Hepatitis C virus antibody prevalence among injecting drug users at selected needle and syringe programs in Australia, 1995-1997

Australia adopted a harm reduction approach as part of its national drug strategy in 1985.⁴ An extensive network of needle and syringe programs (NSPs) has been established in Australia; in the financial year 1994-95, around 700 NSPs distributed six million syringes nationally and an additional four million were distributed through pharmacies.⁵ Early and vigorous implementation of harm reduction measures, such as methadone maintenance, peer-based education and NSPs, has successfully maintained low seroprevalence of HIV infection among people who inject drugs in Australia.⁶ In contrast, prevalence and incidence of HCV infection among this population remain high.^2,7

To assess the impact of prevention activities on HIV and HCV infection, a monitoring system was established at selected NSPs throughout Australia.⁶ Cross-sectional surveys at NSPs among people who inject drugs offer a practical and repeatable mechanism for recruiting large samples of injecting drug users. This provides a basis for systematic monitoring of HCV and HIV infection, and of injecting behaviours associated with transmission of these viruses. We report the results of the first three years of national HCV monitoring among NSP attenders in Australia.

Sex and age group were recorded for all clients who attended participating sites during the survey week. As some clients attended more than once, a record was kept as to whether the attendance was the first for that week to measure survey response rate. Participants completed a brief questionnaire on basic demographic characteristics and injecting and sexual behaviour.

Capillary blood was collected on blotting paper by finger prick with single-use lancets. Venous blood was not obtained, as few NSPs had appropriate facilities for collecting and storing large amounts of blood. Specimens were tested for HCV antibody by a modified third-generation enzyme immunoassay (Abbott HCV 3.0, Chicago, USA). A modified cutoff value for optical density was calculated to capture more than 95% of the seronegative population. Specimens were considered positive for HCV antibody if the optical density to cutoff ratio was greater than or equal to one on initial and subsequent testing. In 1998, as a quality control measure, 60 samples were randomly selected from each year's survey and retested for HCV antibody as described. Retest results were 100% concordant for the 1995 and 1997 surveys but only 88% for 1996. Subsequently, all available stored 1996 samples (83%) were retested.

Ethical approval for this study was obtained from relevant ethics committees associated with the investigators and with participating sites.

Associations between HCV prevalence and participants' characteristics were evaluated by the χ² test and χ² test for linear trend. To control for intralaboratory variation with HCV antibody tests, we calculated HCV prevalence for 1996 participants using retested results where available. In addition, for participants with insufficient stored sera for retesting, the original estimates of HCV prevalence were adjusted for each category of variables used in the analysis, by the overall difference between original and retest test results. Associations between HCV prevalence and participants' characteristics in 1997 were also assessed using multivariate logistic regression. Factors significantly associated with HCV infection on univariate analysis (with an inclusion criterion of P ≤ 0.05) or factors that were considered a priori to be predictors of HCV risk were included in the logistic regression model.

The age ranges for respondents in the three years were 14-54 years in 1995, 13-53 years in 1996 and 15-58 years in 1997. The median age of respondents was significantly higher in 1995 than in 1996 (29 years v. 28 years; P = 0.009) and 1997 (29 years v. 27 years; P < 0.001). Median age at commencement of injecting drug use was 18 years in the three surveys. Consequently, the median number of years that respondents had injected drugs was significantly higher in 1995 than in 1996 (10 years v. 8 years; P = 0.001) and 1997 (10 years v. 7 years; P < 0.001).

Similar proportions of males (65%, 67% and 66%), respondents reporting being heterosexual (79%, 78% and 79%), and those reporting having been imprisoned in the past year (15%, 13% and 14%) participated in the 1995, 1996 and 1997 surveys. Almost a third of respondents in each year reported no contact with health services such as counselling, detoxification or methadone maintenance treatment. The proportion of respondents on methadone treatment, however, was significantly lower in 1997 (33%) than in 1995 (40%; P = 0.001) or 1996 (36%; P = 0.02).

More respondents reported that heroin was the last drug they injected in 1996 and 1997 than in 1995 (53% and 56% v. 44%) and fewer reported last injecting amphetamines (19% and 18% v. 21%) and methadone (12% and 11% v. 19%; P < 0.001). Daily or more frequent injection increased from 41% of respondents in 1995 and 1996 to 51% in 1997 (P = <0.001).

