Hepatitis C Virus Projections Working Group: Estimates and Projections of the Hepatitis C Virus Epidemic in Australia 2006

http://www.health.gov.au/internet/main/publishing.nsf/Content/phd-hepc-estimates-project-06-l~phd-hepc-estimates-project-06-l-2

2. Review of data sources on HCV in Australia

2.1. National surveillance

HCV infection has been a notifiable disease (doctor and/or laboratory) in most Australian State and Territory Health jurisdictions since 1990, and all States and Territories since 1995. Notifications increased rapidly to between 17,000 and 20,000 new HCV diagnoses annually during the period 1995 to 2001, but have since declined to around 13,000 to 15,000 notifications in the period 2002 to 2005 (AGDHA, 2006). Cumulative HCV notifications over the period 1990-2004 are over 225,000 (AGDHA, 2006). The extent to which there have been duplicate HCV notifications is uncertain. In each State/Territory, new HCV diagnoses have been notified with case-identifying data, so that within each State/Territory duplicate notifications are likely to be limited. However, new HCV diagnoses are forwarded by State/Territory Health Departments to the National Notifiable Diseases Surveillance System, maintained by the Australian Government Department of Health and Ageing, in anonymous aggregate format. This means that it is not possible to assess duplicate notifications between state/Territories.

Around 65% of HCV notifications have been in people aged 20-39 years, with approximately 35% of notifications in females (AGDHA, 2006). However, the proportion of female HCV notifications is larger than 35% in younger age- groups, particular age 15-19 years where there are similar numbers of men and women notified (AGDHA, 2006).

The vast majority of HCV notifications have been prevalent HCV diagnoses, with newly acquired HCV cases constituting 100 to 450 cases per year over the period 1997-2001 (NCHECR, 2001). Enhanced surveillance mechanisms to improve ascertainment of newly acquired HCV cases have been introduced in most State and Territory health jurisdictions (Spencer et al, 2002), but newly acquired HCV notifications remain low at around 300 cases per year between 2002 and 2005 (AGDHA, 2006). However, generally fewer than 25% and probably nearer 10% of HCV infections are associated with acute symptoms (van der Poel et al, 1994). Furthermore, identifying recent HCV infections solely on the basis of a recent previous negative test is limited because negative HCV test results cannot be cross-checked between different testing laboratories. These difficulties, in combination with the irregular testing of some people at increased risk of HCV infection and large resources required for enhanced HCV surveillance, mean that only a small minority of HCV infections can be diagnosed and notified close to the time of HCV exposure.

2.2. Risk factors for HCV infection

Studies examining the risk factors for HCV infection in Australia are summarised in Table 1. Routine HCV notification data, and the survey of antenatal patients, indicate that around 80% of prevalent HCV infections have occurred through injecting drug use. Recent studies of incident HCV notifications since 1995 indicate that of those cases where the transmission route was determined, the proportion of incident HCV infections due to injecting drug use was even higher, at around 90% (Andrews and Curran, 1996; Copland, 2002; Robotin et al, 2004; NCHECR, 2005). This may indicate that the proportion of HCV infections due to injecting drug use has increased in more recent years, but may also be biased upwards if HCV negative injecting drug users (IDUs) are retested more frequently than other population groups.

The proportion of HCV infections due to injecting drug use was lower in blood donors, probably because people with a history of injecting drug use are asked to exclude themselves from blood donation. Rates of HCV infection due to injecting drug use were also lower in liver clinic patients, between 51% and 75%, although this proportion had increased in one Melbourne liver clinic, from 51% during 1990-1993 (Strasser et al, 1995) to 64% during 1990-1998 (Ostapowicz et al, 2001). In the study by Li et al (1998), of liver clinic patients infected with HCV through routes other than injecting drug use, around half were immigrants to Australia from countries of high HCV prevalence.

 

2.3. Prevalence and incidence of HCV in injecting drug users

Studies of the HCV prevalence among IDUs in Australia appear in Table 2. Taken together, these studies indicate that the proportion of regular IDUs with HCV infection in Australia has been consistently in the range 50% to 70% since the early 1970s. Repeated surveys of people attending needle and syringe program (NSPs) showed some decrease in HCV prevalence from 63% in 1995 to 51% in 1996, but prevalence has remained around 50% over the period 1996 to 2004. Among IDUs, prevalent HCV infections have been found to be very strongly associated with duration of injecting (Crofts et al, 1997c). Age has also been found to be associated with HCV seroprevalence, but primarily through its association with duration of injecting (Crofts et al, 1997c).

Studies assessing the incidence of HCV infection in IDUs are summarised in Table 3. In a combined analysis, based on one cohort of IDUs in Melbourne and rural Victoria and one cohort in Sydney, the incidence of HCV infection among IDUs in the 1980s and early 1990s was estimated to be around 15 new infections per 100 person years of follow-up (Crofts et al, 1997c). These studies also provided an indication, though not statistically significant, of a reduction in HCV incidence among IDUs from around 18 infections per 100 person years in IDUs who started injecting prior to 1987, to 13 infections per 100 person years in IDUs who started injecting since then, coincident with the introduction of needle and syringe exchange programs, and other preventive campaigns, which were aimed at reducing the risk of human immunodeficiency virus (HIV) infection among IDUs. Data from people attending the Kirketon Road Centre (KRC) in Sydney indicate that HCV incidence in clinic attendees has remained broadly consistent, at around 15 to 20 new HCV infections per 100 person years, over the period 1992 to 2004 (van Beek et al, 1998; NCHECR, 2001; Gilmour et al, 2002; NCHECR, 2004; NCHECR 2005).

