Tu1001 Intensive Vaccination Strategies vs. Standard of Care in the Hepatitis C Clinic: a Pragmatic Randomized 2x2 Factorial Trial Amy Hirsch, Yang Liu, Perica Davitkov, Apoorva Chandar, Renee H. Lawrence, Brook Watts, Yngve Falck-Ytter Background: Immunization for Hepatitis A (HAV) and Hepatitis B (HBV) is recommended for Hepatitis C (HCV) patients as superinfection can increase the risk of adverse events. Since immunogenicity increases with each successive dose in the vaccine series, completing the series provides the best chance of maximal protection. Standard of care (SOC) approaches such as provider driven vaccination after serologic testing results in poor series completion rates. Universal vaccination (UV), where patients are vaccinated for HAV/HBV without prior serologic testing, is a commonly identified as an effective vaccination strategy. Patient and provider outreach interventions have also been recommended to improve vaccination rates. Studies of the effectiveness and implementation of these strategies is lacking. Methods: Veteran patients were eligible if they attended a HCV group medical appointment and had evidence of chronic HCV without history of immunization/serologic evidence of immunity to HBV alone or HAV and HBV. Using a 2x2 factorial design assuming no interactions, consenting subjects were randomized to one of four vaccination strategies: 1) UV alone 2) UV plus outreach 3) outreach plus SOC 4) or SOC alone. UV was initiated at enrollment prior to serologic testing. Outreach interventions included a phone and mailed reminder to subjects about the need for vaccination and an electronic reminder sent to subjects' primary care provider. Vaccine series completion rates were compared amongst the two strategies using a chi-square analysis. A series completion failure was defined as failure to receive all three doses of the HBV or combined HAV/HBV series within 14 months of study enrollment. The implementation of each strategy was studied including recording the time per subject required to perform each strategy. Results: Those who received the UV strategy had a reduction in series completion failures compared to SOC RR: 0.46 (0.27; 0.81); p < 0.01. The mean time needed to conduct UV was 7.3 (± 3) minutes. Unnecessary vaccinations based upon follow up serologies were observed. Those receiving UV were more likely to receive an unnecessary vaccination RR: 4.4 (1.7; 12). Subjects receiving the outreach interven- tion did not experience a reduction in series completion failures compared to SOC RR: 0.82 (0.49; 1.67); p=0.3. The mean time needed to perform outreach interventions was 7.75 (± 2) minutes. Conclusions: The UV strategy resulted in a large reduction of series completion failures compared to SOC, with time required similar to the outreach strategy. Outreach efforts were labor intensive and did not result in lower rates of series completion failures. Further study of clinic workflow is needed to reduce the time required to perform UV and to identify methods to avoid unnecessary vaccinations. Tu1002 Non Liver Disease-Related Factors Significantly Impact "Liver Disease- Specific" Quality of Life Scores From the Chronic Liver Disease Questionnaire (CLDQ) Sujit V. Janardhan, Jason Morris, Archita P. Desai, Helen S. Te, Nancy Reau, K. G. Reddy, Donald M. Jensen, Andrew Aronsohn INTRODUCTION: Chronic liver disease (CLD) patients suffer from diminished quality of life (QOL) caused by a multitude of factors. The CLDQ is a 29-item liver disease-specific QOL instrument designed to provide an overall QOL score as well as 6 QOL subdomain scores in CLD patients. While CLD severity is known to influence CLDQ scores, we sought to identify non liver disease-related factors that could also affect CLDQ-measured QOL and to determine how these factors compare to CLD severity in their ability to influence CLDQ scores. METHODS: CLDQ responses from 578 patients enrolled in the Chronic Liver Diseases Database at the University of Chicago were analyzed along with their clinical and socio- demographic data. Univariate ANOVA was performed to determine if demographic factors, CLD diagnoses, Childs-Turcotte-Pugh (CTP) class, comorbid diseases, or recreational drug use correlated with significantly different QOL scores. Factors with ANOVA p values < 0.1 were incorporated into multivariate linear regression to identify independent predictors of changes in QOL scores. Regression coefficients were analyzed to determine which factors