Share this article on:

Effects of Hepatitis C Virus Coinfection on Survival in Veterans With HIV Treated With Highly Active Antiretroviral Therapy

Backus, Lisa I MD, PhD*; Phillips, Barbara R PhD*; Boothroyd, Derek B PhD*; Mole, Larry A PharmD*; Burgess, Jane RN; Rigsby, Michael O MD; Chang, Sophia W MD, MPH*

JAIDS Journal of Acquired Immune Deficiency Syndromes: August 15th, 2005 - Volume 39 - Issue 5 - p 613-619
doi: 10.1097/01.qai.0000156853.99769.6b
Epidemiology and Social Science

Background: With highly active antiretroviral therapy (HAART) available for patients with HIV, hepatitis C virus (HCV) infection has emerged as a potentially important cause of mortality in coinfected patients. Several studies have investigated the effect of coinfection on mortality, with conflicting results.

Methods: The study cohort consisted of HIV-infected veterans on HAART receiving care at US Department of Veterans Affairs facilities. Inclusion was based on first HAART prescription between January 1997 and February 2003, HCV antibody test result, and baseline CD4 and HIV viral load results within 1 year of starting HAART. We fitted Cox proportional hazards models to study the effect of HCV serostatus on survival time from HAART initiation, controlling for patient demographic and clinical characteristics, facility characteristics, HAART exposure, HAART response, and HCV treatment.

Results: Of 12,216 patients in the study cohort, 38% were HCV-seropositive. During an observation time averaging 3.5 years, 2087 patients died. The adjusted hazard ratio for HCV-seropositive patients was 1.56 (95% confidence interval [CI]: 1.42-1.70; P < 0.0001) without a HAART exposure measure and 1.38 (95% CI: 1.26-1.51; P < 0.0001) with the measure. We obtained similar results in analyses also controlling for HAART response.

Conclusions: HCV seropositivity was independently associated with increased risk of death in a large cohort of HAART-treated HIV-infected veterans. Given the success of HAART in extending the lives of HIV patients, HCV has become an important predictor of their mortality.

From the *Center for Quality Management in Public Health, Veterans Health Administration, Palo Alto, CA; and †HIV and Hepatitis C Program Office, Veterans Health Administration, Palo Alto, CA.

Received for publication February 21, 2004; accepted January 11, 2005.

Reprints: Lisa I. Backus, Center for Quality Management in Public Health, Veterans Health Administration, 3801 Miranda Avenue (132), Palo Alto, CA 94304 (e-mail:

In the United States, up to 300,000 persons are coinfected with HIV and hepatitis C virus (HCV), representing 15% to 30% of all HIV-infected persons and 5% to 10% of all HCV-infected persons.1,2 With the widespread adoption of highly active antiretroviral therapy (HAART), HCV infection has emerged as a potentially important contributor to mortality in coinfected persons.3-6 HAART era studies report conflicting results on the effect of HCV on survival, however.7-10

The US Department of Veterans Affairs (VA) is the single largest provider of HIV care in the United States.11 The objective of this study was to investigate the effect of HCV infection on mortality for HIV-infected patients receiving HAART in VA care. We conducted survival analyses controlling for HAART exposure, HAART response, and other potentially confounding factors.

Back to Top | Article Outline


Data Sources and Study Cohort

Our major data source was the Immunology Case Registry (ICR), which is an automated extract of VA electronic medical record information for HIV-infected patients receiving care at all VA medical facilities.12 Deaths were ascertained using ICR entries, the VA's Beneficiary Information Record Locator System (BIRLS), and the Social Security Administration's (SSA) Master Beneficiary File. The study protocol was approved by the Stanford University Institutional Review Board.

Subjects were veterans who received their first VA-prescribed HAART between January 1, 1997 and February 28, 2003, had an HCV antibody test result, and had a CD4 cell count and HIV viral load result within 1 year before their first VA-prescribed HAART. HAART was defined as receipt of a US Food and Drug Administration-approved protease inhibitor, nonnucleoside reverse transcriptase inhibitor, abacavir, or tenofovir. We excluded veterans with VA-prescribed HAART before 1997 because practice patterns before that year differed appreciably from later practice; in the early HAART era, patients often received newly approved antiretroviral medications sequentially. Patients were classified as HCV-seropositive if they had at least 1 positive HCV antibody result.

