Higher HCV antibody prevalence among Indigenous clients of needle and syringe programs
James Ward, Aboriginal and Torres Strait Islander Health Program, National Centre in HIV Epidemiology and Clinical Research University of New South Wales, 45 Beach Street, Coogee, NSW 2010; e-mail: Jward@nchecr.unsw.edu.au
Objective: To compare prevalence of hepatitis C virus (HCV) antibody and associated risk behaviours among Indigenous and non-Indigenous participants in the Australian Needle and Syringe Program Survey.
Methods: During 1 or 2 weeks each October from 1998 to 2008, clients of participating needle and syringe programs (NSPs) completed a self-administered questionnaire on demographics and risk behaviour and provided a capillary blood sample for HIV and HCV antibody testing. After de-duplication, 16,132 individuals participated during the 11 years, of whom 1,380 (8.6%) identified as Indigenous.
Results: Higher proportions of Indigenous than non-Indigenous participants were HCV antibody positive (57% versus 51%, p<0.001). In an overall multivariable analysis, Indigenous status (OR 1.17; CI 1.03–1.32) and female gender (OR 1.25; CI 1.16–1.35) were independently associated with HCV antibody seropositivity. Indigenous participants also reported higher rates of risk behaviour, including receptive sharing of needle syringes (21% vs 16%; p<0.001), receptive sharing of ancillary injecting equipment (38% vs 33%; p<0.001), having been injected by others (18% vs 13%; p<0.001), and injecting in public (54% vs 49%; p<0.001).
Conclusion and implications: These results highlight the need for targeted, culturally appropriate programs to minimise risks for bloodborne viral transmission among Indigenous people who inject drugs.
People who inject drugs (PWID) are at increased risk of acquiring bloodborne viral infections (BBVIs), including HIV and hepatitis C virus (HCV).1 As part of Australia's public health response, more than 3,000 needle and syringe programs (NSPs) have been established nationally. Compelling evidence exists of their public health benefits: since their establishment in 1986, NSPs have been credited with preventing an estimated 57,050 cases of HIV infection and 117,667 cases of HCV infection, resulting in health-care cost savings of more than $4 for every $1 invested in such programs.2,3
Despite this, BBVIs continue to pose a risk to people who inject drugs.4 HCV remains one of the most common notifiable diseases in Australia,4 notified at 10 times the rate of HIV; more than 90% of all diagnoses of HCV infection are attributed to injecting drug use. Although the prevalence of HIV infection remains low among PWID in Australia, there is a continued threat of rapid escalation in this group, particularly for marginalised populations such as Aboriginal and Torres Strait Islander people.5 At the end of 2008, an estimated 284,000 people had been exposed to HCV in Australia, including 212,000 with chronic infection.6 In 2005, Aboriginal and Torres Strait Islander people were estimated to account for 22,000 or 8.3% of all Australian HCV diagnoses, with 16,000 chronically infected.7 Given that Aboriginal and Torres Strait Islander people constitute just 2.4% of Australia's total population, the disproportionate HCV-related burden of disease borne by this group is striking.8
Despite this clear health inequity, comprehensive examination of HCV seroprevalence and associated risk behaviour among Indigenous PWID is yet to be undertaken. Using data collected over 11 years (1998–2008) through the Australian NSP Survey (ANSPS),9,10 we compared patterns of risk behaviour and predictors of HCV exposure among Indigenous and non-Indigenous survey participants.
Since 1995, the ANSPS has estimated annual point prevalence in order to monitor changes over time in patterns of infection and risk behaviour among NSP clients.6,9,10 During 1 or 2 weeks each October, staff at selected NSP sites in all Australian jurisdictions recruit clients to complete a brief questionnaire covering demographics, injecting drug use and risk behaviour for bloodborne viral transmission, and to provide a capillary blood sample, which is subsequently screened for antibodies to HIV and HCV. The demographics section of the ANSPS questionnaire includes an item in which participants can identify as Aboriginal and/or Torres Strait Islander. Responses to this item were used to classify participants as Indigenous or non-Indigenous.
