Introduction

People living with HIV (PWH) experience a disproportionate burden of infection with hepatitis C virus (HCV) and hepatitis B virus (HBV) [1, 2]. In the Asia-Pacific region, research has established a high prevalence of hepatitis coinfection with HIV [3], the association of injection drug use with coinfection, particularly HCV [4–6], and the co-occurrence of hazardous alcohol use in these populations [5]. HIV, hepatitis, drug, and alcohol use each independently impair liver health; their confluence results in high risks of liver-related morbidity and mortality [6, 7]. Understanding health behaviors related to the syndemic conditions can inform potential areas of integrated intervention.

In Vietnam, the prevalence of HBV and HCV coinfection with HIV has been estimated to be 14% and 40%, respectively [8], and HCV coinfection ranges as high as 95–98% in some populations who use injection drugs [9, 10]. However, these prevalence estimates may be outdated [10] and an in-depth examination of drug and alcohol use in these populations is warranted. Additionally, recent work in Vietnam has found discrepancies in HCV coinfection status when self-reports are compared to the medical records of PWH, suggesting that HCV awareness among patients and the screening practices of providers remain inconsistent [11].

In this analysis, we estimate the prevalence of HBV and HCV coinfection among antiretroviral therapy (ART) clinic clients and examine differences in self-reported drug and alcohol use between clients with and without coinfection.

Materials and methods

Results

A total of 440 participants living with HIV enrolled into the trial. Of those, 43 (10%) participants had no hepatitis coinfections, 355 (81%) had an HCV coinfection, 5 (1%) had an HBV coinfection, and 37 (8%) had HBV and HCV coinfections (Table 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Baseline characteristics of participants by HIV/hepatitis coinfection status.

https://doi.org/10.1371/journal.pgph.0003744.t001

The median age of participants was 40 years old (interquartile range (IQR): 36, 44) and the vast majority of participants were male (n = 426 of 440, 97%; Table 1). There was a higher percentage of women in the no HBV or HCV group (n = 11 of 43, 26%) compared to the HBV or HCV(n = 3 of 360, 1%) and HBV and HCV groups (n = 0 of 37, 0%; p<0.01). Because recruitment occurred at ART clinics, nearly all participants reported taking ART (n = 438 of 440, 100%; p = 1.00), and most (n = 370 of 440, 84%; p = 0.47) were HIV virally suppressed.

Among the 42 participants who had a positive HBV surface antigen test at enrollment, 15 (36%) reported receiving a previous HBV diagnosis, all of whom were in the HBV and HCV group (15 of 37, 41%). Among the 392 participants who had a positive HCV antibody test at enrollment, 71 (18%) reported having been previously diagnosed with HCV (HCV only: n = 59 of 355, 17%; HBV and HCV: n = 12 of 37, 32%). Few participants in the entire study population (n = 15 of 440, 3%, p = 0.18) reported having been previously diagnosed with Hepatitis A.

We observed notable differences between coinfection groups in ever having injected drugs (p<0.01) and recent injection drug use (p<0.01; Table 1). Among those without hepatitis coinfection, few participants reported ever having injected drugs in their lifetime (n = 3 of 43, 7%) and none (n = 0 of 43, 0%) reported injection drug use in the last 3 months. In the HBV or HCV group, the majority (n = 317 of 360, 88%) of participants reported ever injecting drugs, with a minority (n = 102 of 360, 28%) reporting injecting drugs in the past 3 months. In the HBV and HCV group, nearly all participants reported ever injecting drugs (n = 36 of 37, 97%) with about half reporting injecting drugs in the past 3 months (n = 16 of 37, 43%). An increasing trend was observed across coinfection groups regarding non-injection drug use: 7% (n = 3 of 43), 42% (n = 151 of 360), and 49% (n = 18 of 37) of participants in the no HBV or HCV, HBV or HCV group, and HBV and HCV groups reported non-injection drug use in the past 3 months, respectively (p<0.01).

