ĐÁNH GIÁ ĐỊNH NGHĨA CA BỆNH NHIỄM KHUẨN LIÊN QUAN ĐẾN CHĂM SÓC Y TẾ CỦA WHO TẠI CÁC CƠ SỞ CÓ NGUỒN LỰC HẠN CHẾ

Abstract

Background: In healthcare facilities with limited resources, the application of European Centre for Disease Prevention and Control (ECDC) healthcare-associated infection (HAI) case definitions faces challenges due to their high technical requirements. The World Health Organization (WHO) has developed simplified definitions to enhance the feasibility of surveillance. This study aimed to evaluate the diagnostic performance and practical applicability of the WHO definitions compared to ECDC standards at the University Medical Center of Ho Chi Minh City (UMC). Methods: A cross-sectional point prevalence survey (PPS) was conducted on 612 inpatients at UMC from October 12-17, 2023. Data were collected from medical records, concurrently applying both WHO and ECDC HAI case definitions. The diagnostic performance of WHO definitions was assessed using sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV), with ECDC definitions serving as the gold standard. Agreement between the two definitions was analyzed using McNemar's test. Results: Among the 612 patients evaluated, WHO definitions identified 223 HAI events, nearly double that of ECDC, reflecting a clear difference in the stringency of the two systems. The level of agreement between the two definitions varied by infection type: statistically significant differences were observed for bloodstream infections (BSI) and urinary tract infections (UTI) (p < 0.01), while hospital-acquired pneumonia (PNEU) and surgical site infections (SSI) showed high concordance. WHO definitions achieved 100% sensitivity for most HAI types, indicating excellent detection capability. However, low specificity for BSI (83.1%) resulted in a low positive predictive value (15.7%), suggesting a risk of over-diagnosis. Conversely, SSI and UTI demonstrated near-perfect specificity and accuracy, while PNEU achieved a good balance between sensitivity and specificity. Conclusion: The simplified WHO HAI case definitions exhibit high detection capability for HAIs, making them suitable for surveillance in resource-limited settings. However, the lower specificity for BSI increases the risk of over-diagnosis and requires careful consideration for broader implementation. The WHO definitions show higher reliability for PNEU, SSI, and UTI, indicating their practical potential in facilities unable to fully implement ECDC standards.