Analysis was restricted to the first participation record for each person. Respondents who reported participation in a previous survey were removed from our dataset (n=4,542). Additional previous participants were identified and removed by use of a unique identifier created by combining the first two letters of the first name, last name, birth month, birth year, Indigenous status and gender (n=1,261). The dataset was further refined by removal of people for whom data on key variables were missing, including participants for whom a unique identifier could not be created (n=2,701) and respondents who had not injected in the month before ANSPS participation and therefore did not answer questions on injecting risk behaviour (n=1,666). Of the 26,302 ANSPS records between 1998 and 2008, 16,132 individual cases were retained for analysis.
Data were collapsed into two groups: participants who identified as Aboriginal, Torres Strait Islander or both; and those who identified as neither Aboriginal nor Torres Strait Islander. Statistical analysis was performed using STATA Release 10.0. The Wilcoxon rank-sum test and Fisher's exact tests were used to analyse continuous and categorical variables, respectively. Univariate and multivariate logistic regression were used to model the associations between demographic, behavioural and drug use characteristics and HCV antibody serostatus among (i) the total sample; (ii) Indigenous participants only; and (iii) non-Indigenous participants only. Wald χ2 tests, odds ratios (ORs) and related 95% confidence intervals (CI) were calculated to assess statistical significance and precision.
Demographic and survey participation rates
Of the 16,132 individual cases, 1,380 participants identified as Aboriginal, Torres Strait Islander or both, representing 9% of our total sample; and 14,752 identified as neither Aboriginal nor Torres Strait Islander. Two-thirds of participants were male (Table 1), although a higher proportion of females reported Indigenous status (42% versus 33%; p<0.001). The proportion of ANSPS respondents for whom Indigenous status was not reported was low and stable at 2–3% over the period 1998–2008. During the period, the overall response rate for completion of the survey was 43% (36%-50%). The number of sites where surveys were collected has varied during the study period, however the majority of respondents (74%) were recruited from sites that participated in all years. The majority of both Indigenous and non-Indigenous respondents were surveyed in state/territory capital cities (78% and 80% respectively, p= 0.232)
Table 1. Demographic characteristics of people who inject drugs attending Australian Needle and Syringe Programs by Indigenous Status 1998–2008.
|Age at survey (yrs), median (IQR)|| ||30 (25–38)||29 (24–37)||0.003|
|Age <25 yrs old||4,045 (36)||386 (28)||3,659 (25)||0.009|
|Female||5,499 (34)||575 (42)||4,924 (33)||<0.001|
|Sexual orientation|| || || ||<0.001|
|Heterosexual||12,922 (80)||1,017 (74)||11,905 (81)|| |
|Bisexual||1,561 (10)||161 (12)||1,400 (9)|| |
|Homosexual||821 (5)||74 (5)||746 (5)|| |
|Not reported||829 (5)||128 (9)||701 (5)|| |
|Anti-HCV positive||8,293 (51)||791 (57)||7,502 (51)||<0.001|
|Anti-HIV positive||144 (1)||13 (1)||131 (1)||0.838|
|Age at first injection (yrs), median (IQR)||18 (16–22)||17 (15–21)||18 (16–22)||<0.001|
|Duration of injecting (IQR)|| ||10 (5–17)||10 (5–17)||0.046|
|<3 years||2,233 (14)||174 (13)||2,059 (14)||0.004|
|3 to 8 years||4,987 (31)||395 (29)||4,592 (31)|| |
|9 to 15 years||4,194 (26)||412 (30)||3,782 (26)|| |
|16+ years||4,718 (29)||399 (29)||4,319 (29)|| |
