Epidemiology of Hepatitis C Virus Infection Among Injecting Drug Users in Australia
Total Page:16
File Type:pdf, Size:1020Kb
69262ournal ofEpidemiology and Community Health 1997;51:692-697 J Epidemiol Community Health: first published as 10.1136/jech.51.6.692 on 1 December 1997. Downloaded from Epidemiology of hepatitis C virus infection among injecting drug users in Australia Nick Crofts, Damien Jolley, John Kaldor, Ingrid van Beek, Alex Wodak Abstract is the case also with hepatitis B virus (HBV) Study objective-To review the epidemiol- and human immunodeficiency virus (HIV). ogy of hepatitis C virus (HCV) infection Other exposures that have been found to be among injecting drug users (IDUs) in important are the receipt of contaminated Australia, and consider needs for further blood or blood products,4 tattooing,5 and den- research and prevention policies and pro- tal procedures.6 Since the introduction of grammes. universal donor screening for HCV antibody in Design-(1) Review of the results of 1990, transmission through the blood supply surveillance for HCV; (2) review of pub- has become rare.7 lished literature on prevalence, incidence, It is currently estimated that over 80% of and risk factors for HCV among IDUs; those exposed to HCV will become carriers,8 at and (3) reconstruction of incidence rates risk of long term disease including cirrhosis from prevalence studies of HCV in IDUs. and hepatocellular carcinoma.9 While actual Setting and Participants-Field and clinic rates are still unclear, it is possible that 20% or based studies of IDUs in Australia. more of those chronically infected will develop Main results-HCV has been present at such disease over 20 years or more.'0 The only high prevalences (of the order of 60-70%) available treatment, alfa interferon, is ineffec- in populations ofAustralian IDUs since at tive in most patients and is specifically not least 1971. Duration of injecting and main available to current injecting drug users drug injected were the main predictors of (IDUs) in Australia." Prospects of a vaccine seropositivity, the latter possibly a surro- against HCV are not promising in the gate for frequency of injecting and both short-term. 12 together as surrogate for cumulative There have been a number of surveys of dif- numbers of times injected. Risk of infec- ferent populations of IDUs in Australia for tion begins with first injection and contin- HCV exposure. Surveillance for diagnosed ues as long as injecting does. Current HCV infection at a national level was intro- The Macfarlane incidence is approximately 15 per 100 per- duced and has been gradually refined since Burnet Centre for son years, and up to 40 per 100 person testing began in 1990. To establish as complete Medical Research, a as of the pattern of HCV Fairfield, Victoria, years in some subpopulations. Incidence picture possible Australia may have decreased through the 1980s as a infection among IDUs in Australia, we under- N Crofts result of behaviour change in relation to took a review of available epidemiological data. D Jolley HIV, as it has for hepatitis B, but not We have also sought to identify research needs http://jech.bmj.com/ significantly so. and opportunities, and effectiveness of and National Centre in HIV Conclusions-Control ofHCV infection in requirements for further prevention pro- Epidemiology and of grammes to decrease its spread in these popu- Clinical Research, Australia will depend on effectiveness Darlinghurst, measures to control HCV spread among lations. New South Wales, IDUs. This will be a greater challenge than Australia the control of HIV in this population has J Kaldor been. Needs identified include improved Methods surveillance, especially for recently ac- NATIONAL SURVEILLANCE on October 1, 2021 by guest. Protected copyright. Kirketon Road Centre, of There are two national surveillance systems for Darlinghurst, quired infection, better understanding exact transmission modes, and urgent communicable diseases that collect informa- New South Wales, tion on the occurrence of HCV infection: the Australia improvement in prevention strategies. I van Beek National Notifiable Diseases Surveillance Sys- (7 Epidemiol Community Health 1997;51:692-697) tem (NNDSS), which receives notification Alcohol and Drug data from State and Territory Health Depart- Services, St Vincent's ments; and the Communicable Diseases Intel- Hospital, Since the identification of hepatitis C virus ligence Laboratory Reporting Scheme (Lab- Darlinghurst, (HCV) in 1989' and the availability of assays VISE), which receives reports of laboratory New South Wales, laboratories. Australia for the detection of antibody to HCV in 1990,2 diagnoses from sentinel Reports A Wodak the epidemiology of HCV infection has been of HCV infections were available for the years investigated in many populations. The major 1991 to 1994 for the first system, and for the Correspondence to: group infected and at risk of continuing infec- period 