Secondary Logo

Journal Logo


Estimating the cost-effectiveness of needle-syringe programs in Australia

Kwon, Jisoo A.; Anderson, Jonathan; Kerr, Cliff C.; Thein, Hla-Hla; Zhang, Lei; Iversen, Jenny; Dore, Gregory J.; Kaldor, John M.; Law, Matthew G.; Maher, Lisa; Wilson, David P.

Author Information
doi: 10.1097/QAD.0b013e3283578b5d



In many parts of the world, the sharing of injecting equipment, including needle-syringes, among injecting drug users (IDUs) is a major mode of transmission of blood-borne viruses such as HIV and hepatitis C virus (HCV) [1,2]. To reduce the spread of HIV and HCV infections through needle-syringe sharing, Australia introduced needle-syringe programs (NSPs) in 1986 [3]. HIV prevalence among IDUs in Australia has since remained relatively low and stable at approximately 1% [4]; however, HCV prevalence has remained relatively high at 50–70% [4]. At present, there are more than 3000 NSP sites across Australia, comprising primary and secondary NSP outlets, mobile and outreach services, syringe vending machines, and pharmacies that sell or provide injecting equipment free of charge.

Globally, there are significant differences in HIV epidemics among IDUs. Ecological studies suggest that where NSPs are not easily accessible, HIV prevalence tends to be substantially higher than in settings where NSPs are available and easily accessible. For example, Hurley et al.[5] identified studies which estimated HIV seroprevalence among IDUs in the United States between 1988 and 1993 and estimated an average annual increase in HIV prevalence of 5.9% in cities without NSPs, whereas seroprevalence decreased by an average of 5.8% per year in cities with NSPs. HIV acquisition among IDUs has been shown to be significantly associated with nonuse of NSPs [6,7].

In this study, we aimed to, first, estimate the likely impact of NSPs in preventing HIV and HCV transmission at the population level in Australia over the period 2000–2010, and second, estimate the cost utility of NSPs. To estimate the effect of an intervention, one could compare epidemiological indicators in the population of interest with a relevant control population. Another approach to impact evaluation is to compare observed conditions with a counterfactual scenario described by an epidemiological model. The counterfactual is intended to represent what the epidemics would look like in the absence of the intervention. The epidemiological model is used to generate an estimate for the counterfactual in lieu of a ‘real’ comparison case. Here, we examined available empirical data to inform assumptions about the counterfactual with respect to the dominant behavioral factor of HIV and HCV transmission, namely, extent of receptive syringe sharing (RSS). We developed an epidemiological model, which we then calibrated to reflect the historical HIV and HCV epidemics among Australian IDUs. This model was then compared with the epidemics that would have been expected according to the counterfactual conditions.


Relationship between needle-syringe distribution and receptive syringe sharing

NSPs aim to reduce the rate of RSS. We examined the relationship between NSP distribution and reported levels of RSS with others in the past month. The numbers of syringes distributed across Australia through NSPs were collected from each Australian state and territory health department for the period 1993–2008 [8]. Syringe distribution in Australia has increased over time (Fig. 1a, circles). Currently, approximately 200 syringes are distributed per IDU per year on average, based on an estimated IDU population size of 170 000 [8]. The frequency of RSS among IDUs has been estimated annually since 1995 through the Australian Needle and Syringe Program Survey (ANSPS), a serial cross-sectional seroprevalence survey of IDUs attending NSPs throughout Australia [4,9]. Over the period 1995–1999, reported prevalence of RSS decreased from 31 to 17% but has since remained steady [4]. During 2000–2010, levels of pharmacy purchasing of needle syringes (Fig. 1a, rectangles) have been relatively constant despite large increases in the distribution of syringes from NSPs (Fig. 1a, triangles); therefore, we assume that increases or decreases in NSP distribution are not offset by changes in pharmacy purchasing.

Fig. 1:
Syringe distribution trends and relationship between needle-syringe distributed per capita and receptive needle-syringe sharing rate.(a) Syringe distribution trends in Australia from needle-syringe programs (NSPs) and pharmacies. (b) Percentage of people who receptively shared needle-syringes in the last month versus annual per capita needle-syringes distributed through NSPs; the red bar on the y-axis indicates the assumed range of sharing rates in the absence of NSPs (20–50%); blue diamonds indicate regional estimates of sharing rate in Sydney [10] (27%) and Perth [11] (32%) in 1993, and in Melbourne [12] (37%) and Perth [13] (27%) in 1995; black circles indicate national estimates of sharing rates from 1995–2010.

