Secondary Logo

Journal Logo

Clinical Science

Factors Associated With Successful Referral For Clinical Care of Drug Users With Chronic Hepatitis C Who Have or Are At Risk For HIV Infection

Fishbein, Dawn A MD*‡; Lo, Yungtai PHD§; Reinus, John F MD; Gourevitch, Marc N MD, MPH‡§∥; Klein, Robert S MD*‡§

Author Information
JAIDS Journal of Acquired Immune Deficiency Syndromes: November 1st, 2004 - Volume 37 - Issue 3 - p 1367-1375
doi: 10.1097/01.qai.0000131932.21612.49


Hepatitis C virus (HCV) infection is the leading chronic blood-borne infection in the United States. The overall prevalence is estimated at 1.8%, with ∼2.7 million persons chronically infected.1 Most persons infected are 40 to 59 years old.2 Injection drug users are among those with the highest rates of HCV infection, with an estimated prevalence of 79% 2; they account for 60% of HCV transmission in the United States.1 Of those persons with HCV infection, 75%–85% develop chronic infection; cirrhosis develops in 10%–20% over 20–30 years.2 HCV infection in persons with coexisting alcohol abuse progresses more rapidly to cirrhosis, end-stage liver disease, and hepatocellular carcinoma.3 An accelerated course may also be associated with HIV infection.4 With improved survival and aging of persons with HIV infection resulting from the availability of highly active antiretroviral therapy,5 the prevalence of end-stage liver disease will continue to rise.

Treatment of hepatitis C has become increasingly effective with the introduction of pegylated interferon in combination with ribavirin. The overall efficacy is estimated at 50%–60%, with higher rates associated with infection due to HCV genotypes 2 and 3.6 Until recently, drug users (DUs) and persons with HIV infection had largely been excluded from clinical trials and treatment because of concerns about active substance abuse, coexisting psychiatric illness, nonadherence, perceived difficulty engaging in treatment, and inadequate control of HIV coinfection. In addition, because of complications from drug use behaviors, HIV infection, or antiretroviral therapy, some such persons are not eligible for treatment of HCV infection because of comorbid illness. Even for those individuals who are eligible, therapy for HCV infection may be poorly tolerated due to adverse effects such as depression, anxiety, and hematopoietic suppression.

Individualizing treatment, especially for DUs and those with psychiatric illness, has been proposed,7 and in 2002 the National Institutes of Health Hepatitis C Consensus guidelines were updated with this recommendation, along with the recommendation to increase the availability of treatment to those persons previously ineligible for clinical trials.8

It is estimated that the morbidity and mortality resulting from hepatitis C will increase dramatically over the next few decades, as will the cost of treating this chronic disease. Therefore, the ability to successfully treat HCV infection in DUs, especially those who may continue to be active substance abusers or have coexisting HIV infection, is an urgent challenge9 that has only recently begun to be addressed. We have been prospectively studying the natural history of HCV infection in a research cohort of DUs, with expedited referral of those with active infection for clinical care. This report describes the rates of and factors associated with successful evaluation and subsequent treatment of HCV infection among DUs in the Bronx, New York City.


Study Participants

We enrolled 557 current and former opiate-addicted HIV-seropositive and HIV-seronegative DUs in a prospective study of the natural history of HCV infection beginning in November 1999, as previously reported.10 Subjects were recruited from participants in a longitudinal study of HIV infection; all had been enrolled in methadone maintenance treatment at Montefiore Medical Center at the time of enrollment in that study.11 Study participants underwent standardized interviews to gather data on demographics, sexual and drug-using behaviors, and medical history, including symptoms and treatment of hepatitis; laboratory assays included HIV-1 (HIV) antibody and viral load testing, determination of CD4+ lymphocyte subsets, hepatitis B surface antigen and core antibody determinations, HCV antibody testing, RNA level determination, and genotyping. Serologic assays were done on batched specimens stored at −70°C. Participants received reimbursement of $35.00 and assistance with transportation when necessary for each research visit. All participants received pretest and posttest counseling with respect to the meaning of and possible treatments for both HIV and HCV infections and ways to minimize transmission risk. Subjects with detectable HCV RNA were offered expedited referral for HCV evaluation and possible treatment. Appointments were scheduled by research personnel for evaluation by or under the supervision of a single hepatologist. Evaluations were made by a hepatologist in either his clinic or private office, both of which are within a 2– to 3–city block radius from the research interview site.