HCV incidence among IDUs is of particular interest for the period 1999 to 2005, following the reduction in the heroin supply in Australia from around late-2000. Incidence of HCV among IDUs in a cohort in urban, regional and rural settings in NSW was examined by year of injecting and main drug injected (Table 4). Incidence of HCV was strongly related to main drug injected, with rates of 12.5, 82.6 and 38.1 per 100 person years for amphetamine, cocaine and heroin respectively. Importantly, there was no evidence of a reduction in HCV incidence from 1999 to 2002, with if anything an increase in HCV incidence from 26.6 to 41.9 per 100 person years respectively.

Although generalisation from these selected populations is difficult, these data at least do not provide any evidence of a decrease in HCV incidence among IDUs during either the period from the mid- to late-1990s, or the period 1999 to 2002.

 

2.4. Prevalence and incidence of HCV in other populations

Studies estimating HCV prevalence among other populations in Australia are summarised in Table 5. HCV prevalence rates were around 1% or lower in antenatal patients, first time blood donors and health care workers. Rates were higher in renal transplant recipients, dialysis patients, and much higher in people with haemophilia, as a result of receiving blood and blood products contaminated with HCV prior to the availability of screening tests for blood donations.

HCV incidence has been estimated among repeat blood donors in Victoria to be 1.9 per 100,000 (Whyte and Savoia, 1997).
There has been one national serosurvey of HCV prevalence, based on 2,800 blood samples collected during 1996 to 1998 for other diagnostic tests, and stored at diagnostic laboratories (Amin et al, 2004). This study found a HCV prevalence of 2.3%, which if applied to the whole Australian population would translate into some 430,000 people living with HCV antibodies during the period 1996 to 1998.

2.4.1. HCV in prisons

The total prison population in Australia on 30 June 2005 was 25,353 persons, of whom 93% were men (ABS, 2006). Prison populations by State/Territory were: ACT 162; NSW 9,819; NT 820; QLD 5,354; SA 1,473; TAS 551; VIC 3,692; WA 3,482. Of the total prison population on 30 June 2005, 20,220 (79.8%) were sentenced prisoners while 5,133 (20.2%) were being held on remand prior to trial or sentencing.

The total number of people held in prison at anytime during 2005 is uncertain, but would certainly be greater than the 25,353 people held on 30 June 2005. Of the 20,220 sentenced prisoners held on 30 June 2005, 3,456 (17.1%) had an aggregate sentence length of less than one year, while 954 (4.7%) were periodic detainees, who are held in prison for two consecutive days each week but are at liberty for the remainder (ABS, 2005b). Furthermore, total sentenced receptions in prisons during 2005 were 24,532 (ABS, 2006). Taken together, these data suggest that of the order of 30,000 to 35,000 people were held in prisons in Australia at some time during 2005.

HCV prevalence among entrants to or inmates in prisons was assessed nationally in 2004, with a number of other assessments in particular States and Territories, largely Victoria and New South Wales, prior to this (Table 5). The national study of prison entrants, and studies in New South Wales and Victoria indicate HCV prevalence rates of the order of 40-60% of all prisoners over the period 1991-2004. Furthermore, a consistent finding in these studies was that HCV prevalence rates were 50% to 100% greater in women than men, perhaps reflecting that a greater proportion of women than men in prison are injecting drug users (Crofts et al, 1995; Butler et al, 1999; Awofeso et al, 2000; Butler et al, 2004; Hellard et al, 2004). A similar pattern of higher rates of HCV infection in females compared to males was also seen in juvenile detainees in New South Wales (NSW Department of Juvenile Justice, 2003). Rates appear somewhat lower in Western Australian prison inmates, at around 23% of male and 46% of female prisoners (Watson J, personal communication), and much lower in Darwin at 8% (Huffam et al, 1999). National rates of HCV prevalence in prison entrants have been estimated to be 34%, slightly lower than the estimates seen in New South Wales and Victoria, and hence are probably in the range 30% to 40% of all male prisoners. Available data suggest that HCV prevalence rates in female prisoners are higher than this, probably in the range 50% to 70%. These prevalence estimates among prisoners suggest that between 7,500 and 10,000 people being held in prison on 30 June 2005 were HCV antibody positive (5,500 to 7,500 with chronic HCV), of whom between 800 and 1,200 were women (600 to 900 with chronic HCV). It is probable that between 9,000 and 14,000 people with HCV antibodies (7,000 to 11,000 people with chronic HCV) were held in prisons at some time during 2005, of whom between 1,000 and 1,700 (750 to 1,300 with chronic HCV) were women. HCV transmission in prisons is difficult to assess as HCV prevalence among prisoners who have injected drugs is so high at entry. HCV transmission in prisoners in continuous imprisonment has been the subject of several recent papers (Haber et al, 1999; Post et al, 2001; Dolan et al, 2003; Butler et al, 2004). Rates of transmission of HCV in prisons are difficult to estimate with any certainty based on these data. Among prisoners assessed as eligible for methadone maintenance treatment, HCV incidence during continuous imprisonment has been estimated to be as high as 30 per 100 person-years (Dolan et al, 2003), while in prisoners in continuous imprisonment between two health surveys HCV incidence was 4.5 per 100 person years (Butler et al, 2004). It is, however, highly likely that HCV transmissions occur in prisons, and that unsterile injecting drug use and tattooing are the most common routes of transmission.

2.5. HCV genotype prevalence rates

Studies assessing HCV genotype prevalence rates are summarised in Table 6. A consistent feature of studies in Melbourne and Sydney is a trend to a decreasing prevalence of genotype 1 between the 1970s to the 1990s, with an increase in genotype 3 over the same period. During the 1990s, all the studies in Table 5 taken together indicate that the prevalence of HCV genotype 1 was around 55%, genotype 3 around 35%, genotype 2 between 5-10%, with other genotypes below 5% combined.