We also collated data on RSS among Australian IDUs from other surveys conducted during 1970–1995 which are comparable to data from the ANSPS. Four studies were identified which also reported on proportions of surveyed IDU populations who receptively shared in the last month: 27% in Sydney [10] and 32% in Perth [11] in year 1993, 37% in Melbourne [12] during 1990–1995 and 27% in Perth [13] in 1995, corresponding with an estimated NSP distribution of 39 and 48 syringes per IDU per year in year 1993 and 1995 [14]. In Fig. 1b we presented the relationship between RSS and NSP distribution of needle syringes for years in which both RSS prevalence and syringe distribution data are available [10–13]. In comparison to data presented in Fig. 1b, it has been estimated that RSS levels in the 1970s and 1980s, with no or limited NSP coverage, could have been 70–90% among groups of IDUs [15–17]. However, considering the impact that increased awareness of blood-borne infections would likely have had on-risk behaviors, we assumed that RSS prevalence would be 25–50% in the absence of NSPs during 2000–2010 (Fig. 1b). For the counterfactual scenario we assumed that parameters other than RSS would remain unchanged.


A mathematical model was developed to describe HIV and HCV transmission, the distribution of the IDU population across health states over time and the associated healthcare costs under actual conditions and according to the counterfactual scenario. Full details of the model are provided in the Appendix, Briefly, the model consists of a number of linked subpopulations (‘compartments’), categorized by infection, diagnosis, disease progression (in terms of CD4 categories for HIV and stages of fibrosis and clinical states for HCV) and treatment states, as shown in Fig. 2. IDUs who are successfully treated for HCV become susceptible again, and can thus be reinfected. The rates of transition between compartments are defined by parameters based on available data. These can be divided into biological parameters (including HIV and HCV transmission probabilities, HIV-related and HCV-related death rates, transition probabilities and treatment effectiveness) and behavioral parameters (including the frequency of injections, the fraction of IDUs who engage in RSS and the fraction of syringes that are cleaned prior to sharing). The model estimates the change in the number of IDUs in each compartment due to disease progression, initiation of treatment, death or aging and incidence of infection. The model also calculates the expected numbers of HIV and HCV transmissions through probabilities associated with risk behaviors. The model incorporates time-dependent parameters for factors in which comparable trend data are available, including the frequency of injections, RSS, testing rates and the number of HIV-infected/HCV-infected current and former IDUs initiating HIV or HCV treatments. Full descriptions of all model parameters and their defined range of values and justifications, are provided in the Appendix,

Fig. 2:
Schematic diagram of HIV and hepatitis C virus disease progression among injecting drug users in Australia.Each arrow represents the rate of change in the number of people in the population moving from one health state compartment to another health state compartment.

Economic analysis

Economic analysis was carried out to estimate the cost-effectiveness of NSPs from a health sector perspective. NSP costs were collated from all state and territory health departments in Australia in 2008 for 2000–2008; 2009–2010 costs were assumed to be the same as the 2008 costs. Costs included consumable items (syringe distribution, syringe disposal, and safe sex packs) and support costs (primary NSP operations, support for secondary NSPs, transport and vending machines). We also derived estimates of average annual costs for different HIV and HCV health states using published literature and local data (estimated from activity-based analysis and national databases) [8]. The annual healthcare cost for a person living with HIV varied from A$1523–2731 (excluding antiretroviral drugs), depending on the health state. In addition, the annual cost of antiretroviral therapy (ART) was estimated to be A$14 613 for first-line therapy, A$15 178 for second-line and A$27 776 for subsequent lines of ART per person per year. The annual healthcare cost for a person living with HCV was estimated to range A$288–114 411, from early HCV infection to liver transplantation, respectively. The unit costs of hospital admission were estimated by searching health department data on the frequency and proportions of admission to hospital with different health states of HCV and HIV [18] and then deriving a weighted average cost per admission in a health state using cost weights for admission to an Australian public hospital [19]. Outpatient items were valued from the Australian Medicare Benefits Schedule [20] and Pharmaceutical Benefits Scheme [21]. Client costs for the purchase of injection equipment were estimated from data on the number of sterile injection equipment packs provided through pharmacies and average client out-of-pocket cost for these packs. However, indirect healthcare costs were not included in the analyses. All costs were estimated in 2008 Australian dollars and inflated to 2010 Australian dollars using the health consumer price index [22]. The annual healthcare costs per person are summarized in the Appendix,