Of 28,280 veterans in the ICR with VA health care utilization during the study period, 12,216 (43%) were included in the study cohort. Seven thousand eighty-one (25%) were excluded because they had no VA-prescribed HAART, 4244 (15%) received VA-prescribed HAART before 1997, 1386 (5%) had no interpretable HCV antibody test result, and 3353 (12%) were missing baseline CD4 and/or HIV viral load results.

Back to Top | Article Outline

Outcome Measure and Explanatory Variables

The outcome measure was all-cause mortality, which was measured through March 31, 2003, with near-complete death ascertainment for the study cohort using BIRLS, SSA, and ICR data. Other VA studies have ascertained more than 96% of deaths using BIRLS and SSA data combined.13,14

In the survival analysis, we controlled for other factors, including VA facility characteristics, patient demographic and clinical characteristics, timing and duration of HAART exposure, HAART response, and HCV treatment. VA facilities were categorized by patient caseload (total and HIV) and by geographic region: Northeast, Northcentral, South, and West.15 Patient demographic variables included age at first VA-prescribed HAART, sex, race/ethnicity, and period of first military service. Baseline HIV status was defined by CD4 cell count and HIV viral load results closest to and within 1 year before a patient's first VA prescription for HAART. A patient was considered to have had an AIDS-defining opportunistic infection (AIDS-OI) if he or she had a recorded date of an AIDS-OI on or before the date of first VA-prescribed HAART. Dates of the first AIDS-OI for patients are manually entered into the ICR, with such illnesses defined in the 1993 Centers for Disease Control and Prevention (CDC) AIDS surveillance case definition.16 Patients were also categorized as having psychiatric illness, alcohol abuse, and/or hard drug use (amphetamine, cocaine, or opioid) based on inpatient and outpatient International Classification of Diseases, Ninth Revision (ICD9) codes (see Appendix for ICD9 codes).

We controlled for several variables related to HAART initiation and HAART exposure. Variables related to HAART initiation included year of VA-prescribed HAART initiation and use of a nucleoside reverse transcriptase inhibitor (NRTI) before first VA-prescribed HAART. HAART exposure was modeled as a time-dependent measure based on month of VA prescription fills, accounting for prescriptions covering more than 1 month. Months not covered by a VA HAART prescription were classified as months “off HAART.” HCV treatment was defined as receipt of a VA prescription for interferon α-2b, pegylated interferon, and/or ribavirin.

We constructed several measures of the virologic and immunologic response to HAART. For virologic response, we measured whether HIV viral replication was ever undetectable and whether it was well controlled after starting HAART. For patients with at least 2 follow-up HIV viral load determinations, “well-controlled viral replication” was defined as an undetectable HIV viral load in 75% or more of determinations. Because various HIV viral load tests were used at different VA facilities over time, we judged HIV viral load results of less than 500 copies/mL as “undetectable.” We also assessed the near-term virologic response to HAART as measured by HIV viral load closest to 6 months after starting treatment (and within 3-9 months). For immunologic response, we determined the maximum CD4 cell count at any time after starting HAART and the CD4 cell count closest to 6 months after starting HAART (and within 3-9 months).

Back to Top | Article Outline

Statistical Analysis

We conducted bivariate and multivariate analyses. Comparison between the HCV- seropositive and -seronegative groups used the Pearson χ2 test (categoric variables) and the Wilcoxon rank sum test (continuous variables). Kaplan-Meier survival functions and the log-rank test were used to compare time from first VA-prescribed HAART to death for the 2 groups. Because our information on death did not depend on receipt of VA health care and was essentially complete, observations were censored only at the end of the study period. We also fitted multivariate Cox proportional hazards models and tested the proportional hazards assumption using graphics methods and scaled Schoenfeld residuals.17,18 Baseline HIV viral load met the proportionality assumption when categorized as <50,000, 50,000 to 99,999, and ≥100,000 copies/mL. Numeric baseline CD4 count failed to satisfy the proportionality assumption, prompting a stratified analysis (with a CD4 count <50, 50-199, 200-349, 350-499, and ≥500 cells/μL).