For persons who accepted referrals but who did not keep their clinical appointments, research study personnel scheduled additional appointments, to a maximum of 3 appointments. Participants who declined referral were asked the reasons for their decision, and those stating that they were receiving clinical care for HCV infection elsewhere were asked to provide consent for research staff to contact their clinicians for details of their care. Participants were evaluated for treatment by, or under the supervision of, the hepatologist using a clinical protocol developed at the onset of the study, based on generally accepted clinical criteria at the time. Liver biopsy was recommended for all patients considered as possible candidates for treatment of HCV infection who did not have a contraindication to biopsy. Treatment would be recommended for all subjects with histologic grade 2, stage 2 or more severe disease, elevated serum alanine aminotransferase level, and age of younger than 60 years, unless there were specific contraindications. For this analysis, medical records were reviewed in April 2003 to determine whether appointments were kept, the extent of clinical evaluations completed, and the details of any treatment given for HCV infection.

The Institutional Review Board for the Protection of Human Subjects of Montefiore Medical Center approved the study, and written informed consent was obtained from all participants.

Statistical Analysis

Four outcomes were analyzed for participants offered referral for clinical evaluation: accepted referral for clinical care, arrived at appointment for clinical care, had liver biopsy, and started treatment of hepatitis C.

CD4+ lymphocyte count was analyzed both as a categorical variable (<200, 200–500, and >500/mm3) and as a continuous variable. HIV viral load and HCV RNA levels were log10 transformed to more closely approximate a normal distribution. χ2 or Fisher exact tests were used in univariate analysis of categorical variables. Logistic regression models were used to identify factors associated with the aforementioned outcomes. Variables with P < 0.25 on a univariate basis or that were biologically plausible were included in multivariate models. Separate logistic regression models were fitted to data for all participants and for those who were HIV seropositive. SPSS statistical software (version 10; SPSS, Inc., Chicago, IL) was used for analysis.

Independent factors examined for associations with study outcomes included the following: age; sex; race; HCV RNA level; history of injection or snorting drug use; self-reported current heroin, cocaine, and alcohol use (including heavy use of >12 drinks per week); age at first injection; HCV genotype; and HIV serostatus. Education level (greater than or less than high school education), children living at home, domicile (eg, living in shelter or homeless), employment, financial and marital status, and current life stressors (eg, recent change in living, relationship, or employment situation) were also examined. For HIV-seropositive participants, we also examined CD4+ lymphocyte count, HIV RNA level, and self-reported antiretroviral treatment. All multivariate models retained alcohol and drug use variables, so that results were also adjusted for substance abuse behaviors.


Study Sample

Baseline characteristics of the 228 subjects with detectable HCV RNA are shown in Table 1. Seventy-two percent of the subjects were male; the mean age was 46 years. Nearly two thirds of subjects were Hispanic, with the remainder approximately equally divided among black and white. On average, >2 decades had passed since those participants who had ever injected first injected drugs. HCV genotype 1a and 1b were most common, found in three quarters of the sample; the median HCV RNA level was 500,500 IU/mL. Slightly more than one half of subjects were HIV seropositive; the median CD4+ lymphocyte count was 256/mm3, and the median HIV RNA level was 909 copies/mL. Of HIV-seropositive participants, 27% had undetectable HIV viral loads, and 22% had CD4+ lymphocyte counts of ≥500/mm3.