Cost-utility analysis was performed to compare the costs and outcomes between the counterfactual no-NSP scenario and the status quo. Health state utilities were used to adjust the absolute life expectancy to derive quality-adjusted life years (QALYs) [23,24]. Healthcare costs were also calculated based on the number of people in each health state. The incremental cost-effectiveness ratio (ICER), expressed as cost per QALY gained, is the difference in costs between the counterfactual no-NSP scenario (HIV and HCV healthcare costs) and the status quo (NSP costs along with HIV and HCV healthcare costs) divided by the difference in QALYs between the same two scenarios. Analyses were performed for two time frames: 2000–2010 to study the return obtained over the period of investment, and 2010–2100 to examine the lifetime costs and consequences for the population related to the investment over the period 2000–2010. Costs were adjusted by an annual rate of 3% for the past 10 years and then discounted over future time by three choices of rate: 0, 3 and 5% for sensitivity analysis.

Calibration, uncertainties and sensitivity analysis

The model was calibrated using a Bayesian melding procedure [25] to reflect the HIV and HCV epidemics among Australian IDUs over the period 2000–2010. Credible intervals and upper and lower bounds for each parameter were derived from empirical data. These were used to define cumulative distribution functions (CDFs) for each model parameter (i.e., the priors). CDFs for each prevalence data point were defined similarly. A total of 1000 simulations were conducted, with parameter values chosen randomly based on the parameters’ inverse CDFs. The likelihood, as a measure of goodness-of-fit, of each simulation was defined as the product of the likelihood of each parameter value and the likelihood of each prevalence point, as defined by their respective CDFs. The posterior parameter distribution was obtained by randomly sampling 1000 times over the original simulations, with each simulation weighted by its likelihood. This resulted in 100 unique simulations being used for the final analysis. The values and interquartile ranges (IQRs) quoted in the remainder of the article are derived from this posterior distribution of simulations.

For sensitivity analysis, partial rank correlation coefficients (PRCCs) were calculated between incidence in 2010 and the sampled model parameters. All simulations and analyses were performed in MATLAB 2010a (The Mathworks, Natick, Massachusetts, USA).


The model accurately reproduced historical trends in the prevalence among IDUs for HIV and HCV in Australia from 2000 to 2010 [4] (Figs 3 and 4). With the current distribution of syringes, modeled HIV prevalence was relatively low and stable, reaching 1.1%, with IQR of 1.0–1.3% in 2010 (Fig. 3a). The model also matched HCV prevalence, which changed over time between 50 and 60% (Fig. 4a).

Fig. 3:
Annual number of cases of HIV among injecting drug users given historical needle-syringe program distribution trends versus projected cases in the absence of needle-syringe programs (NSPs).(a) HIV prevalence, (b) annual number of HIV diagnoses, (c) annual number of incident cases of HIV, and (d) number of people on antiretroviral therapy (ART); grey curve, model trajectory; black dots, observed data [4,41]; shaded region, model trajectories according to 25–50% receptive syringe sharing counterfactual scenario of no NSPs. Median model simulations are shown.
Fig. 4:
Annual number of cases of hepatitis C virus among injecting drug users given historical needle-syringe program distribution trends versus projected cases in the absence of needle-syringe programs.(a) HCV prevalence, (b) annual number of HCV diagnoses, (c) annual number of incident cases of HCV, and (d) number of people on HCV treatment; grey curve: model trajectory; black dots: observed data [4,41]; shaded region: model trajectories according to 25–50% receptive syringe sharing counterfactual scenario of no NSPs. Median model simulations are shown.

The model estimated substantial increases in both HIV and HCV infections according to the counterfactual assumptions in the absence of NSPs (with RSS levels between 25 and 50%). During 2000–2010, approximately 559–1240 new HIV infections and 120 000–178 000 new HCV infections would have occurred among IDUs without NSPs, resulting in prevalence levels of 1.2–1.5 and 66–80%, respectively (the lower and upper bounds represent the medians of the counterfactual scenarios with 25 and 50% levels of RSS, respectively). Thus, we estimate that NSPs resulted in the aversion of 192–873 HIV cases (34–70%) and 19 000–77 000 HCV cases (15–43%). The prevention of these infections would have resulted in a gain of 20 000–66 000 QALYs during 2000–2010, and 48 000–145 000 additional QALYs over a lifetime horizon (3% discounted).

