This is the first global systematic review of HCV and HBV infection among people who inject drugs and the first to produce regional and global population estimates. We estimate midpoint populations of IDUs who may be HCV-positive of around 10.0 million, and those who are HBsAg-positive of around 1.3 million. Clear geographic differences exist in prevalence. Eastern Europe, and East and South East Asia are estimated to hold the largest populations of IDUs infected with viral hepatitis.
It is important to note that the population size estimates reported here refer to the estimated number of anti-HCV, anti-HBc or HBsAg-positive people who are current or recent injecting drug users not people who have ever injected drugs. This is the same methodology undertaken for HIV previously1
. Many people who inject drugs cease injecting at some point25
, so the current estimates cannot be interpreted as the total number of cases of HCV or HBV attributable to IDU. Given the limitations in current knowledge of the natural history of IDU (such as the range in duration of injecting, and the likelihood and timing of resuming IDU following cessation), particularly in low and middle-income countries, it is not possible to make defensible regional and global estimates of the number of former
IDUs, and the numbers of whom may be anti-HCV, anti-HBc and HBsAg-positive. An estimate of the burden of chronic viral hepatitis among current IDUs is essential for assessing secular trends in the risk of infection, the impact of control strategies, and the importance of implementing these, as well as implications for future burden of disease and health care requirements.
Efforts to prevent, treat and reduce harms related to liver disease among IDUs are essential, particularly in situations where HIV has successfully been prevented or managed, since the larger numbers of IDUs infected with HCV, and significant morbidity arising from this infection, mean that the health and economic costs of HCV transmitted by IDU may be as high as (or higher than) those of HIV. HCV treatment remains under-utilised, however10
. Part of the reason for this is the high cost, which will remain a significant barrier to increasing treatment coverage in resource poor settings until this is reduced. There is increasing attention on this issue among international groups who are advocating for cost reductions, generic manufacturing, and changes to licensing conditions10, 26
. Not long ago, the high cost of HIV antiretrovirals similarly prevented access in high prevalence, low-income countries: in recognition of this barrier, there are growing efforts to bring viral hepatitis treatments into the same (lower cost) access framework as HIV antiretrovirals10
. Another barrier is the toxicity of HCV treatment, although a large number of new HCV drugs are being developed, which will revolutionise HCV treatment in the next few years27
Greater attention needs to be paid to reducing the impact of other causes of liver disease progression in people chronically infected with viral hepatitis. This includes addressing alcohol use problems, and providing HAV and HBV vaccination, particularly given evidence that liver related disease will become a major cause of mortality as IDUs age28
Evidence regarding the impact of needle and syringe programs29
and provision of other injecting paraphernalia upon prevention of HCV infection is limited, yet reduction of risk remains paramount, particularly during the period of initiation to injecting, when HCV incidence is highest6, 14
. The potential for HCV treatment to reduce HCV prevalence among IDU populations and in this way reduce the force of infection acting on susceptible members of these populations has been supported by mathematical modelling30
. This potential role of HCV treatment in the prevention of HCV transmission among IDU populations warrants further investigation.
Although significant variability in HBsAg prevalence reports was observed, prevalence typically reflected the differences in the prevalence of HBV infection in the general population. In low-intermediate prevalence countries, the prevalence of HBsAg in IDU was typically <10%, whereas in high-HBV prevalence countries, prevalence of HBsAg among IDUs was in the order of 10–20% (e.g. East and South-East Asia). Given the high prevalence of chronic HCV infection in IDUs, HBV infection is particularly likely to represent HBV/HCV co-infection, which is associated with more rapid progression of liver disease and attendant mortality31
; this is similarly the case for coinfection between HIV and viral hepatitis32
Effective treatments for CHB are available, which reduce progression of liver disease and complications such as HCC33
. However, antiviral therapy for CHB is often of indefinite duration, and access to modern, potent drugs with high resistance barriers is limited in many high prevalence, resource-poor settings. Barriers to accessing treatment and care for CHB result in poor outcomes for those affected, and ongoing transmission to susceptible contacts.