Table 1
Table 1:
Baseline Characteristics of 228 Study Participants With Detectable HCV RNA

Clinical Referrals

The referral outcomes for subjects offered clinical evaluation are shown in Figure 1. Clinical referral was offered to 228 subjects with chronic HCV infection; 127 (55.7%) accepted the referral, but only 54 (42.5%) of those given ≥1 appointments actually arrived for clinical care. Four additional patients actually had been clinically evaluated under the supervision of the study hepatologist before referral but had not reported this fact, and 1 reported having been evaluated by another clinician. Participants were given an average of 1.4 appointments. Of the 54 referred participants who arrived for clinical evaluation, 40 (74.1%) did so at the first appointment, 11 (20.4%) did so at the second, and 3 (5.5%) did so at the third. Sixty-eight (53.5%) of the 127 subjects who accepted referral never underwent clinical evaluation before the end of the study period.

Outcomes of referral for clinical evaluation of hepatitis C among 228 DUs with detectable serum levels of HCV RNA. Appts indicates appointments; Rx, treatment.

Reasons Referrals Declined

Of the 101 participants with detectable HCV RNA who declined referral for clinical evaluation, 99 (98%) provided explanations, including self-reported clinical care elsewhere (62.4%), not interested or too busy (15.8%), and not ready but would reconsider decision in the future (8.9%). Additional reasons reported by from 1% to 3% of individuals each included active substance abuse, fear of biopsy or treatment, told by other physician that treatment was not necessary, unable to cope, unable to keep appointments, end-stage liver disease, and prior evaluation by the study hepatologist. No confirmation of most of these self-reports was available. However, for 9 of 10 participants reporting evaluation or care elsewhere at our medical center, review of medical records failed to corroborate the self-reports.

Reasons for Failure to Keep Appointments

Of the 68 individuals who accepted referral for evaluation but failed to keep any appointments, 43 (63.2%) provided explanations, of which the most common was that the individual forgot (13.7%). Additional reasons provided were inability to take time off from work (8.2%), desire to deal with other medical or social problems first (6.8%), not ready but would reconsider decision in the future (5.5%), frightened (4.1%), not interested or too busy (4.1%), and previously evaluated by the study hepatologist (4.1%). From 1% to 3% of subjects each reported active substance abuse, no means of transportation to the clinical site, end-stage liver disease, told by another physician that treatment was not necessary, and clinical care at another site as reasons for failure to keep any appointment; 1 patient died before a scheduled appointment.

Results of Clinical Evaluation

The results of clinical evaluations of the 54 patients seen by or under the supervision of the hepatologist as a result of study referral are shown in Figure 2. The median time between initial appointment date for clinical evaluation and outcome assessment was 6 months (range, 0–22 months). In 12 patients (22.2%), hepatitis C treatment was considered contraindicated due to medical comorbidities, undetectable HCV RNA when retested by the clinician, or advanced age. For 14 patients (25.9%), it was decided to defer hepatitis C treatment until other active clinical conditions were brought under control, including psychiatric illness, active substance abuse, and advanced immunodeficiency. Of 28 patients (51.8%) recommended for treatment of HCV infection, 16 (57.1%) failed to keep follow-up appointments. Only 12 (42.8%) of these subjects had liver biopsies; for 6, treatment of HCV infection was scheduled to begin, and 2 were lost to follow-up. On the basis of liver biopsy results, treatment was deferred in 4 patients.

Results of clinical evaluations of 54 subjects evaluated as a result of study referral. ETOH indicates alcohol.

Five additional participants received documented clinical evaluation by or under the supervision of the study hepatologist before this study. These subjects included 4 who accepted referral but were found on record review to have been evaluated before enrollment and 1 who did not accept referral, indicating prior evaluation elsewhere. Of these 5 subjects, 3 had liver biopsies and 3 received hepatitis C treatment.