In 2000–2010, an average of approximately 30 million syringes was distributed annually through NSPs in Australia, at an estimated cost of A$245 million (Table 1). The total healthcare costs attributed to HIV and HCV infections among IDUs over the same period were A$2.8 billion (IQR: $2.3–3.7 billion) and A$5.1 billion (IQR: $4.4–6.5 billion) when including future costs for current IDUs living with these infections (3% discounted). If NSPs did not exist in Australia, an estimated extra A$70–220 million would have been spent over 2000–2010 on HIV and HCV healthcare and A$340–950 million on additional future healthcare (3% discounted). The ICER for the current program compared to no program ranged from A$416 to 8750 per QALY (25–50% RSS) over 2000–2010. If one uses a maximum willingness to pay for healthcare interventions threshold of A$50 000 per QALY gained, as is commonly used in Australia, NSPs in Australia are cost-effective. Comparing future costs and outcomes, the investment in NSPs over the period 2000–2010 would be cost saving from 2032 onwards. In the lifetime model, each dollar spent on NSPs in 2000–2010 would have a return on investment of A$1.3–5.5 [(cost benefit − cost investment)/cost investment] in averted healthcare costs. Our uncertainty analysis findings are provided in the Appendix,

Table 1:
Estimated HIV-related and hepatitis C virus-related outcomes with and without needle-syringe programs.

Our sensitivity analysis revealed that the transmission probability per injecting episode using a contaminated syringe is the most important biological factor in determining both HIV and HCV incidence (PRCC: 0.78 for HIV and 0.60 for HCV), whereas the RSS level was the most sensitive behavioral factor (PRCC: 0.62 for HIV and 0.42 for HCV), followed by the frequency of injecting (PRCC: 0.51 for HIV and 0.27 for HCV). A full listing of parameter sensitivities is provided in the Appendix,


In this study, we evaluated the cost-effectiveness of syringe distribution in Australia, a country in which NSPs have been well established for over two decades. Using available biological, behavioral and program data in a mathematical modeling framework, we showed that NSPs are likely to have averted a substantial number of HIV and HCV infections.

Importantly, our evaluation suggests that NSPs have facilitated consistently low prevalence of HIV among IDUs in Australia. In many other regions of the world, HIV prevalence is very high among IDUs [26]. In contrast to their success in preventing an epidemic of HIV among IDUs, NSPs have not effectively controlled HCV in Australia. This is likely due to greater biological transmissibility of HCV compared with HIV [27] and that HCV was already well established in the Australian IDU population before NSPs were introduced [28]. Despite high NSP coverage in Australia, RSS is still relatively common among new IDUs and has been demonstrated to be strongly associated with increased risk of HCV infection [29].

In 2002, a cost-effectiveness analysis of the effect of NSPs in preventing HIV and HCV transmission among Australian IDUs was conducted to assess impact over the period 1991–2000 [30]. It was estimated that around 25 000 HIV and 21 000 HCV infections were prevented, with a net cost saving of A$20 million. In a recent report [8] we estimated that greater numbers of infections and healthcare costs would be saved according to less conservative assumptions regarding the counterfactual RSS.

Our model demonstrates that NSPs have been beneficial in preventing both HIV and HCV transmission among IDUs, are cost-effective in the short term and are cost saving when future health outcomes and costs are considered. It is important to note that this study assesses only the impact of NSPs in averting HIV and HCV infections among IDUs. It does not include other benefits such as preventing injecting-related injuries, psychosocial support and referral, education and prevention. Several other studies have demonstrated similar results in other contexts. In a systematic review of the international literature published in 2006, 13 economic evaluation studies of NSPs were identified [31]. The studies all concluded that NSPs were cost saving or cost-effective compared to the lifetime cost of HIV.

Representatives from all Australian jurisdictions were asked to provide the costs of NSPs for this analysis that related only to NSPs, but this was not always possible. It should also be noted that there are only small changes in some outcomes over the period 2000–2010, such as number of people who receive ART, develop hepatocellular carcinoma or liver failure and receive liver transplantation. The benefits in these outcomes become more marked over a long period as infections averted lead to aversions of these clinical and disease-related outcomes.