|Drug last injectede|| || || ||<0.001|
|Heroin||6,649 (41)||524 (38)||6,125 (42)|| |
|Amphetamine/methamphetamine||5,048 (31)||430 (31)||4,618 (31)|| |
|Cocaine/cocaine and heroin||658 (4)||88 (6)||570 (4)|| |
|Methadone/buprenorphine/subuxone||934 (6)||76 (6)||858 (6)|| |
|Other opiatesb||1,755 (11)||143 (10)||1,612 (11)|| |
|Not reported||1,088 (7)||119 (9)||969 (7)|| |
|Injecting frequency (last month)e|| || || || |
|Daily or more||8642 (54)||814(61)||7,828 (53)||<0.001|
|Injected in public (last month)e||7,909 (49)||748 (54)||7,161 (49)||<0.001|
|New needle syringe for all injections (last month)ce|
|Yes||11,272 (70)||1,006 (73)||10,266 (70)||<0.001|
|No||4,440 (27)||333 (24)||4,107 (28)|| |
|Not reported||420 (3)||41 (3)||379 (3)|| |
|Receptive needle syringe sharing (last month)e||2,632 (16)||293 (21)||2,339 (16)||<0.001|
|Receptive sharing ancillary equipment (last month)def||5,689 (35)||3,113 (38)||2,576 (33)||<0.001|
|Injected by anothere||2,134 (13)||222 (18)||1,912 (13)||<0.001|
|History of incarcerationg|
|Last year||2,667 (17)||367 (27)||2,300 (16)||<0.001|
|Ever||4,051 (25)||500 (36)||3,551 (24)||<0.001|
|Currently in opioid substance treatment program||5,311 (33)||441 (32)||4,870 (33)||0.425|
Indigenous participants were, on average, older than their non-Indigenous counterparts at the time of ANSPS completion (30 versus 29 years; p<0.001), although a higher proportion of Indigenous participants were under 25 years of age (28% vs 25%; p<0.009). A higher proportion of Indigenous participants reported having been incarcerated, both during their lifetime (36% vs 24%; p<0.001) and during the preceding 12 months (27% vs 16%; p<0.001). About one-third of each group reported current opioid substitution therapy (OST) (32% Indigenous vs 33% non-Indigenous).
Half (51%) of all participants tested positive to HCV antibody. Seroprevalence was significantly higher among Indigenous than non-Indigenous participants (57% vs 51%; p<0.001). Just 1% of each group tested positive to HIV antibody.
Drug use characteristics
Indigenous participants reported first injecting drugs at a younger age than non-Indigenous participants (17 vs 18 years; p<0.001). Although median duration of injecting (10 years) was similar for Indigenous and non-Indigenous participants, a higher proportion of Indigenous than non-Indigenous participants reported injecting daily or more frequently in the preceding month (61% vs 53%; p<0.001). Heroin was the most commonly reported drug last injected by both Indigenous (38%) and non-Indigenous participants (42%). Recent injecting risk behaviour was more commonly reported by Indigenous participants, including receptive needle syringe sharing (21% vs 16%; p<0.001); receptive sharing of ancilliary injecting equipment (38% vs 33%; p<0.001); having been injected by others (18% vs 13%; p<0.001); and injecting in public (54% vs 49%; p<0.001).
Risk factors for HCV infection
In the overall multivariable analyses, Indigenous status (OR 1.17; CI 1.03–1.32) and female gender (OR 1.25; CI 1.16–1.35) were independently associated with HCV antibody seropositivity (Table 2). Other independent predictors of anti-HCV seropositivity in the overall sample were daily or more frequent injection (OR 1.18; CI 1.10–1.27), reporting cocaine as the last drug injected (OR 1.12; CI 0.93–1.34), recent receptive needle syringe sharing (OR 1.14; CI 1.04–1.25) and receptive sharing of ancillary equipment (OR 1.23; CI 1.14–1.33), recent public injecting, (OR 1.24; CI 1.17–1.32), incarceration in the last 12 months (OR 2.14; CI 1.93–2.95) and current opioid substitution therapy (OR 1.70; CI 1.56–1.83).
