|| Checking for direct PDF access through Ovid
The objective of this study was to assess the impact of universal versus risk-based hepatitis C (HCV) testing of source patients' (SPs) postexposure to blood and body fluids on the HCV exposure rates among healthcare workers.Exposure and test result information between 1993 and 2004 was abstracted from the Johns Hopkins Bloodborne Pathogen Database. A Poisson regression model estimating HCV infection among underlying SPs based on partial testing was developed and applied.After adjusting for the effect of partial testing of SPs, the estimated underlying prevalence of HCV-positive SPs increased slightly during the study period, from 11.9% to 15.1%, but the trend was not statistically significant. Yield curve of HCV-positive SPs rose quickly when SPs' testing rates were low but became flat when SPs' testing rates were high.Reliance on HCV risk factors to screen SPs resulted in an underestimation of the prevalence of HCV in SPs before 1997 when the testing rates were between 15.4% and 25.6%. When SPs' testing rates were above 65%, our model predicted no additional yield of HCV-positive SPs.