Although the assumptions of this analysis were based on the best available data, these data consisted of nonrandom samples and case notifications. This study was limited by reliance on self-reports of behavior from serial cross-sectional surveillance studies, which were based on self-administered questionnaires. Self-administration of surveys has been shown to reduce social desirability bias but is still not an ideal measure of actual risk [32–34]. Where possible, multiple prospective observational studies using varied methods and sampling techniques were also considered. The most crucial component of our analysis was the counterfactual assumption in the absence of NSPs. Studies conducted prior to the introduction of NSPs could not be used for this analysis due to their use of incomparable survey questions, however, the 70–90% levels of RSS reported in these studies provides an indication of high pre-NSP RSS levels in Australia. International studies provide some support for our counterfactual scenario [35–37]. Canada RSS recently increased from 10 to 23% following the closure of the only fixed NSP in the city of Victoria, but there was no change in RSS in Vancouver where NSPs remained open [38].

Although we incorporated uncertainty in parameter estimates where possible, many of the differences that exist between IDUs could not be captured in this population-based mathematical model. Our model did not include large heterogeneity, such as distinguishing current and former IDUs, and used weighted averages across the distributions of injecting frequencies. We believe the population-based model used in the current analysis is most appropriate for the data available and represents an advance over previous models of HIV and HCV among IDUs in Australia.

This study used single estimates of HIV and HCV healthcare costs based on national medical benefits schemes. Ranges in utility weights were used to calculate uncertainties in QALYs. Conservatively, the model did not consider the increases in HCV costs that are likely to occur in the future associated with direct-acting antiviral therapy. As it is unlikely that HCV prevalence will decrease substantially over the next 10 years due to the impact of improved HCV treatment unless uptake is enormously enhanced, the increase in the costs should not affect the conclusions related to the cost-effectiveness of prevention programs.

Our model did not consider the reduction of RSS due to opioid substitution therapy, which has been associated with reductions in opiate use, mortality and the transmission of HIV [39]. However, as opioid substitution therapy is not an effective treatment for people who inject drugs other than opiates, its potential impact depends on the proportion of IDUs who inject opioids. During the period 2006–2010 in Australia, approximately 30% of IDUs injected opioids [4]. Studies have also shown that HCV transmission can be reduced when NSPs and opioid substitution therapy are delivered in combination [40].

Although NSPs in Australia are effective, there is also a need for a range of complementary evidence-based public health responses to further prevent HIV and HCV transmission among IDUs. These include biomedical and behavioral prevention interventions which target injecting risk behaviors, and interventions designed to increase access to and early uptake of HCV treatment. In conclusion, our study provides evidence that the ongoing implementation of NSPs in Australia has led to substantial public health benefits and cost savings.


The authors would like to thank the members of a Reference Group who provided advice and guidance in the project's initial facilitation: Liz Sutton, Lisa Ryan, Deborah Warneke-Arnold, Fiona Poeder, Kate Dolan, Stuart Roberts, Nick Walsh, Michael Wooldridge, Don Baxter, Patrick Smith, Robyn Davies. The authors would like to thank each of the state and territory health department representatives for provision of data and other input: John Didlick, Owen Westcott, Rose Mason, James Broadfoot, Robert Kemp, Stephen Lymb, Francine Smith, Roland Jauernig, and Judith Bevan. The authors thank the organizations and people involved in the development and conduct of the Australian NSP Survey, and Karen Schneider and Anne Magnus for useful discussions.

J.A.K. implemented methods, produced results, and wrote the manuscript; J.A. designed health economic analyses, collated programmatic, costing and utility data and co-lead the study; C.C.K. assisted in the design of the mathematical model, model development and writing the manuscript; H.H.T. assisted with the design and implementation of analyses of impact and economic evaluation; L.Z. assisted in model design and implementation; J.I. provided data; G.J.D. assisted in the study design and interpretation of results; J.M.K. instigated the study and acted in an advisory role; M.G.L. facilitated stakeholder meetings, assisted in the study design and interpretation of results; L.M. provided data and advice, assisted in interpretation of results and editing the manuscript; D.P.W. led the study, providing oversight in the design, implementation, interpretation of findings, and manuscript writing.