HCV antibody prevalence was also significantly lower in 1996 and 1997 (13%) than in 1995 (22%) among respondents who reported less than three years of drug injection (P = 0.03). This difference also remained significant after adjustment for sex, age, last drug injected, frequency of drug injection and health service contact (adjusted odds ratio, 0.4; 95% CI, 0.2-0.7). HCV antibody prevalence was lower in 1996 and 1997 than in 1995 among respondents aged less than 25 years, regardless of whether the last drug injected was heroin (trend test, P = 0.03) or amphetamine (trend test, P = 0.002; Box 3a).

Box 3b shows that, when respondents were grouped according to the year they started injecting, HCV antibody prevalence was higher among those reporting most recent injection of methadone or heroin than among those reporting most recent injection of amphetamine. HCV antibody prevalence was also significantly higher among respondents from New South Wales and Victoria than those from Queensland, and remained higher when the analysis was restricted to respondents reporting heroin as the last drug injected and stratified according to the year drug injection started (Box 3c). Multivariate logistic regression analysis showed that other factors significantly associated with presence of HCV antibody in 1997 included being female, having been imprisoned in the past year, having a history of methadone treatment, being aged 25 years or more, having injected drugs for more than five years, and daily or more frequent injection (see Box 2).

Variation in the populations surveyed in the three years of the study cannot be excluded as an explanation for our observations. Respondents in 1997 were younger and newer to injecting than in 1995. Nonetheless, the decline in HCV antibody prevalence remained statistically significant when multivariate logistic regression analysis was used to control for differences in demographic characteristics, when analysis was restricted to respondents who reported less than three years of drug injection, and when analysis was restricted to sites that participated in all three surveys.

It is also conceivable that HCV infection status influenced participation in the survey. For example, people infected with HCV might be more likely to participate than those not infected because of the services provided by NSPs or because they have an interest in blood-borne infections by virtue of having one. Alternatively, people with HCV infection might be reluctant to provide a blood sample in a non-clinical setting because of concerns about inadvertent spread of infection. It is not possible to determine whether people with HCV infection were more or less likely than those without infection to participate in the surveys. If such bias occurred it is unlikely that the direction changed from 1995 to 1996 and 1997. However, the extent to which the magnitude of such bias may have changed over the three surveys is not known.

Comparison of our data with those of other Australian studies reporting HCV antibody prevalence according to type of drug injected and duration of injecting supports our observation that HCV antibody prevalence is declining among people who inject drugs. Among opiate injectors who had been injecting for less than three years, HCV antibody prevalence was 70% in the late 1980s⁹ and almost 50% in the early 1990s,¹⁰ compared with 20% in our study. Declining HCV antibody prevalence has also been reported from other cities that implemented HIV prevention measures in the mid-1980s, namely Geneva¹¹ and Glasgow.¹²

We also found that the proportion of respondents who reported using a syringe after someone else was significantly lower in 1997 than in 1995 and 1996. There has been a marked decline in reported sharing of syringes in Australia since 1984, when more than 90% of respondents reported having done so in the month before interview.¹³ Of concern was the higher rate of sharing reported among respondents aged less than 25 years than among older respondents in 1997. Younger injecting drug users are probably more recent initiates to injecting; this is a subpopulation previously identified as being at extremely high risk of acquiring HCV infection.¹⁴ An extremely high incidence of HCV infection has also been reported recently among young people with a history of drug injecting in Sydney.⁷

It is difficult to explain the marked geographic variation in HCV antibody prevalence detected in our study. The sample recruited from Queensland was significantly different from other States and Territories with regard to known correlates of HCV infection such as duration of injecting and type of drug injected. Nonetheless, significantly lower prevalence persisted among respondents from Queensland compared with those from New South Wales and Victoria when the sample was stratified according to these factors.