HIV/HCV-Coinfected Subjects

Of the 228 subjects referred for clinical evaluation, 125 were HIV seropositive; 61 (48.8%) of the 125 subjects accepted referral. Twenty-one (34.4%) of the 61 subjects kept appointments for evaluation; 5 underwent liver biopsy, and the only individual treated for HCV infection had been treated before enrollment in our study.

Factors Associated With Study Outcomes

HIV coinfection was the only factor significantly associated with participant acceptance of referral for clinical evaluation by univariate (odds ratio [OR], 0.53; 95% confidence interval [CI], 0.31–0.91) and multivariate (adjusted OR [ORadj], 0.51; 95% CI, 0.30–0.88) analyses (Table 2). Of coinfected persons, those with a history of injection drug use were more likely to accept a referral (ORadj, 3.60; 95% CI, 1.08–11.96) (Table 3).

Table 2
Table 2:
Factors Associated With Accepting Referral (n = 127) Among 228 Persons Referred
Table 3
Table 3:
Factors Associated With Accepting Referral (n = 61) Among 125 HIV-Seropositive Persons Referred

For persons accepting clinical referrals, factors associated with keeping appointments are shown for all persons in Table 4 and for those with HIV coinfection in Table 5. Age (ORadj, 1.06/y; 95% CI, 1.00–1.12), and among those with HIV coinfection, HIV-1 viral load (ORadj, 0.50/log10; 95% CI, 0.26–0.94) and Hispanic ethnicity (ORadj, 0.26; 95% CI, 0.07–0.89) were independently associated. CD4+ lymphocyte count (<200/mm3 [OR, 0.19; 95% CI, 0.04–0.86], 200–500/mm3 [OR, 0.21; 95% CI, 0.05–0.91], and >500/mm3 [OR, 1.0]) was associated by univariate analysis but was no longer significant after adjusting for HIV-1 viral load, ethnicity, and drug or alcohol use. HIV-seropositive persons and unemployed subjects were less likely to arrive at a clinical appointment, but the findings did not achieve statistical significance. For persons who underwent clinical evaluation, liver biopsy was less likely for those who had recently used heroin or cocaine (OR, 0.71; 95% CI, 0.58–0.87). Because of the small sample size for liver biopsy and treatment outcomes, no significant associations were found by multivariate analysis.

Table 4
Table 4:
Factors Associated With Arriving at Clinical Appointment (n = 54) Among All 127 Participants Who Accepted Referral
Table 5
Table 5:
Factors Associated With Arriving at Clinic Appointment (n = 21) Among 61 HIV-Seropositive Subjects Who Accepted Referral

To determine the effect that inclusion of the additional 5 patients who had documented clinical care under the supervision of the study hepatologist before enrollment in this study would have on our findings, we repeated the analysis to include these patients. The only additional significant independent association found was that among HIV-seropositive participants: recent drug snorting was associated with arrival for clinical evaluation (ORadj, 7.02; 95% CI, 1.12–44.02). In addition, the association of Hispanic ethnicity with not arriving for clinical evaluation was no longer significant. No significant interactions remained in the final regression models.


Despite counseling about HCV infection and the need for medical evaluation, only one quarter of DUs with active HCV infection who were offered expedited referral for clinical care actually arrived at an appointment. Even taking into account the small number of participants with documented prior clinical evaluation and the substantial number who reported evaluation elsewhere, nearly one half of individuals with active HCV infection did not receive any clinical evaluation. Of those subjects accepting referral and given expedited appointments, most never received clinical evaluations during the study period.