This study was funded from the following sources: the Australian Government Department of Health and Ageing, the University of New South Wales. The Kirby Institute 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. The Kirby Institute is affiliated with the Faculty of Medicine, University of New South Wales. G.J.D. is supported by an NHMRC Practitioner Fellowship. L.M. and J.M.K. are supported by NHMRC Research Fellowships. D.P.W. is supported by an ARC Future Fellowship. H.H.T. is supported by the Ontario Institute for Cancer Research Health Services Research Program New Investigator Award.

Conflicts of interest

All authors have no conflicts of interest.


1. Wasley A, Alter MJ. Epidemiology of hepatitis C: geographic differences and temporal trends. Semin Liver Dis 2000; 20:1–16.
2. Aceijas C, Stimson GV, Hickman M, Rhodes T. Global overview of injecting drug use and HIV infection among injecting drug users. AIDS 2004; 18:2295–2303.
3. Wodak A, Dolan K, Imrie AA, Gold J, Wolk J, Whyte BM, et al. Antibodies to the human immunodeficiency virus in needles and syringes used by intravenous drug abusers. Med J Aust 1987; 147:275–276.
4. Iversen J, Topp L, Maher L. Australian NSP Survey National Data Report 1995–2010: prevalence of HIV, HCV and injecting and sexual behaviour among NSP attendees. Sydney, Australia: The Kirby Institute, The University of New South Wales; 2011.
5. Hurley SF, Jolley DJ, Kaldor JM. Effectiveness of needle-exchange programmes for prevention of HIV infection. Lancet 1997; 349:1797–1800.
6. Monterroso ER, Hamburger ME, Vlahov D, Des Jarlais DC, Ouellet LJ, Altice FL, et al. Prevention of HIV infection in street-recruited injection drug users. The Collaborative Injection Drug User Study (CIDUS). J Acquir Immune Defic Syndr 2000; 25:63–70.
7. Des Jarlais DC, Marmor M, Paone D, Titus S, Shi Q, Perlis T, et al. HIV incidence among injecting drug users in New York City syringe-exchange programmes. Lancet 1996; 348:987–991.
8. Return on investment 2: evaluating the cost-effectiveness of needle and syringe programs in Australia. Sydney, Australia: The Kirby Institute, The University of New South Wales; 2009.
9. Topp L, Day CA, Iversen J, Wand H, Maher L. Fifteen years of HIV surveillance among people who inject drugs: the Australian Needle and Syringe Program Survey 1995–2009. AIDS 2011; 25:835–842.
10. Darke S, Cohen J, Ross J, Hando J, Hall W. Transitions between routes of administration of regular amphetamine users. Addiction 1994; 89:1077–1083.
11. Carruthers S, Loxley W. Hepatitis C and young drug users: are they about to join the epidemic?. Aust J Public Health 1995; 19:421–424.
12. Crofts N, Aitken CK. Incidence of bloodborne virus infection and risk behaviours in a cohort of injecting drug users in Victoria, 1990–1995. Med J Aust 1997; 167:17–20.
13. Lenton S, Tan-Quigley A. The fitpack study: a survey of ’hidden’ drug injectors with minimal drug treatment experience. Perth, Australia: National Centre for Research into the Prevention of Drug Abuse, Curtin University of Technology; 1997.
14. MacDonald M. Monitoring HIV and HCV infection among injecting drug users: the Australian approach. Unpublished Thesis (PhD); University of New South Wales; 2001.
15. Blacker P, Tindall B, Wodak A, Cooper D. Exposure of intravenous drug users to AIDS retrovirus, Sydney, 1985. Aust N Z J Med 1986; 16:686–690.
16. Boughton CR, Hawkes RA. Viral hepatitis and the drug cult: a brief socio-epidemiological study in Sydney. Aust N Z J Med 1980; 10:157–161.
17. Philpot CR, Harcourt CL, Edwards JM. Drug use by prostitutes in Sydney. Br J Addict 1989; 84:499–505.
18. National Admitted Patient Care Collection (NAPCC) 2004–05 Clinical Profiles 2004–05. Department of Health and Ageing; 2005.
19. Round 11 (2006–07) Cost Report: Public version 5.1. Department of Health and Ageing; 2007.
20. Medicare Australia. MBS online. Medicare Australia; 2008.
21. Pharmaceutical Benefits Schedule. Department of Health and Ageing; 2008.