The basic Cox model provided estimates of the effect of HCV serostatus on survival, controlling for facility caseload, geographic region, patient age, sex, race/ethnicity, first period of military service, baseline CD4 cell count, baseline HIV viral load, prior AIDS-OI, psychiatric illness, alcohol abuse, hard drug use, year of first VA-prescribed HAART, NRTI pretreatment, whether the patient was off HAART in a given month, and receipt of HCV medications.

To identify the strongest predictors of mortality, we developed a parsimonious model. The study cohort was randomly split into 2 subsamples, and a model was fitted on each. Covariates were selected from the variables in the basic Cox model by a backward stepwise procedure. Those with a probability value less than 0.05 in both subsample models were included in a final parsimonious model estimated on the full sample. To permit crude comparison of the hazard ratios for key factors, we estimated a variant of the parsimonious model with baseline CD4 cell count as a numeric variable rather than a stratifying variable.

We re-estimated our basic model with HAART exposure excluded. HCV infection increases HAART-induced hepatotoxicity,19-22 which may prompt HAART discontinuation.23 Thus, controlling for HAART exposure is likely to understate the effect of HCV. Other factors, such as alcohol abuse, that are correlated with but independent of HCV infection may also limit HAART exposure, however. If HAART exposure is excluded from the model, an effect of these correlated factors on survival through a reduction in HAART exposure may be erroneously attributed to HCV and the effect of HCV on survival overstated. We estimated the basic model including and excluding HAART exposure to limit our estimate of the effect of HCV on survival.

To explore the possibility of a differential response to HAART as a possible cause of differences in mortality for HCV-seropositive and -seronegative patients, we conducted several additional analyses. We re-estimated our basic model, limiting the analysis to patients who ever achieved an undetectable HIV viral load and to patients with well-controlled viral replication. Limiting the analysis to patients with at least 1 follow-up CD4 cell count, we also re-estimated our basic model, including maximum follow-up CD4 count as an explanatory variable. Finally, limiting the analysis to patients with at least 6 months of observed survival and 6-month HIV viral loads and CD4 cell counts, we re-estimated our basic model, including 6-month HIV viral load and CD4 cell count as explanatory variables.

Data were analyzed using SAS software version 8.2 (SAS Institute, Cary NC) and S-PLUS version 6.1 (Insightful Corporation, Seattle, WA).

Back to Top | Article Outline


Characteristics of the 12,216 study patients distributed by HCV serostatus are shown in Table 1. A total of 4668 (38.2%) were HCV-seropositive. With such a large sample size, the 2 groups differed statistically for nearly all baseline characteristics investigated. Notably, HCV-seropositive patients were slightly older; more likely to be black or Hispanic; and more likely to have diagnoses of psychiatric illness, alcohol abuse, and hard drug use. Although the distributions of baseline CD4 counts differed, HCV-seropositive patients had a median baseline CD4 count comparable to that of HCV-seronegative patients. HCV-seropositive and -seronegative patients were equally likely to have a prior AIDS-OI. The HCV-seropositive group had a lower median baseline HIV viral load than the HCV-seronegative group. On average, HCV-seropositive patients were in VA care longer before their first VA-prescribed HAART (7.0 vs. 4.5 years; P < 0.0001) and were more likely to have been prescribed NRTIs before HAART. On average, HCV-seropositive patients were exposed to HAART for fewer months than HCV-seronegative patients (21.3 vs. 24.9; P < 0.0001). Overall, 167 patients received HCV treatment.