Despite our finding of declining HCV antibody prevalence among people who inject drugs in Australia, and even though an epidemic of HIV infection has so far been prevented, the prevalence and incidence of HCV infection in this group remain high.^2,7 Percutaneous transmission is more efficient for HCV than for HIV infection.¹ More importantly, the carriage rate of HCV among injecting drug users was already considerably higher than that for HIV when harm reduction policies were first introduced.⁹ It is likely that occasional instances of shared injection equipment and other blood contact during injection have been sufficient to maintain high levels of HCV transmission without an increase in HIV transmission.

It can not be assumed that the results of our surveys are generalisable to all people who inject drugs in Australia. However, NSP clients represent a heterogeneous population of injecting drug users who are readily accessible for targeted prevention initiatives. Prevention efforts encompassing education, drug treatment and needle exchange need to be enhanced to improve consistency and coverage so that transmission of HCV infection is reduced further and the current low prevalence of HIV infection is sustained.

1. MacDonald M, Crofts N, Kaldor J. Transmission of hepatitis C virus: rates, routes and cofactors. Epidemiol Rev 1996; 18: 137-148.
2. Crofts N, Jolley D, Kaldor J, et al. Epidemiology of hepatitis C virus infection among injecting drug users in Australia. J Epidemiol Community Health 1997; 51: 692-697.
3. Seeff LB. Natural history of hepatitis C. Hepatol 1997; 3 Suppl: 21S-28S.
4. Blewett N. National Campaign against drug abuse: Assumptions, arguments, and aspiration. Canberra: AGPS, 1987. (NCADA Monograph No. 1.)
5. Wodak A, Lurie P. A tale of two countries: Attempts to control HIV among injecting drug users in Australia and the United States. J Drug Issues 1996; 27: 117-134.
6. MacDonald M, Wodak A, Ali R, et al, on behalf of the Collaboration of Australian Needle Exchanges. HIV prevalence and risk behaviour in needle exchange attenders: a national study. Med J Aust 1997; 166: 237-240.
7. van Beek I, Dwyer R, Dore GJ, et al. Infection with HIV and hepatitis C among injecting drug users in a prevention setting: retrospective cohort study. BMJ 1998; 317: 433-437.
8. Stata Statistics/Data Analysis [computer program]. Version 5.0. Texas: Stata Corporation, 1997.
9. Bell J, Batey RG, Farrell GC, et al. Hepatitis C virus in intravenous drug users. Med J Aust 1990; 153: 217-273.
10. Crofts N, Hopper JL, Milner R, et al. Blood-borne virus infections among Australian injecting drug users: Implications for spread of HIV. European J Epidemiol 1994; 10: 687-694.
11. Broers B, Junet C, Bourquin M, et al. Prevalence and incidence rate of HIV, hepatitis B and C among drug users on methadone maintenance treatment in Geneva between 1988 and 1995. AIDS 1998; 12: 2509-2066.
12. Goldberg D, Cameron S, McMenamin J. Hepatitis C antibody prevalence among injecting drug users in Glasgow has fallen but remains high. Commun Dis Public Health 1998; 1: 95-97.
13. Crofts N, Webb-Pullman J, Dolan K. An analysis of trends over time in social and behavioural factors related to the transmission of HIV among injecting drug users and prison inmates. Evaluation of the National HIV/AIDS Strategy 1993-94 to 1995-96. Technical Appendix 4. Canberra: AGPS, 1996.
14. Garfein RS, Vlahov D, Galai N, et al. Viral infections in short-term injection drug users: the prevalence of hepatitis C, hepatitis B, human immunodeficiency, and human t-lymphotropic viruses. Am J Public Health 1996; 86: 655-661.

Alcohol and Drug Services, St Vincent's Hospital, Sydney, NSW.
Alex D Wodak, FRACP, FAFPHM, Director.

National Drug and Alcohol Research Centre, University of New South Wales, Sydney, NSW.
Kate A Dolan, PhD, Senior Lecturer.

Kirketon Road Centre, Sydney, NSW.
Ingrid van Beek, MBA, FAFPHM, Director.

Centre for Immunology, St Vincent's Hospital, Sydney, NSW.
Philip H Cunningham, BAppSc, Senior Hospital Scientist.

Reprints: Ms M A MacDonald, National Centre in HIV Epidemiology and Clinical Research, Level 2, 376 Victoria Street, Darlinghurst, NSW 2010.
mmacdATnchecr.unsw.edu.au

Make a comment