methods A point-prevalence study was conducted among inpatients at tertiary (1), secondary (1), and primary-level (3) public hospitals in Southern Province, Sri Lanka. From June-Aug 2017, all patients in medical, surgical, pediatric, and intensive care wards were included. Charts were assessed for systemic antibiotic, antifungal, and antiviral use in 1-day point-prevalence studies. Demographics, clinical characteristics, and antimicrobial use data were recorded from charts. Hospital and patient characteristics associated with antimicrobial therapy were assessed using the Chi-square test and Kruskall-Wallis tests. Potentially inappropriate antimicrobial use was defined as the receipt of two beta-lactam antibiotics or two antibiotics with activity against anaerobes or Pseudomonas aeruginosa.Results A total of 1,709 patients were included (69.6% tertiary, 21.7% secondary, and 8.7% primary). Patients were in medical (55.2%), surgical, (27.2%), pediatric (12.9%), and intensive care wards (4.7%). Overall, 54.7% (95% CI, 52.3%-57.1%) of patients were receiving antimicrobials. Antimicrobial use prevalence was similar across hospital type (p= 0.439), but varied by ward type: 43.1% in medical, 68.0% in surgical, 61.1% in pediatric, and 97.6% in intensive care wards (p<0.001). Commonly used antimicrobials were amoxicillin/clavulanate (33.8%), 3rd-generation cephalosporins (23.6%), metronidazole (16.6%), narrow and extended-spectrum penicillins (15.8%), clarithromycin (12.3%), 2nd-generation cephalosporins (10.7%), and carbapenems (10.4%). Common antimicrobial indications were lower respiratory infections (20.7%), soft tissue infections (9.4%), urinary tract infections (9.7%), and surgical prophylaxis (7.8%). Of patients receiving antimicrobials, potential inappropriate use was seen: 9.2% double anaerobic coverage, 7.6% double beta-lactam coverage, and 1.5% double P. aeruginosa coverage.Conclusions We report the first point-prevalence study of antimicrobial use in public hospitals in southern Sri Lanka. Over half of inpatients were receiving antimicrobials. High antimicrobial use and potentially inappropriate antimicrobial use should be addressed by future antimicrobial stewardship efforts.