HCV-seropositive and -seronegative patients had a similar virologic response to HAART; however, the former had a diminished immunologic response. Among 11,546 patients with 1 or more follow-up HIV viral load tests, there was no difference in the percentage of HCV-seropositive and -seronegative patients with at least 1 undetectable HIV viral load (81.1% vs. 82.0%; P = 0.2). Of the 10,800 patients with at least 2 follow-up HIV viral load tests, there was no clinical difference in the percentage of patients with well-controlled viral replication (36.9% vs. 38.6%; P = 0.07). Of the 11,584 patients with at least 1 follow-up CD4 cell measurement, HCV-seropositive patients had a lower maximum CD4 cell count than HCV-seronegative patients (median of 447 [interquartile range (IQR): 260-690] vs. 496 [IQR: 290-747] cells/μL; P < 0.0001). This CD4 maximum reflected a net gain of 199 cells/μL for the HCV-seropositive patients compared with 239 cells/μL for the HCV-seronegative patients. Among the 9062 patients with a 6-month HIV viral load and CD4 cell count, HCV-seropositive patients had similar 6-month HIV viral loads (median of <500 [IQR: <500-5974] vs. <500 [IQR <500-5502] copies/mL; P = 0.23) but lower 6-month CD4 counts (median of 307 [IQR: 169-407] versus 336 [IQR: 177-514] cells/μL; P < 0.0001).

Back to Top | Article Outline


During the study period (mean follow-up of 3.5 years, range: 0.1-6.2 years), 2087 patients died. There were proportionally more deaths among HCV-seropositive patients than among HCV-seronegative patients (22.2% vs. 13.9%; P < 0.0001). The unadjusted death rates were 6.4 deaths per 100 person-years for HCV-seropositive patients and 4.0 deaths per 100 person-years for HCV-seronegative patients (P < 0.0001). Kaplan-Meier survival curves with 95% confidence intervals (CIs) for the 2 groups are shown in Figure 1.



We found an impact of HCV on survival in all our multivariate models. In the basic Cox model, the hazard ratio for death for HCV-seropositive patients relative to HCV-seronegative patients was 1.52 (95% CI: 1.37-1.68; P < 0.0001) excluding any measure of HAART exposure and 1.43 (95% CI: 1.29-1.58; P < 0.0001) including the time-dependent measure of HAART exposure. In the parsimonious model, the hazard ratio for HCV-seropositive patients was 1.56 (95% CI: 1.42-1.70; P < 0.0001) excluding HAART exposure and 1.38 (95% CI: 1.26-1.51; P < 0.0001) including it. Hazard ratios for the 5 other covariates in the parsimonious model are shown in Table 2. The impact of HAART exposure was clearly predominant; when off HAART, patients had a markedly increased risk of death. The impact of HAART may be overestimated using the time-dependent measure if patients opt to discontinue HAART immediately before death; we consider this issue further elsewhere in this article. The other predictors of increased mortality were older age, higher baseline HIV viral load, an AIDS-OI before HAART, and NRTI treatment before HAART (likely a proxy for longer standing HIV infection and less optimal early antiretroviral use).



The findings of the parsimonious model were little changed when we measured HAART exposure not as a time-dependent measure but as the HAART refill fraction, defined as the number of months in which the patient had HAART refills divided by the months of survival. The HAART refill fraction failed to satisfy the proportionality assumption and thus was used as a crude measure. In the parsimonious model with HAART refill fraction, the hazard ratio for death for HCV-seropositive patients relative to HCV-seronegative patients was 1.36 (95% CI: 1.24-1.50; P < 0.0001). The other predictors of increased mortality were unchanged, except that a history of alcohol abuse also predicted mortality.

For each of the methods that we used to control for response to HAART, we continued to find a similar negative impact for HCV on survival. When the analysis was limited to the 9427 patients with at least 1 undetectable HIV viral load, the hazard ratio for death for HCV-seropositive patients relative to HCV-seronegative patients was 1.77 (95% CI: 1.54-2.04; P < 0.0001). When the analysis was limited to the 4102 patients who had well-controlled viral replication, the hazard ratio for death for HCV-seropositive patients was 1.69 (95% CI: 1.32-2.15; P < 0.0001). Limiting the analysis to patients with at least 1 follow-up CD4 cell measurement and controlling for their maximum follow-up CD4 cell count yielded a hazard ratio for death for HCV-seropositive patients of 1.34 (95% CI: 1.19-1.49; P < 0.0001). Finally, limiting the analysis to patients with 6-month HIV viral loads and CD4 cell counts and controlling for these measures yielded a hazard ratio for HCV-seropositive patients of 1.50 (95% CI: 1.31-1.70; P < 0.0001).

Back to Top | Article Outline


Our analysis showed that HCV infection increases the risk of death in HIV patients who received HAART, controlling for numerous demographic and clinical factors, including exposure to HAART and response to HAART. Depending on the factors for which we controlled, we found that the risk of death among HAART-treated HIV patients was between 30% and 80% higher for those also infected with HCV.