Reported alcohol consumption across all three infection groups varied slightly with median AUDIT scores of 10 (IQR: 8, 13), 12 (IQR: 9, 16), and 11 (IQR: 9, 17) for the no HBV or HCV, HBV or HCV, and HBV and HCV, respectively (p = 0.01). Correspondingly, self-reported alcohol consumption from the past 30 days also differed slightly (Fig 1). Though the number of drinks on drinking days (median range: 3–4; p = 0.46) and the number of heavy drinking days (median range: 2–3; p = 0.66) were similar across groups, participants in the no HBV or HCV group reported a median of 10 (IQR: 6, 21) drinking days in the previous 30 days, while those in the HBV or HCV group reported a median of 20 (IQR: 10, 30) drinking days and those in the HBV and HCV group reported a median of 22 (IQR: 13, 28) drinking days (p<0.01).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Participant alcohol use by HIV/hepatitis coinfection status.

https://doi.org/10.1371/journal.pgph.0003744.g001

Efforts to reduce injection drug use among those who had ever injected drugs varied across coinfection groups (p = 0.01), though there were no statistically significant differences when we looked at increasing alcohol while reducing injection drug use (p = 0.20). Among those who reported ever having injected drugs in the no HBV or HCV group, 1 of 3 (33%) reported trying to reduce injection drug use, and that person (100%) reported increasing alcohol use while reducing injection drug use. In the HBV or HCV group, 235 of 317 (74%) reported trying to reduce injection drug use, 80 (34%) of whom reported increasing alcohol use during those attempts. Among those in the HBV and HCV group who ever injected drugs, 33 of 36 (n = 92%) reported trying to reduce drug use, 14 (42%) of whom reported increasing alcohol use during their attempts.

Discussion

Historically, HCV prevalence estimates in the general Vietnamese population range from 0.2–3.3% but have been much higher among those living with HIV (22.9–89.0%), as we saw in our population [18–21]. Both HIV and HCV can be transmitted through IDU, and while our sample limits our inferences to those living with HIV, our results support a confluence of HIV, HCV, and IDU as illustrated by the high prevalence of HCV among PWH and a higher reported prevalence of IDU among those with HIV and hepatitis coinfection [11]. Of note, HCV testing in our study was conducted with an antibody test, and we are unable to determine whether the positive results were from active or previously cleared infections. However, nearly 9 of 10 participants tested positive for HCV antibodies, indicating a high lifetime incidence of HCV in this population.

Prevalence of HBV coinfection in our study population was similar to estimates from the general population [9, 22, 23]. While HBV can be transmitted intravenously, in Vietnam, most incident HBV cases are thought to be transmitted perinatally [24]. Our sample size of persons with HBV and HIV (n = 5) did not enable us to examine differences in IDU compared to other coinfection statuses (HIV alone or HCV coinfection) and as such, we are unable to infer HBV-related behavioral associations from this population [24, 25]. However, understanding HIV and HBV coinfection is critical given the overlap in antiviral treatment: choosing antiviral regimens active against both HIV and HBV infections as well as any other concomitant condition is imperative for long term care [26].

Notably, among those who tested positive for HBV antigen and HCV antibodies, the majority had not received a diagnosis prior to the study. Only 36% and 18% of participants who tested positive had previously received a positive diagnosis for HBV and HCV, respectively. Screening for these infections is important particularly in this population given the high coinfection rates and benefits of early treatment [27]. Additionally, because participants were recruited from ART clinics, our results highlight the importance of this point of engagement in care for both coinfection testing and treatment referral as well as substance use assessment and outreach.

We observed more reported recent drinking days in the HBV or HCV group and HBV and HCV group as compared to the no HBV or HCV group. Furthermore, we found that among those who reported trying to reduce injection drug use, one in three reported increasing alcohol consumption during those attempts. Our data are cross-sectional and largely derived from self-report and thus we are unable to assess any causal or temporal relationships. Furthermore, our study population limits our inferences to PWH seeking ART care and reporting hazardous alcohol use. However, the confluence of alcohol use and hepatitis coinfection in this population highlights the entwined nature of these syndemic conditions and the potential for increased hepatic morbidity. As HIV coinfection has shown to hasten liver disease progression among those with HCV, and hazardous alcohol consumption increases the risk of cirrhosis and fibrosis, cooccurring HCV, HIV, and hazardous alcohol use have the potential for a synergistically negative effects [28, 29]. Therapies for HBV and HCV are becoming more widely available in Vietnam, and successful interventions for alcohol and drug use have been developed and adapted locally [12, 26, 30, 31]. Integrating screening and treatment for these syndemic conditions as coordinated care will benefit patient care and prognostic outcomes [27].

The syndemics of HIV, hepatitis, IDU, and alcohol use are deeply entangled and challenging to parse out. However, the differences we observed between coinfection groups highlights patterns of substance use within a vulnerable population. Future research should assess barriers to substance use and hepatitis screening and care referral programs in the HIV care setting in Vietnam.

Supporting information

Acknowledgments

We would like to thank the study staff in Thai Nguyen, Vietnam for their dedication, as well as all participants for their time and contributions.