DUs have a long history of decreased access to medical care12–14; associated factors include social instability15 and active drug use.16 Enrollment in methadone maintenance treatment has been associated with improved access to health care services.12,17 It has been demonstrated that when primary care services are included on-site in a methadone maintenance treatment program, >80% of DUs voluntarily use these services.16 Patients with a longer history of drug use and those with a stable income were more likely to remain in treatment.15 Among the most important predictors of access to health services for the inner city population and of receiving preventive care are having a regular source of care and the relationship with a primary care provider.18 However, even with the greatly increased access to care provided by our referral process, a substantial proportion of DUs did not use the available services, suggesting that factors other than access are important and require further investigation. Of great interest is that even among the selected cohort of research visit–adherent individuals studied, utilization of important clinical care interventions was poor. This suggests the need also for further studies on the possible role in clinical care, including effectiveness and cost benefits, of inducements typically used in research studies such as ours: continuity in interactions with the same health care provider, provision when necessary of transportation to and from visits, and cash reimbursements.

HIV-seronegative status was associated with accepting referral for clinical evaluation in this study. It seems possible that persons with HIV infection may be overwhelmed by many clinical appointments, a need to focus predominantly on HIV infection, or the prospect of additional therapies complicating antiretroviral regimens (although the reported use of antiretroviral drugs was not associated with any of the study outcomes). Another possibility may be that patients with HIV infection have difficulty establishing a new relationship with yet another health care provider. The frequency and possible rapid progression of hepatitis C in persons coinfected with HIV19 highlight the urgent need to make delivery of hepatitis C care acceptable to DUs.

It has been suggested that enrollment in a methadone treatment program increases compliance with treatment of HCV infection.20 Although all study participants previously had been in methadone treatment programs, ongoing participation in methadone treatment was not predictive of arriving for clinical evaluation or of being treated. Compliance with research visits in our study was ∼85%; however, compliance with clinic visits remained poor. It seems likely that DUs not participating in research studies would be even less likely to keep clinical appointments. Concern about the possible cost of care was unlikely to be an important consideration, since 90% of study participants had Medicaid, Medicare, or private insurance.

Only older age among all persons accepting clinical referral was independently associated with arriving at a scheduled appointment. For HIV-seropositive participants, lower HIV viral load and ethnicity other than Hispanic were also associated with arriving for clinical care. Knowledge that increased age at acquisition and duration of infection are adverse risk factors for progression of hepatitis C8 may account for older age being a predictor of arriving at a scheduled clinical evaluation. Early successful treatment of hepatitis C for persons with milder liver disease before onset of cirrhosis, however, should improve prognosis. Therefore, efforts to engage younger DUs with hepatitis C in care are warranted. Lower HIV viral load predicted greater adherence with scheduled clinical care. It may be not only that lowering HIV viral load with successful antiretroviral therapy improves the likelihood of a patient being eligible for hepatitis C treatment as a result of improvement in the degree of immunodeficiency but also that the adherence skills demonstrated by patients who have successfully controlled HIV replication will improve their chances of adhering to scheduled clinical appointments. It is not clear why Hispanics were less likely to arrive at appointments. Research interviewers were bilingual, and our medical center has a large Latino population and Spanish translators available. Nonetheless, additional attention to possible cultural differences and language barriers between health care providers and patients may be important.

It is of interest that when the sample size of this study was increased slightly by inclusion of 5 additional persons receiving care before study referral, recent drug snorting became a significant predictor of arrival for evaluation for HIV-seropositive participants. As previously reported, snorting of drugs was associated with increasing age in this study cohort, but snorting remained independently associated with HCV infection after adjustment for age.10 We previously postulated that there may be an inverse relationship of snorting with unmeasured correlates of injection frequency,10 so that it is possible that it is really unmeasured drug injection that predicts receipt of clinical evaluation.

Several study limitations should be noted. Since only small numbers of individuals had liver biopsy or received treatment of hepatitis C, we were unable to identify factors independently associated with these outcomes. Furthermore, this study included only current or former opiate addicts, and the results should not be generalized to other groups. Finally, we were unable to confirm most self-reported receipt of clinical evaluations by health care providers at other sites.