The published literature provides conflicting results on the impact of HCV infection on mortality among coinfected patients.7-10,24-26 These studies differ in method; differences include the outcome measured (mortality alone vs. mortality combined with an AIDS-OI), inclusion of a measure of HAART and the type of measure, and the time origin. Four of the studies (those from the Italian, Valme, McGill, and Australian HIV Observational Database [AHOD] cohorts) have a small number of outcome events for one of the HCV status groups (less than 25 outcome events) or have a relatively small number of outcome events overall (less than 100 outcome events) and no report of the number by group. Consequently, we cannot have confidence in the results of these 4 studies.7,9,25,26

Among the 4 studies reporting a material number of outcome events for each HCV status group, a likely explanation for some of the conflicting findings involves the extent of HAART use, and thus the prevention of otherwise dominant HIV-related mortality. Two studies (the present study and a study of the Swiss cohort) found a detrimental HCV effect.24 Both set the time origin at HAART initiation and thus include only patients who received HAART. The other 2 studies (those for the Johns Hopkins University and HIV Outpatient Study [HOPS] cohorts) found no HCV effect on survival. Both set the time origin at clinic entry and included patients who never received HAART.8,10 The inclusion of such patients and of survival time before HAART use for other patients may obscure the effect of HCV on survival.

In the Johns Hopkins University study, only 61% of patients received HAART; reported results suggest that the median duration of HAART for the cohort patients was less than 6 months. Although the precise death rate is not given, the authors acknowledge that the death rate in their cohort was high. Based on the percentage of patients who died (15.5%) and the duration of follow-up (2.0 years), a crude estimate is that the death rate among the HCV-seronegative patients in the Johns Hopkins University cohort was nearly twice that observed in the present study.

In addition to the facts that some HOPS patients never received HAART (16%) and that survival time before HAART initiation was included for others, the HOPS study excluded patients who had HCV diagnosed before 1996. These patients likely had longer durations of HCV infection and/or evidence of liver disease that prompted early HCV testing. Thus, the excluded patients may have been more likely to die of liver disease than patients diagnosed later with HCV.

We found that HCV was associated with a decreased immunologic (CD4 cell recovery) but similar virologic response to HAART. This pattern has been seen in some24,25 but not all prior studies.8,9,22,27 After we controlled for immunologic response to HAART, the estimated impact of HCV on survival was little changed, suggesting that the negative effect of HCV on survival does not occur through the mechanism of CD4 cell response. In addition, among those with well-controlled HIV viral replication, the increased risk of death persisted for HCV-seropositive patients, suggesting that HCV does not exert its negative effect in HAART-treated patients through ineffective HIV suppression.

Back to Top | Article Outline

Strengths and Limitations

Our study has 2 major strengths. First is the large cohort. Even with a low mortality rate, we observed many deaths, enabling investigation of the impact of HCV on survival rather than on a combined outcome of time to an AIDS-OI or death. This distinction is important because the observed impact of HCV-HIV coinfection on mortality is likely to occur in large part through progressive liver disease and not through accelerated HIV clinical progression. Because of the large cohort, we were also able to split our sample and estimate multivariate models on each half, thereby minimizing the impact of sample idiosyncrasies. The second major strength is the ability to control for many potentially confounding factors, including HAART exposure, which was the most important predictor of mortality in our parsimonious model, and HAART response, which was defined by measures that have been shown in other studies to predict the survival benefit of HAART therapy.28,29

The use of all-cause mortality as opposed to liver-associated mortality as the outcome may be seen as a limitation. All-cause mortality is an undeniably important outcome, however, and is difficult to misclassify. Conversely, liver-related mortality is difficult to define and may easily be misclassified, because the proximate cause of death from liver failure may be related to infection, neoplasm, encephalopathy, hemorrhage, or hemodynamic consequences, including renal failure.

The use of data collected in an administrative database during routine clinical care has inherent limitations. For instance, differences in the timing and frequency of measuring HIV viral loads after initiation of HAART make it difficult to characterize the longitudinal control of viral replication as a potential confounding variable. A similar problem arises for immunologic recovery. In response, we used several different measures of response to HAART, with similar results for the hazard ratio for mortality associated with HCV.