Vaccination against HBV must be prioritised for all susceptible IDU, especially those already infected with HCV. However, selective vaccination programs against HBV in this group have often been characterised by low uptake and difficulty reaching most at-risk individuals34
. A significant reduction in the burden of HBV infection amongst IDU is expected in countries with universal infant vaccination programs once these individuals reach the age at which acquisition through IDU is most common. Correctional facilities provide one opportunity to vaccinate, treat and reduce the transmission of viral hepatitis in a population with high levels of IDU, HBV, and HCV, many of whom cycle in and out of the community35
Limitations of existing data
There are a number of important limitations of existing data. One issue concerns the way in which HCV and HBV infection are measured and reported across studies: reporting was typically on only one (or perhaps two) markers, making estimates of the true prevalence of chronic HBV and HCV difficult. Without the measurement and reporting of multiple markers (anti-HCV plus HCV RNA-PCR, or HBsAg plus anti-HBc and ideally anti-HBs) it is not possible to make more accurate estimates of chronic infection, past infection, susceptibility or immunity. For HCV, we have estimated the number of current IDUs who are anti-HCV positive: this is not a measure of total chronic HCV infection, but rather HCV exposure
among IDUs, given that a minority (perhaps 20%) of those infected with HCV (who would test positive for anti-HCV) are likely to have cleared the virus4
For HBsAg, wide ranges in reports that met inclusion criteria were observed. In addition, where anti-HBc was not reported it was not possible to determine what proportion of HBsAg positive individuals were acutely infected and within the window period prior to anti-HBc seroconversion. Future studies should include both markers to allow a more accurate understanding of study results. Additional sample details including country of birth and ethnicity would also assist interpretation36
. An additional limitation of existing data concerns a lack of data on the age range of samples and individuals duration of drug injecting history and hence period of elevated exposure to viral hepatitis, which would permit more accurate understanding of varying prevalence of both HCV and HBV among samples. The reliance upon older studies, with less accurate serological testing modalities, small sample sizes, and those conducted in countries where laboratory capacity is limited, increases uncertainty about the validity of both HCV and HBV reports.
A final issue concerns the representativeness of IDU samples. Some studies of HBV and HCV recruited “lifetime” IDUs, whereas others recruited “current” or “past year” IDUs. They also were recruited from a variety of locations: prisons, drug treatment centres, outpatient clinics and other medical settings, where IDUs may differ in their risk behaviour and exposure to viral hepatitis. Further, convenience sampling is most often used, so it is possible that the samples do not represent the IDU population from which they are drawn. Data were also typically sub-national, from a limited number of locations that may or may not be representative of the epidemic nationally, particularly in larger countries where there may be considerable geographic variation, potentially limiting national representativeness.
Limitations of this review
We have used the same methods as in our previous reviews, which have been detailed elsewhere1
. As those previous investigations also revealed, limitations include the current concentration of peer-reviewed literature from high-income countries, the review being conducted by a small team, and the greater potential for papers in languages other than English to be overlooked. As we did for the earlier reviews we attempted to address these limitations by consulting widely with experts and stakeholders, seeking unpublished reports and verification of the data and reports included and enlisting the support of UN and other agencies, who facilitated access to data and contact with relevant in-country personnel.
The response to blood borne virus transmission among IDUs has primarily focussed on HIV. Maintaining and strengthening the response to HIV among IDUs remains crucial, but the significance of viral hepatitis must receive greater recognition than is currently the case. Investment in, and development of, comprehensive and effective strategies to prevent the transmission of viral hepatitis, and reduce resultant morbidity and mortality among IDUs, are urgently required. The Viral Hepatitis resolution of the 63rd
World Health Assembly10
requested that the Director-General collaborate with all relevant stakeholders in supporting surveillance, prevention, and treatment goals, especially in developing countries. It is essential that policies and strategies for addressing viral hepatitis include IDUs, who are at elevated risk and often have poorer access to services. The current review has provided estimates of the scale of this problem at country, regional and global levels. These findings should inform efforts to efforts to accurately scale and appropriately target the response.