Table 2. Multivariate analysis of risk factors for HCV seropositivity among people attending needle syringe programs in Australia by Indigenous status 1998–2008.
|Non-Indigenous||1|| ||–||–||–|| |
|Indigenous||1.17(1.03–1.32)||0.013||–|| ||–|| |
|Male||1|| ||1|| ||1|| |
|Female||1.25(1.16–1.35)||<0.001||1.38 (1.08–1.76).||0.01||1.26 (1.16–1.36)||<0.001|
|Heterosexual||1|| || ||–||1|| |
|Bisexual||0.96 (0.85–1.08)||0.030||–|| ||0.96 (0.85–1.09)||0.599|
|Homosexual||0.84 (0.72–0.98)||0.485||–|| ||0.84 (0.71–0.99)||0.048|
|Duration of injecting|
|<3 years||1|| ||1|| ||1|| |
|3–8 years||1.52 (1.35–1.71)||<0.001||1.46 (0.99–2.16)||0.059||1.52 (1.35–1.72)||<0.001|
|9–15 years||2.90 (2.57–3.26)||<0.001||2.98 (2.00–4.42)||<0.001||2.87 (2.54–3.26)||<0.001|
|16+ years||7.64 (6.76–8.63)||<0.001||6.66 (4.38–10.12)||<0.001||7.75 (6.82–8.80)||<0.001|
|Drug last injected|
|Heroin||1|| ||1|| ||1|| |
|Amphetamine/methamphetamine||0.44 (0.40–0.48)||<0.001||0.50 (0.38–0.66)||<0.001||0.43 (0.39–0.48)||<0.001|
|Cocaine/cocaine and heroin||1.12 (0.93–1.34)||0.227||0.96 (0.57–1.60)||0.864||1.17 (0.96–1.41)||0.170|
|Methadone/buprenorphine/subuxone||1.14 (1.07–1.34)||0.115||0.93 (0.53–1.63)||0.800||1.17 (0.99–1.39)||0.0935|
|Other opiatesa||0.76 (0.68–0.85)||<0.001||0.95 (0.63–1.43)||0.804||0.75 (0.66–0.84)||<0.001|
|Not reported||0.77 (0.67–0.88)||0.002||0.84 (0.54–1.30)||0.42||0.76 (0.65–0.88)||0.003|
|Injecting frequency (last month)b|
|Less than daily||1|| || || ||1|| |
|Daily||1.18 (1.10–1.27)||<0.001||–||–||1.18 (1.10–1.27)||<0.001|
|Public injecting last monthb|
|No?||1|| || || ||1|| |
|Yes?||1.24(1.17–1.32)||<0.001||1.08 (0.84–1.33)||0.494||1.08 (1.01–1.17)||<0.001|
|Receptive needle syringe sharing (last month)b|
|No||1|| ||–|| ||1|| |
|Yes||1.14 (1.04–1.25)||0.0082||–|| ||1.18 (1.07–1.31)||0.0014|
|Receptive sharing ancilliary equipmentabc|
|No||1|| || || ||1|| |
|Yes||1.23 (1.14–1.33)||<0.001||1.26 (1.01–1.61)||0.05||1.25 (1.15,1.35)||<0.001|
|Incarceration last yeard|
|No/Missing||1|| ||1|| ||1|| |
|Yes||2.14 (1.93–2.35)||<0.001||1.61 (1.23–2.13)||0.005||2.23 (2.01–2.48)||<0.001|
|Currently in opioid substance treatment programd|
|No/Not reported||1|| ||1|| ||1|| |
|Yes||1.70 (1.56–1.83)||<0.001||2.66 (1.26–2.17)||0.003||1.69 (1.55–1.83)||<0.001|
In comparison with participants who reported having injected for less than three years, those with longer injecting histories were more likely to test HCV antibody positive (3–8 years: OR 1.52, CI 1.35–1.71; 9–15 years: OR 2.90, CI 2.57–3.26; 16 years or longer: OR 7.64, CI 6.76–8.63). Compared to participants who reported last injecting heroin, those who reported last injecting methamphetamine were less likely to test positive to HCV antibody (OR 0.44; CI 0.40–0.48), as were participants who reported last injecting opioids other than methadone, buprenorphine, or subuxone.