Our study has other possible limitations. Despite numerous covariates, other unknown or unmeasurable confounding factors may exist. In addition, we relied on HCV antibody tests to identify patients with HCV. Such tests, as opposed to HCV RNA assays, may misclassify patients who resolve their HCV infection. Because only 10% to 15% of HIV-infected HCV-seropositive patients do so, however, we believe that our measure of HCV status is reasonable.30,31 Our use of an overly sensitive measure to define HCV infection would weaken the effect of HCV on mortality; therefore, the true effect may be stronger. Another limitation is that we have no data on the receipt of medications from outside the VA. Also, HCV treatment may have been prescribed for indications other than HCV, such as Kaposi sarcoma. These other indications likely explain why some HCV-seronegative patients seemed to receive HCV treatment. Any measurement error introduced by these limitations is likely to have little effect on the estimate of the impact of HCV infection, however, because few VA patients receive HIV medications outside the VA and only a small percentage of those in the study cohort received HCV treatment. A final limitation is that our cohort consists of US veterans in VA care, who are predominantly men; thus, our results may not be generalizable to populations with different socioeconomic conditions or more women.

Back to Top | Article Outline


With the success of HAART in extending the lives of patients with HIV, understanding how HCV infection affects mortality is essential for improving the care and prognosis of coinfected patients. HCV seropositivity was associated with an increased risk of death in HIV-infected patients who received HAART independent of numerous factors, including HAART exposure and response to HAART. This finding raises the pressing question of whether HCV treatment can ameliorate the observed increase in risk of death. The study period largely preceded the availability of pegylated interferon, and only a small percentage of the HCV-infected patients in our study cohort received any treatment of HCV. Preliminary evidence from other studies indicates that the currently recommended regimen of pegylated interferon and ribavirin has a lower success rate in coinfected patients than in those infected with HCV alone.32,33 In the future, new medications, such as HCV RNA polymerase, helicase, and serine protease inhibitors, may increase the effectiveness and tolerability of HCV treatment in coinfected patients.34 In the meantime, we note that many of our patients carry a diagnosis of alcohol abuse, hard drug use, or psychiatric illnesses in addition to HIV and HCV. Alcohol abuse certainly accelerates the clinical course of HCV,31,35-38 and other drug use and psychiatric illnesses contribute directly and indirectly to poor outcomes in HIV and HCV infection.39-42 Greater emphasis on treatment of these frequent comorbidities may be warranted, especially while awaiting improvements in pharmacologic HCV treatment.

Back to Top | Article Outline


The authors thank Lawrence Deyton and Susan Mather, who persevered in their support of the ICR and the Center for Quality Management in Public Health, and Sergey Gavrilov, whose technical expertise has made the ICR a reality. They extend special thanks to all the VA facility ICR coordinators, without whom none of this work would be possible.