were most influential on QOL scores. RESULTS: On univariate analysis, 12 factors correlated with different overall QOL scores while 10 additional factors only correlated with different subdomain scores. On multivariate analysis, factors independently associated with higher overall QOL score included male gender {mean difference (MD) 0.415, p<0.01} and having a private health insurance plan (MD 0.721, p<0.01). These factors were also associated with higher scores in 5/6 and 4/6 QOL subdomains, respectively (p<0.05). Factors independently associated with lower overall QOL score included decompensated (CTP class B or C) cirrhosis (MD -0.955, p<0.01), depression (MD -1.016, p<0.01), illegal drug use (MD -0.957, p<0.01), diabetes (MD -0.600, p<0.01) and anemia (MD -0.654, p=0.04). Decompensated cirrhosis, depression and illegal drug use were associated with decreases in 6/6 subdomain scores, while diabetes and anemia were associated with decreases in 5/6 and 2/6 subdomain scores, respectively (p<0.05). The negative effects of illegal drug use {regression coefficient ± 95% confidence interval (RC) -0.864±0.51} and depression (RC -0.859±0.32) were similar in magnitude to the effect of decompensated cirrhosis (RC -0.816±0.42) on overall QOL score. This was also noted in 6/6 (illegal drug use) and 4/6 (depression) QOL subdomain scores. CONCLUSION: Non CLD-related factors such as depression and illegal drug use are as predictive of diminished "liver disease-specific" QOL as decompensated cirrhosis in CLD patients. These data highlight the need to identify and treat non-CLD related factors that negatively impact QOL in patients with chronic liver disease. Tu1003 Population-Based HIV Screening Rates in Patients With Viral Hepatitis Amanda M. Lynn, Joseph J. Larson, Michael D. Leise Introduction: In the U.S. at least 500,000 persons are coinfected with viral hepatitis and HIV. If HIV remains undiagnosed even in 10% of this population, due to poor screening practices, this would be significant. Therefore, ensuring adherence to current guidelines is S-1003 AASLD Abstracts important to reduce the burden of disease related to coinfection. There is a paucity of data examining population based screening rates for HIV in patients with hepatitis B or C. Aims: (1) Determine screening rates for HIV in a population-based cohort infected with HBV (1994 - 2010) and HCV (1990 - 2010). (2) Examine the change in HIV screening patterns in a viral hepatitis cohort during three eras (1997 - 2000, 2001 - 2004, and 2005 - 2008). (3) Evaluate differences in HIV screening rates between patients with hepatitis B and C, female vs. male gender, and race. Methods: A previously identified group of patients with hepatitis B (1994-2013) and hepatitis C (1990-2013) have been identified through the Rochester Epidemiology Project (REP). The REP affords population-based studies of diseases as it provides access to records of medical care provided to >96% of Olmsted County, MN residents for > 90 years. All HIV screening tests and their results were electronically captured for these population based-cohorts. Screening rates were compared by era, gender, and race using Kruskal-Wallis and Chi-Squared analysis as appropriate. Results: There were 1519 patients with hepatitis B (n=607) or C (n=965), of which 54% received testing for HIV. In the hepatitis B cohort, 273 (39%) were screened for HIV, while 553 (68%) received HIV screening in the hepatitis C cohort, which was a significant difference (p < 0.01). Among all patients with viral hepatitis, screening rates were no different for females (329/630,52%) versus males (489/926, 53%). Screening rates for black patients (124/188,66%) were signifi- cantly higher than whites (419/775,54%) and Asians (109/224,49%) with a p<0.01 for both comparisons. Over the three eras (1997-2000, 2001-2004, 2005-2008) the screening rates were relatively stable for patients with hepatitis B (36%, 34%, 39%, p=0.58) and hepatitis C (65%, 67%, 62%, p=0.55). Conclusion: In a population-based cohort of patients with hepatitis B or C, the HIV screening rates remain sub-optimal in this at risk group, despite societal recommendations for one time HIV screening in all patients between ages 15-65 (USPSTF). Patients with hepatitis C are more likely to receive HIV testing than hepatitis B, likely due to more frequent IV drug use and higher perceived risk of co-infection in this population. Blacks have higher HIV testing rates compared to Asians and