22. CPI: groups, weighted average of eight capital cities, index numbers and percentage changes. Australian Bureau of Statistics; 2011.
23. Tengs TO, Lin TH. A meta-analysis of utility estimates for HIV/AIDS. Med Decis Making 2002; 22:475–481.
24. Thein HH, Krahn M, Kaldor JM, Dore GJ. Estimation of utilities for chronic hepatitis C from SF-36 scores. Am J Gastroenterol 2005; 100:643–651.
25. Hallett TB, Gregson S, Mugurungi O, Gonese E, Garnett GP. Assessing evidence for behaviour change affecting the course of HIV epidemics: a new mathematical modelling approach and application to data from Zimbabwe. Epidemics 2009; 1:108–117.
26. Mathers BM, Degenhardt L, Phillips B, Wiessing L, Hickman M, Strathdee SA, et al. Global epidemiology of injecting drug use and HIV among people who inject drugs: a systematic review. Lancet 2008; 372:1733–1745.
27. Thomas DL, Astemborski J, Rai RM, Anania FA, Schaeffer M, Galai N, et al. The natural history of hepatitis C virus infection: host, viral, and environmental factors. JAMA 2000; 284:450–456.
28. Moaven LD, Crofts N, Locarnini SA. Hepatitis C virus infection in Victorian injecting drug users in 1971. Med J Aust 1993; 158:574.
29. Maher L, Li J, Jalaludin B, Chant KG, Kaldor JM. High hepatitis C incidence in new injecting drug users: a policy failure?. Aust N Z J Public Health 2007; 31:30–35.
30. Return on investment in needle and syringe programs in Australia report. Sydney, Australia: National Centre in HIV Epidemiology and Clinical Research, University of New South Wales; 2002.
31. Jones PL, Sumnall H, McVeigh J, Bellis M. A review of the effectiveness and cost-effectiveness of needle and syringe programmes for injecting drug users. Liverpool, UK: Centre for Public Health, Liverpool John Moores University; 2006.
32. Des Jarlais DC, Paone D, Milliken J, Turner CF, Miller H, Gribble J, et al. Audio-computer interviewing to measure risk behaviour for HIV among injecting drug users: a quasi-randomised trial. Lancet 1999; 353:1657–1661.
33. White B, Day C, Maher L. Self reported risk behaviour among injecting drug users: Self versus assisted questionnaire completion. AIDS Care 2007; 19:441–447.
34. De Irala J, Bigelow C, McCusker J, Hindin R, Zheng L. Reliability of self-reported human immunodeficiency virus risk behaviors in a residential drug treatment population. Am J Epidemiol 1996; 143:725–732.
35. Keene J, Stimson GV, Jones S, Parry-Langdon N. Evaluation of syringe-exchange for HIV prevention among injecting drug users in rural and urban areas of Wales. Addiction 1993; 88:1063–1070.
36. Donoghoe MC, Stimson GV, Dolan K, Alldritt L. Changes in HIV risk behaviour in clients of syringe-exchange schemes in England and Scotland. AIDS 1989; 3:267–272.
37. Watters JK, Estilo MJ, Clark GL, Lorvick J. Syringe and needle exchange as HIV/AIDS prevention for injection drug users. JAMA 1994; 271:115–120.
38. Ivsins A, Chow C, Marsh D, Macdonald S, Stockwell T, Vallance K. Drug use trends in Victoria and Vancouver, and changes in injection drug use after the closure of Victoria's fixed site needle exchange (CARBC Statistical Bulletin). Victoria, British Columbia: University of Victoria; 2010.
39. Darke S, Ross J, Teesson M, Ali R, Cooke R, Ritter A, et al. Factors associated with 12 months continuous heroin abstinence: findings from the Australian Treatment Outcome Study (ATOS). J Subst Abuse Treat 2005; 28:255–263.
40. Turner KM, Hutchinson S, Vickerman P, Hope V, Craine N, Palmateer N, et al. The impact of needle and syringe provision and opiate substitution therapy on the incidence of hepatitis C virus in injecting drug users: pooling of UK evidence. Addiction 2011; 10:1978–1988.
41. HIV/AIDS, Viral Hepatitis & Sexually Transmissible Infections in Australia Annual Surveillance Report: 2000–2010. National Centre in HIV Epidemiology and Clinical Research, University of New South Wales; 2011.

    Australia; cost-effectiveness; hepatitis C virus; HIV; injecting drug use; mathematical model; needle-syringe programs

    Supplemental Digital Content

    © 2012 Lippincott Williams & Wilkins, Inc.