Back to Top | Article Outline


1. Alter MJ, Kruszon-Moran D, Nainan OV, et al. The prevalence of hepatitis C virus infection in the United States, 1988 through 1994. N Engl J Med. 1999;341:556-562.
2. Sherman KE, Rouster SD, Chung RT, et al. Hepatitis C virus prevalence among patients infected with human immunodeficiency virus: a cross-sectional analysis of the US Adult AIDS Clinical Trials Group. Clin Infect Dis. 2002;34:831-837.
3. Bica I, McGovern B, Dhar R, et al. Increasing mortality due to end-stage liver disease in patients with human immunodeficiency virus infection. Clin Infect Dis. 2001;32:492-497.
4. Monga HK, Rodriguez-Barradas MC, Breaux K, et al. Hepatitis C virus infection-related morbidity and mortality among patients with human immunodeficiency virus infection. Clin Infect Dis. 2001;33:240-247.
5. Quintana M, Del Amo J, Barrasa A, et al. Progression of HIV infection and mortality by hepatitis C infection in patients with haemophilia over 20 years. Haemophilia. 2003;9:605-612.
6. Rosenthal E, Poiree M, Pradier C, et al. Mortality due to hepatitis C-related liver disease in HIV-infected patients in France (Mortavic 2001 study). AIDS. 2003;17:1803-1809.
7. Macias J, Melguizo I, Fernandez-Rivera FJ, et al. Mortality due to liver failure and impact on survival of hepatitis virus infections in HIV-infected patients receiving potent antiretroviral therapy. Eur J Clin Microbiol Infect Dis. 2002;21:775-781.
8. Sulkowski MS, Moore RD, Mehta SH, et al. Hepatitis C and progression of HIV disease. JAMA. 2002;288:199-206.
9. Klein MB, Lalonde RG, Suissa S. The impact of hepatitis C virus coinfection on HIV progression before and after highly active antiretroviral therapy. J Acquir Immune Defic Syndr. 2003;33:365-372.
10. Tedaldi EM, Baker RK, Moorman AC, et al. Influence of coinfection with hepatitis C virus on morbidity and mortality due to human immunodeficiency virus infection in the era of highly active antiretroviral therapy. Clin Infect Dis. 2003;36:363-367.
11. Philips BR, Mole LA, Backus LI, et al. Caring for Veterans with HIV Disease. Washington, DC: Department of Veterans Affairs; 2003.
12. Backus LI, Mole LA, Chang S, et al. The Immunology Case Registry. J Clin Epidemiol. 2001;54(Suppl 1):S12-S15.
13. Page WF, Braun MM, Caporaso NE. Ascertainment of mortality in the U.S. veteran population: World War II veteran twins. Mil Med. 1995;160:351-355.
14. Page WF, Mahan CM, Kang HK. Vital status ascertainment through the files of the Department of Veterans Affairs and the Social Security Administration. Ann Epidemiol. 1996;6:102-109.
15. Centers for Disease Control and Prevention. Update: impact of the expanded AIDS surveillance case definition for adolescents and adults on case reporting-United States, 1993. MMWR Morb Mortal Wkly Rep. 1994;43:160-161, 167-170.
16. Centers for Disease Control and Prevention. 1993 revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults. MMWR Recomm Rep. 1992;41(RR 17):1-19.
17. Schoenfeld D. Partial residuals for the proportional hazards regression model. Biometrika. 1982;62:239-241.
18. Grambsch PM, Therneau TM. Proportional hazards tests and diagnostics based on weighted residuals. Biometrika. 1994;81:515-526.
19. Saves M, Vandentorren S, Daucourt V, et al. Severe hepatic cytolysis: incidence and risk factors in patients treated by antiretroviral combinations. Aquitaine Cohort, France, 1996-1998. Groupe d'Epidemiologie Clinique de SIDA en Aquitaine (GECSA). AIDS. 1999;13:F115-F121.
20. den Brinker M, Wit FW, Wertheim-van Dillen PM, et al. Hepatitis B and C virus co-infection and the risk for hepatotoxicity of highly active antiretroviral therapy in HIV-1 infection. AIDS. 2000;14:2895-2902.
21. Nunez M, Lana R, Mendoza JL, et al. Risk factors for severe hepatic injury after introduction of highly active antiretroviral therapy. J Acquir Immune Defic Syndr. 2001;27:426-431.
22. Aceti A, Pasquazzi C, Zechini B, et al. Hepatotoxicity development during antiretroviral therapy containing protease inhibitors in patients with HIV: the role of hepatitis B and C virus infection. J Acquir Immune Defic Syndr. 2002;29:41-48.
23. Sulkowski MS, Thomas DL. Hepatitis C in the HIV-infected person. Ann Intern Med. 2003;138:197-207.
