Citations

INTRODUCTION: Nosocomial infections (NIs) in cirrhosis are associated with high mortality but could be preventable. Logistic regression (LR) models have failed to identify high-risk patients. We aimed to develop machine learning (ML) models to predict NI. METHODS: The CLEARED consortium consists of prospectively enrolled cirrhosis inpatients from >120 centers. Using day-of-admission clinical data, 3 ML approaches (random forest [RF], extreme gradient boosting, and neural networks [NNs]) were used to predict NI. Data were split 80:20 for training and testing stratified by the outcome. Models were compared using area under the receiver operating characteristic curve (AUC). RESULTS: In total, 8,263 patients (55.90 ± 13.34 years; 64.1% men) from 127 centers in 37 countries were included. NI developed in 869 (10.5%), a median of 6 (4–11) days of postadmission. Major NIs were respiratory (29.6%) and urinary tract infection (15.7%), and spontaneous bacterial peritonitis (13.5%). NIs occurred more frequently in patients from low/low-middle income countries and those with severe liver disease, alcohol etiology, and admission infections. NIs were associated with inpatient mortality (31.9% vs 8.1%, P < 0.001) and liver transplantation (4.8% vs 1.9%, P < 0.001). Although the RF model (AUC 0.69) showed good calibration (Brier score 0.09), outperforming extreme gradient boosting, neural network, and LR models (AUC 0.66 for all; LR comparison P = 0.043), no model achieved AUC ≥0.80 for clinical utility. At 10% predicted probability threshold, the RF model demonstrated only 75.4% sensitivity, 52.9% specificity, and 15.9% positive predictive value (PPV). DISCUSSION: NIs cannot be accurately predicted from day-of-admission data using ML models, even in a large, prospective, global cirrhosis cohort. Every hospitalized patient with cirrhosis should receive protocolized infection control measures.

SummaryDue to the introduction of newer, more efficacious treatment options, there is a pressing need for policy makers and public health officials to develop or adapt national hepatitis C virus (HCV) control strategies to the changing epidemiological landscape. To do so, detailed, country‐specific data are needed to characterize the burden of chronic HCV infection. In this study of 17 countries, a literature review of published and unpublished data on HCV prevalence, viraemia, genotype, age and gender distribution, liver transplants and diagnosis and treatment rates was conducted, and inputs were validated by expert consensus in each country. Viraemic prevalence in this study ranged from 0.2% in Hong Kong to 2.4% in Taiwan, while the largest viraemic populations were in Nigeria (2 597 000 cases) and Taiwan (569 000 cases). Diagnosis, treatment and liver transplant rates varied widely across the countries included in this analysis, as did the availability of reliable data. Addressing data gaps will be critical for the development of future strategies to manage and minimize the disease burden of hepatitis C.

SummaryFactors influencing the morbidity and mortality associated with viremic hepatitis C virus (HCV) infection change over time and place, making it difficult to compare reported estimates. Models were developed for 17 countries (Bahrain, Bulgaria, Cameroon, Colombia, Croatia, Dominican Republic, Ethiopia, Ghana, Hong Kong, Jordan, Kazakhstan, Malaysia, Morocco, Nigeria, Qatar and Taiwan) to quantify and characterize the viremic population as well as forecast the changes in the infected population and the corresponding disease burden from 2015 to 2030. Model inputs were agreed upon through expert consensus, and a standardized methodology was followed to allow for comparison across countries. The viremic prevalence is expected to remain constant or decline in all but four countries (Ethiopia, Ghana, Jordan and Oman); however, HCV‐related morbidity and mortality will increase in all countries except Qatar and Taiwan. In Qatar, the high‐treatment rate will contribute to a reduction in total cases and HCV‐related morbidity by 2030. In the remaining countries, however, the current treatment paradigm will be insufficient to achieve large reductions in HCV‐related morbidity and mortality.

SummaryThe hepatitis C virus (HCV) epidemic was forecasted through 2030 for 17 countries in Africa, Asia, Europe, Latin America and the Middle East, and interventions for achieving the Global Health Sector Strategy on viral hepatitis targets—“WHO Targets” (65% reduction in HCV‐related deaths, 90% reduction in new infections and 90% of infections diagnosed by 2030) were considered. Scaling up treatment and diagnosis rates over time would be required to achieve these targets in all but one country, even with the introduction of high SVR therapies. The scenarios developed to achieve the WHO Targets in all countries studied assumed the implementation of national policies to prevent new infections and to diagnose current infections through screening.