In conclusion, despite expedited referrals for clinical care, few DUs with detectable HCV RNA received an evaluation, and only a very small proportion of these subjects received treatment. Since increasingly effective treatment of HCV infection is available, better methods are urgently needed to improve treatment of HCV-infected DUs, including those coinfected with HIV. Future studies should assess the effectiveness of combining on-site, point-of-care hepatitis C treatment with drug treatment programs and HIV clinical care and explore ways to induce DUs to receive evaluation and treatment of hepatitis C when access to such care is available.


The authors thank Randy Teeter for study coordination and data management, Donna Buono for programming, and all of the study participants and staff.


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. Centers for Disease Control. Hepatitis C. Available at: http://www.cdc. gov/ncidod/diseases/hepatitis/c_training/edu2003.
3. Ostapowicz G, Watson KJR, Locarnini SA, et al. Role of alcohol in the progression of liver disease caused by hepatitis C virus infection. Hepatology. 1998;27:1730–1735.
4. Sulkowski MS, Thomas DL. Hepatitis C in the HIV-infected person. Ann Intern Med. 2003;138:197–207.
5. Fultz SL, Justice AC, Butt AA, et al. Testing, referral, and treatment patterns for hepatitis C virus coinfection in a cohort of veterans with human immunodeficiency virus infection. Clin Infect Dis. 2003;36:1039–1046.
6. Fried MW, Shiffman ML, Reddy R, et al. Peginterferon alfa-2a plus ribavirin for chronic hepatitis C virus infection. N Engl J Med. 2002;347:975–982.
7. Edlin BR, Seal KH, Lorvick J, et al. Is it justifiable to withhold treatment for hepatitis C from illicit-drug users? N Engl J Med. 2001;345:211–214.
8. National Institutes of Health. National Institutes of Health Consensus Development Conference Statement, Management of Hepatitis C. June 10–12, 2002. National Institutes of Health Consensus 2002;1–44.
9. Sylvestre DL. Treating hepatitis C in methadone maintenance patients: an interim analysis. Drug Alcohol Depend. 2002;67:117–123.
10. Strasfeld L, Lo Y, Netski D, et al. The association of hepatitis C prevalence, activity, and genotype with HIV infection in a cohort of New York City drug users. J Acquir Immune Defic Syndr. 2003;33:356–364.
11. Selwyn PA, Alcabes P, Hartel D, et al. Clinical manifestations and predictors of disease progression in drug users with human immunodeficiency virus infection. N Engl J Med. 1992;327:1697–1703.
12. Sambamoorthi U, Warner LA, Crystal S, et al. Drug abuse, methadone treatment, and health services use among injection drug users with AIDS. Drug Alcohol Depend. 2000;60:89.
13. Sherer R. Adherence and antiretroviral therapy in injection drug users. JAMA. 1998;280:567–568.
14. Strathdee SA, Palepu A, Cornelisse PGA, et al. Barriers to use of free antiretroviral therapy in injection drug users. JAMA. 1998;280:547–549.
15. DelRio M, Mino A, Perneger TV. Predictors of patient retention in a newly established methadone maintenance treatment programme. Addiction. 1997;92:1353–1360.
16. Selwyn P, Budner N, Wasserman W, et al. Utilization of on-site primary care services by HIV-seropositive and seronegative drug users in a methadone maintenance program. Public Health Rep. 1993;108:492–500.
17. Gourevitch MN, Wasserman W, Panero MS, et al. Successful adherence to observed prophylaxis and treatment of tuberculosis among drug users in a methadone program. Addict Dis. 1996;15:93–104.
18. Merzel C, Moon-Howard J. Access to health services in an urban community: does sources of care make a difference? J Urban Health. 2002;79:186–199.
19. 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.
20. Davis GL, Rodrigue JR. Treatment of chronic hepatitis C in active drug users. N Engl J Med. 2001;345:215–217.

hepatitis C; drug users; HIV; HIV/hepatitis C virus (HCV) coinfection; treatment referral

© 2004 Lippincott Williams & Wilkins, Inc.