24. Greub G, Ledergerber B, Battegay M, et al. Clinical progression, survival, and immune recovery during antiretroviral therapy in patients with HIV-1 and hepatitis C virus coinfection: the Swiss HIV Cohort Study. Lancet. 2000;356:1800-1805.
25. De Luca A, Bugarini R, Lepri AC, et al. Coinfection with hepatitis viruses and outcome of initial antiretroviral regimens in previously naive HIV-infected subjects. Arch Intern Med. 2002;162:2125-2132.
26. Lincoln D, Petoumenos K, Dore GJ. HIV/HBV and HIV/HCV coinfection, and outcomes following highly active antiretroviral therapy. HIV Med. 2003;4:241-249.
27. Chung RT, Evans SR, Yang Y, et al. Immune recovery is associated with persistent rise in hepatitis C virus RNA, infrequent liver test flares, and is not impaired by hepatitis C virus in co-infected subjects. AIDS. 2002;16:1915-1923.
28. Chene G, Sterne JA, May M, et al. Prognostic importance of initial response in HIV-1 infected patients starting potent antiretroviral therapy: analysis of prospective studies. Lancet. 2003;362:679-686.
29. Grabar S, Le Moing V, Goujard C, et al. Clinical outcome of patients with HIV-1 infection according to immunologic and virologic response after 6 months of highly active antiretroviral therapy. Ann Intern Med. 2000;133:401-410.
30. Villano SA, Vlahov D, Nelson KE, et al. Persistence of viremia and the importance of long-term follow-up after acute hepatitis C infection. Hepatology. 1999;29:908-914.
31. Thomas DL, Astemborski J, Rai RM, et al. The natural history of hepatitis C virus infection: host, viral, and environmental factors. JAMA. 2000;284:450-456.
32. Chung RT, Andersen J, Volberding P, et al. Peginterferon alfa-2a plus ribavirin versus interferon alfa-2a plus ribavirin for chronic hepatitis C in HIV-coinfected persons. N Engl J Med. 2004;351:451-459.
33. Torriani FJ, Rodriguez-Torres M, Rockstroh JK, et al. Peginterferon alfa-2a plus ribavirin for chronic hepatitis C virus infection in HIV-infected patients. N Engl J Med. 2004;351:438-450.
34. McHutchison JG, Patel K. Future therapy of hepatitis C. Hepatology. 2002;36(5 Suppl 1):S245-S252.
35. Benhamou Y, Di Martino V, Bochet M, et al. Factors affecting liver fibrosis in human immunodeficiency virus- and hepatitis C virus-coinfected patients: impact of protease inhibitor therapy. Hepatology. 2001;34:283-287.
36. Di Martino V, Rufat P, Boyer N, et al. The influence of human immunodeficiency virus coinfection on chronic hepatitis C in injection drug users: a long-term retrospective cohort study. Hepatology. 2001;34:1193-1199.
37. Tural C, Fuster D, Tor J, et al. Time on antiretroviral therapy is a protective factor for liver fibrosis in HIV and hepatitis C virus (HCV) co-infected patients. J Viral Hepat. 2003;10:118-125.
38. Bellentani S, Pozzato G, Saccoccio G, et al. Clinical course and risk factors of hepatitis C virus related liver disease in the general population: report from the Dionysos study. Gut. 1999;44:874-880.
39. Palepu A, Tyndall M, Yip B, et al. Impaired virologic response to highly active antiretroviral therapy associated with ongoing injection drug use. J Acquir Immune Defic Syndr. 2003;32:522-526.
40. Porter K, Babiker A, Bhaskaran K, et al. Determinants of survival following HIV-1 seroconversion after the introduction of HAART. Lancet. 2003;362:1267-1274.
41. Dieperink E, Willenbring M, Ho SB. Neuropsychiatric symptoms associated with hepatitis C and interferon alpha: a review. Am J Psychiatry. 2000;157:867-876.
42. Lucas GM, Cheever LW, Chaisson RE, et al. Detrimental effects of continued illicit drug use on the treatment of HIV-1 infection. J Acquir Immune Defic Syndr. 2001;27:251-259.
Back to Top | Article Outline


The ICD9 codes are as follows:

  1. Psychiatric illness ICD9 codes: 290.xx, 293.xx to 302.xx, and 306.xx to 319.xx
  2. Alcohol abuse ICD9 codes: 291.xx, 303.0x, 303.9x, 305.0x, 357.5, 425.5, 535.3x, 571.0 to 571.3, 790.3, 980.0, 980.8, 980.9, E860.0, E860.1, E860.9, and V11.3
  3. Hard drug use ICD9 codes: 304.0x, 304.2x, 304.4x, 304.7x, and 305.5x to 305.7x

hepatitis C; highly active antiretroviral treatment; survival analysis

© 2005 Lippincott Williams & Wilkins, Inc.