whites, for unclear reasons. Tu1004 Nonalcoholic Fatty Liver Disease Is Underrecognized in the Primary Care Setting Pierre Blais, Nisreen Husain, Jennifer R. Kramer, Marc A. Kowalkowski, Hashem El-Serag, Fasiha Kanwal Background: Nonalcoholic fatty liver disease (NAFLD) is a common condition that is increasing in prevalence and disease burden. In the spectrum of care for patients with NAFLD, however, little is known about the processes related to identification, diagnosis and referral. Specifically, the extent to which NAFLD is recognized as a clinical condition in the primary care setting as well as the degree of subsequent evaluation is largely unknown. Therefore, we sought to examine the level and determinants of meeting early processes of care in a cohort of patients with NAFLD. Methods: Using the VA Corporate Data Warehouse, we identified a random sample of patients who visited a large tertiary care VA facility and had >2 alanine transaminase (ALT) values >40 IU/ml >6 months apart between 2004 and 2011 in the absence of any positive results for hepatitis C RNA, hepatitis B surface antigen, or screens for excess alcohol use in the 24 months prior to elevated ALT. We conducted a structured review of the medical charts and abstracted data for the following measures from the primary care providers' notes: (a) recognition of abnormal transaminase levels, (b) mention of NAFLD as a possible diagnosis, (c) recommendations for diet or exercise, and (d) referral to a specialist for further evaluation. Using a multilevel logistic regression model, we identified patient demographic, clinical, comorbidity, and healthcare utilization factors associated with receipt of early NAFLD care. Results: Of 250 patients with NAFLD, 99 (39.6%) had documentation of abnormal ALT in the primary care providers' notes, 54 (21.6%) had NAFLD mentioned as a possible diagnosis, 37 (14.8%) were counseled regarding diet modification and exercise, and 26 (10.4%) were referred to a specialist for further evaluation. Only 99 patients (39.6%) had any of these processes fulfilled. In the multivariable regression model, only the degree of ALT elevation (adjusted OR for ALT ≥80 IU/ml versus <80 IU/ml = 4.4, 95% CI=2.65-7.30) and chronicity of elevation (adjusted OR for ≥50% of ALT values >40 IU/ml versus <50% values elevated = 1.8, 95% CI=1.03-3.14) were associated with receiving specified NAFLD care. There were no associations between receipt of examined care and age, race, and the NAFLD fibrosis score. Only 3% of patients at high risk of fibrosis (NAFLD fibrosis score >0.675) were referred to the specialists. Conclusions: Most patients in care who may have NAFLD are not being recognized and evaluated for this condition. Our data suggest that providers may be using a wrong heuristic in delivering NAFLD care by concentrating care on those with high transaminase levels. Tu1005 Outcomes Associated With Co-Utilization of VA and Medicare Healthcare Benefits in Patients With Hepatocellular Carcinoma Jessica A. Davila, Yvonne H. Sada, Fasiha Kanwal, Sahil Mittal, Christine M. Hartman, Sarah Temple, Sarah B. May, Hashem El-Serag Background: Increasing numbers of patients with hepatocellular carcinoma (HCC) who use the Department of Veterans Affairs (VA) are eligible for cancer care benefits through Medicare. Utilization of two healthcare systems can lead to duplicated services and delays in diagnosis and treatment, but also can result in obtaining complementary care that would otherwise be absent or delayed in one healthcare system. The aim of this study is to examine HCC diagnostic and treatment patterns in patients who are eligible for both VA and Medicare benefits, and to compare outcomes in VA-only users to those who co-utilize VA and Medicare (VA-Medicare). Methods: We identified a national cohort of patients diagnosed with HCC during 2005-2009 who used the VA healthcare system and were also eligible for Medicare within the 1-year before and after their HCC diagnosis. We identified the earliest HCC diagnosis and treatment dates in VA and/or Medicare. We compared HCC stage at diagnosis, receipt of curative treatment (transplant, resection, ablation), and survival between VA-only and VA-Medicare users. Logistic regression was used to examine differences in receipt of curative treatment between VA-only and VA-Medicare users adjusting for clinical factors. Cox proportional hazards analyses were used to evaluate survival. Results: Among the 1,272 AASLD Abstracts