828 Impact of Hepatitis C, HIV, or Both on Survival in Veterans in Care Before and After the Introduction of HAART (1996) S. L. Fultz*1,2, A. C. Justice1,2, C. H. Chang2,3, A. A. Butt1,2, M. Skanderson1, D. Rimland4, M. O. Rigsby5,6 1Veterans Admin Pittsburgh Hlthcare Sys, PA; 2Univ of Pittsburgh Sch of Med, PA; 3Univ of Pittsburgh Graduate Sch of Publ Hlth, PA; 4Veterans Admin Med Ctr, Atlanta, GA; 5Publ Hlth Strategic Hlthcare Group, Veterans Admin, Washington, DC; and 6Yale Univ Sch of Med, New Haven, CT
Objectives: To determine 1) whether a survival difference exists between patients (pts) diagnosed with hepatitis C, HIV, or both, compared to controls, and 2) whether these differences were affected by the introduction of highly active antiretroviral therapy (HAART) in 1996.
Methods: Male HIV+ pts and age-, race-, and site-matched controls were identified from VA administrative databases. Demographic characteristics and dates of death were extracted. Kaplan-Meier survival curves and Cox proportional hazards regression analysis were performed to determine the impact on survival of infection with HIV, hepatitis C, or both, controlling for age and race.
Results: We identified 35,290 HIV+ pts and 34,626 controls with 10,888 and 3,053 deaths, respectively. In the pre-1996 time period, survival differences occurred across all four groups. In a Cox proportional hazards model, HIV diagnosis (Hazard ratio [HR] 6.5) had a larger impact on survival than hepatitis C diagnosis (HR 1.8), while co-diagnosis had an even greater impact (HR 7.9). Post-1996, HIV and hepatitis C diagnoses had equal hazard ratios (HR 2.5 and 2.2, respectively, p = 0.2 for comparison), while co-diagnosis had a larger detrimental effect on survival (HR 5.0). Both models controlled for age and race. All predictors were significant at the p < 0.0005 level. In both controls and HIV+ subjects, the relative mortality hazard of hepatitis C co-diagnosis increased in the post-1996 time period. There was a trend towards a larger increase in the effect for HIV+ pts (p = 0.06).
Conclusions: Significant differences in the relative hazard of mortality existed in pts diagnosed with hepatitis C, HIV or both, prior to the introduction of HAART in 1996. While the hazard of mortality with HIV infection declined after this time period, the relative hazard of mortality due to hepatitis C has increased in both mono- and co-diagnosed patients, and trended towards a larger effect in co-diagnosed patients. The growing impact of hepatitis C on mortality is of note and may reflect an aging cohort effect. The larger increase in mortality among co-diagnosed pts should spur further research into the interaction between HIV and hepatitis C infections and increased clinical focus on management of co-infection.