In the multivariate model restricted to Indigenous participants (Table 2), factors independently associated with HCV antibody positivity included female gender (OR 1.38; CI 1.08–1.76); injecting drugs for more than three years; injecting in public in the past month (OR 1.08 CI 0.84–1.33); receptive sharing of ancillary injecting equipment in the last month (OR 1.26 CI 1.0–1.61); having been incarcerated in the preceding year (OR 1.61 CI 1.23–2.13) and current OST. Similar risk factors were identified in the multivariate model restricted to non-Indigenous participants (Table 2).
This study reveals higher seroprevalence of HCV antibody among Indigenous than non-Indigenous PWID, as well as more commonly reported risk factors among this group, including receptive sharing of needles and ancillary injecting equipment, having been injected by others, and injecting in public. Although similar factors predicted HCV antibody seropositivity among non-Indigenous participants, these factors were reported more commonly reported by Indigenous respondents. These risk behaviours should be addressed in future prevention campaigns targeting Indigenous PWID.
Furthermore, more Indigenous than non-Indigenous people accessing NSP were female and aged less than 25 years. Women have been identified as being at higher risk for acquiring HCV in this and other studies.11 The literature suggests that men often control the injecting behaviour of their female companions,12,13 and that women may inject drugs after male partners, using the same equipment, or be injected by male partners, thereby increasing their risk for acquiring HCV.12,13 Young Indigenous people have also been identified as vulnerable to risky injecting drug use due to a range of factors including marginalisation, shame and disempowerment14 demonstrating a specific need for HCV prevention interventions designed for Indigenous women and young people.
Indigenous participants were more likely than non-Indigenous participants to report recent incarceration. Several previous studies, along with the current one, have identified incarceration as an independent predictor of anti-HCV prevalence.4,14 The National Prison Entrants Blood Borne Virus Surveys collected in 2004 and 2007 in selected Australian States and Territories and described elsewhere15 comprised a consecutive cross-section of prison entrants recruited over 2 two-week periods. The two surveys merged and compared with this study show higher rates of Indigenous participants (17% vs 9%, n= 228) a younger median age (30 vs 33) and a higher proportion of males (84% vs 66%). Both ANSPS and NPEBBVS participants were tested for HCV antibody. In the ANSPS both non-Indigenous (51%) and Indigenous (57%) participants had a slightly lower rate of anti HCV positivity 51% and 57% respectively compared to participants in the NPEBBVS who identified as a person who injects drugs. Rates of anti HCV positive for Indigenous and non-Indigenous participants in the NPEBBVS were 58% and 61% respectively.
Given that an estimated 1% of the Australian population are estimated to be anti HCV positive this environment poses significant risk for acquisition of HCV for people who are incarcerated and exposed to blood through any known risk factors for HCV acquisition. This is particularly important for Indigenous people, given the over-representation of Indigenous people in Australian prisons. Currently, 26% of all Australian prisoners are Aboriginal and Torres Strait Islander, rates of incarceration are increasing, and rates of re-entering the Australian prison system are significantly higher for Indigenous inmates.16 Clearly, this environment poses significant risks for Indigenous communities, because of these factors combined. Furthermore no Australian prisons currently operate NSPs or safe tattooing spaces.17 This, in association with the fact that prisoners may be exposed to injecting drug use for the first time while incarcerated and the over-representation of Indigenous people in Australian prisons, has implications for the broader Indigenous community as, upon release, there may be an increased risk of further transmission of HCV.14 Prison settings are a unique opportunity to provide HCV education and prevention programs to high-risk individuals.14,16,18 Prison NSPs operating in six European jurisdictions have had favourable results, with stable or decreased drug use and decreased reports of drug initiation, decreased equipment sharing and no reported cases of HCV, hepatitis B or HIV transmission.19
Whereas HCV seroprevalence was higher among Indigenous participants, HIV seroprevalence rates were around 1% for both groups. Given this study has demonstrated higher risk behaviour associated with HCV seroprevalence, there is an urgent need for a heightened awareness among Indigenous PWID highlighting the risks of HIV transmission. Failure to do this could result in a generalised epidemic similar to Canada's First Nations people,20 and that to date has been averted among Indigenous people in Australia.