1990 to September 1995 for the second. Dr N Crofts, The as in other The NNDSS collects information on age, sex, Macfarlane Burnet Centre tion with HCV in Australia, for Medical Research, PO developed countries, is people who currently or and postcode of residence of cases, but not on Box 254, Fairfield, Victoria, previously injected or had been injected with possible route of exposure; since 1994, cases Australia 3078. illicit drugs.3 This increased risk is associated have been classified as incident or otherwise. data on Accepted for publication with the practice of sharing of equipment used The LabVISE system collects age, 5 December 1996 in injecting; especially needles and syringes, as sex, postcode, risk factor (a combination of Hepatitis C among Australian IDUs 693 J Epidemiol Community Health: first published as 10.1136/jech.51.6.692 on 1 December 1997. Downloaded from Table 1 Contributions to the overall log likelihood years in each year. Incident cases were so clas- sified in 1993-4, making up 0.4% of notifica- Status Observed Log-likelihood terms tions (73 of 16 424). Always negative u only - k (u - s) There were data available from the LabVISE Entered positive v only log [1 - exp j- A (v - s) ] system for six years (1990 to 1995), but only to Seroconverted both u & v - k (u - s) + log [ 1 - exp - X (v - u)t] September in the last year. The total number of s= date of starting to inject; u = date of last negative test; v = date of first positive test. reports for the period was 18 124, with annual totals rising from 115 in 1990 to 6118 in 1994. exposure factors and personal characteristics), Details on risk factors were available only for outcome, and method of diagnosis. the years 1992 to 1995, but over these years only 11.8% of cases had such information. Of PREVALENCE AND INCIDENCE STUDIES the 2137 cases with such data, 1847 (86.4%) Published reports on prevalence, incidence, were classified as having a history of IDU, this and predictors of HCV seropositivity among proportion remaining constant over the years. IDUs in Australia were also reviewed. Studies based on self reported HCV status were not PROPORTION OF AUSTRALIAN IDUS TESTED FOR included, because of evidence of uncertainty ANTI-HCV about reliability and validity of such data,'3 but Most of the reported clinic based studies of those using self report to assess whether seroprevalence did not report the participation subjects had had a test for HCV were included. rate of clients in anti-HCV testing. One Published reports were identified by MedLine retrospective study on the stored serum search and contact with researchers known to samples of methadone clinic attenders re- be working in this field. ported testing samples from 172 IDUs of a total of 783 people (22%), based apparently on RECONSTRUCTED INCIDENCE the availability of stored serum samples."' Of To assess the impact of programmes intro- 234 young drug users, which included an duced from 1987 in Australia to decrease the unstated proportion with no history of drug spread of HIV among IDUs on the spread of injecting in Perth in 1993, 57 (24%) reported HCV in the same populations, we modelled the having had a test for HCV; 29% (51 of 176) of incidence of HCV over the 1980s, using preva- those who reported a history of injecting drug lence data. Data from two studies were used. use (mean duration 1.9 years) had been The first was a longitudinal prospective study tested."' A similar study of 275 young IDUs of field recruited IDUs in Victoria, the full (mean duration of injecting 2.5 years) in Mel- method for which has been published bourne in 1994 found 59% reported having elsewhere. 4 The second was of IDUs attending had an HCV test.20 Of 872 respondents of all a primary care clinic in inner Sydney, New ages interviewed in four cities in 1994, 79% South Wales, the method for which has also reported having been tested for hepatitis C been published elsewhere.'5 We estimated a antibodies at least once.2' seroconversion rate as the subject's risk of test- ing positive (even at first test) as a function of HCV PREVALENCE IN AUSTRALIAN IDU the number of years of exposure since first POPULATIONS (TABLE 2) injecting, assuming zero risk for HCV infection A variety of IDU populations has been surveyed for prevalence of HCV antibody before starting to inject. The full method is http://jech.bmj.com/ presented in table 1.16 including those from whom serum samples Ninety five per cent confidence intervals for have been collected for other purposes, IDUs incidence rates were calculated using an expo- in hospital, IDUs attending clinics (metha- nential error factor for incidence rates." done, primary care, and STD), and IDUs par- ticipating in field studies. A variety of antibody tests with different sensitivities and specificities Results were used in the different studies.22 Overall, the The earliest evidence of HCV infection among prevalence among male IDUs ranged from on October 1, 2021 by guest.