Our study suggests that for over a decade, Indigenous people comprised around 10% of all participants attending NSPs, a significant over-representation given that estimates suggest this group currently comprises 2.4% of Australia's total population. In the absence of national representative data in relation to Indigenous PWID, this study provides a proxy of the prevalence of injecting drug use within this population. The great strength of this study is its large sample size and the heterogeneity of the sample, which was recruited from multiple NSPs in a range of geographical areas in all states and territories of Australia. However, given that the majority of participants were recruited from major cities, it is not clear to what extent our results are generalisable to PWID living in regional and remote areas, both Indigenous and non-Indigenous. On the basis of these findings, we have no evidence that Aboriginality in the survey is confounded by remoteness.
The conclusiveness of our study's findings is nevertheless limited by a range of factors. The sample is limited to PWID who access NSPs; and the ANSPS achieves a response rate of less than 50% in most years. Recent work has demonstrated, however, that the overall profile of participants is typical of Australian NSP clients.21 It is possible that participants under-report risk behaviours, particularly those that are illegal or socially sensitive,22,23 although others have demonstrated the reliability of self-report data from PWID,24 and we demonstrated that self-completion reduces social desirability bias in reports of risk behaviours among ANSPS participants.25,26 Lastly, the study is vulnerable to the limitations inherent in all cross-sectional study designs, including the inability to disentangle the cause-effect relationship between risk factors and HCV antibody prevalence.
To conclude, HCV seroprevalence is higher among Indigenous than non-Indigenous PWID attending NSPs in Australia. This may be due partly to higher proportions of women and young people among Indigenous PWID; and because Indigenous PWID are more likely than their non-Indigenous counterparts to have been incarcerated and to engage in risky behaviour, such as sharing syringes and other equipment and being injected by another person. Clearly, specific, culturally appropriate HCV prevention programs are needed to reduce the burden of HCV infection among Indigenous people in Australia.
This work was undertaken on behalf of the Collaboration of Australian Needle Syringe Programs (NSPs)(2009): ACON Hunter; Albury CHC; Barwon Health Service; Biala NSP; Cairns NSP; Central Access Service; Central Coast Harm Reduction Services; Clarence CHC; DASSA; Directions ACT; First Step Program; Health ConneXions; Health Information Exchange; Hindmarsh Centre; Hunter Harm Reduction Services; Inner Space; Kirketon Road Centre; Kobi House; Northcote NSP; North Coast Harm Reduction Services; North Richmond CHC; Northern Territory AIDS Council; NUAA; Nunkuwarrin Yunti CHC; QUIHN NSP Services; REPIDU; Salvation Army Launceston; SAVIVE CNP Services; SHARPS; Sydney West NSP Services; TasCHARD NSP Services; Townsville ATODS; WA AIDS Council; WASUA and West Moreton NSP.
We are grateful to the clients and staff of participating Needle and Syringe Programs and to the Australian Needle and Syringe Program Survey (ANSPS) National Advisory Group. The ANSPS is funded by the Australian Government Department of Health and Ageing. The views expressed in this publication do not necessarily represent the position of the Australian Government. John Kaldor and Lisa Maher are supported by NHMRC Senior Research Fellowships. The Kirby Institute (formerly the National Centre in HIV Epidemiology and Clinical Research) is affiliated with the Faculty of Medicine, University of New South Wales.