Return on investment in needle and syringe programs in Australia: report

3.3 Methodology for estimates of numbers of people living with HCV acquired through injecting drugs

Page last updated: 2002

3.3.1 Assumptions used in models of HCV incidence
3.3.2 Estimates of rates of HCV-related liver disease progression
3.3.3 Estimates of the number of injecting drug users living with HCV without NSPs
3.3.4 Estimated effect of NSPs in reducing numbers of injecting drug users with hCV by disease stage

3.3.1 Assumptions used in models of HCV incidence

The incidence of HCV in Australia was modelled based on the following assumptions regarding the past pattern of injecting drug use in Australia:
  • There were 100,000 regular IDUs in 1997 (Hall et al. 2000), with a constant net increase of 8% per year since 1970, and with 5% stopping injecting each year (Thorley 1981; English et al. 1995).

  • There were 175,000 occasional IDUs in 1997, with a constant net increase of 8% per year since 1970, and with 10% stopping injecting each year (Law 1999).

  • There were no IDUs in 1960, with a linear increase in the number of both regular and occasional IDUs between 1960 and 1970.
Other assumptions made in modelling HCV incidence were the same as those adopted by the HCV Projections Working Group (Law 1999):
  • 65% of IDUs who start injecting regularly have previously injected occasionally (from the Delphi study).

  • The HCV incidence rate in uninfected regular IDUs was taken to be 18% per annum from 1960 until 1985, after which it was taken to decrease linearly to 13% in 1989 and thereafter.

  • The HCV incidence rate in occasional IDUs was taken to be 20% of that in regular IDUs.

  • All people starting or stopping injecting, or becoming regular rather than occasional IDUs, did so independent of their HCV status.

  • HCV incidence due to receipt of infected blood or blood products was taken to be 15% of HCV incidence in IDUs until the early 1980s, after which it was assumed to have gradually decreased following the introduction of donor self-deferral related to injecting drugs (which began in 1983), and to be stopped entirely from 1990 onwards with the introduction of blood donor screening for HCV.

  • HCV incidence through other transmission routes (such as needle stick injuries in health care workers, or tattoos) was taken to be 10% of HCV incidence in IDUs between 1987 and 1997, reflecting the data on risk factors for recent incident HCV infections. Prior to 1987 it was assumed to increase linearly to 20% of HCV incidence in IDUs in 1977, and then fixed at this absolute number of infections per year prior to this, again broadly consistent with data on risk factors for prevalent HCV infections, and for people with HCV infection attending liver clinics.

  • The number of HCV infections between 1950 and 1960 was held constant at a low level proportional to the modelled HCV incidence among IDUs. Any HCV infections prior to 1950 were assumed to have negligible effect on estimates and projections, and were not modelled.
Top of page

3.3.2 Estimates of rates of HCV-related liver disease progression

It was assumed that 75% of people exposed to HCV developed HCV chronic infection (i.e. 25% of exposed people cleared HCV) (Law 1999). Of people with chronic HCV infection, it was assumed that one third had normal ALT values, one third abnormal ALT values, and one third abnormal ALT values with further covariates which indicate they would be at increased risk of progression (eg high alcohol intake). Rates of progression from stage 0/1 liver disease to stage 2/3 liver disease, and from stage 2/3 disease to cirrhosis are shown in Table 3.3.1.

Taken together, these assumptions combine so that of all people exposed to HCV, 5.3% are estimated to develop cirrhosis by 20 years, with 7.1% of people with chronic HCV developing cirrhosis by 20 years. This is consistent with current evidence regarding progression rates to cirrhosis (Freeman et al. 2001).

Rates of developing liver failure or hepatocellular carcinoma (HCC) from cirrhosis were assumed to be 4% and 1% respectively (Fattovich et al. 1997). It was further assumed that HCC could develop following liver failure, but not vice-versa. HCV-related mortality following cirrhosis was taken to be 1.5% per annum (Fattovich et al. 1997).

Mortality unrelated to HCV, both before and after cirrhosis, was assumed to be 1% per year (Thorley 1981; English et al. 1995) due to injecting drug use. Background mortality was based on ABS life tables, assuming that the mean age at HCV seroconversion among IDUs was 25 years, and that there were 2 male HCV-infected IDUs for each female HCV-infected IDU (ABS 1995).

Estimates of the numbers of people living with HCV by disease stage, and the incidence of liver cancer and HCC, were derived by combining these progression rates with the HCV incidence pattern estimated through the models described above.

Table 3.3.1 Annual rates of liver disease progression

Disease stageStage 0/1 to Stage 2/3Stage 2/3 to Cirrhosis
Not chronic HCV
0%
0%
Chronic HCV, normal ALT
1%
1%
Chronic HCV, abnormal ALT
2%
2%
Chronic HCV, normal ALT and further cofactors
3%
3%

Note: Stage 0=no hepatic fibrosis, stage 1=minimal hepatic fibrosis; stage 2=moderate hepatic fibrosis; stage 3=severe hepatic fibrosis; stage 4=cirrhosis.

3.3.3 Estimates of the number of injecting drug users living with HCV without NSPs

The modelled estimate of HCV incidence in Australia that has occurred with NSPs described above corresponds to a gradual increase in HCV prevalence among regular IDUs until the mid- to late-1980s, followed by a gradual decline to around 52% HCV prevalence in 2000. NSPs were first introduced in Australia in late 1987. Hence, NSPs were assumed to have reduced HCV prevalence among IDUs from 1988 onwards. The pattern of HCV prevalence if NSPs had not been introduced was estimated by assuming that HCV prevalence would have remained constant at 1988 levels from 1988 onwards. From this, a pattern of HCV incidence if NSPs had not been introduced was derived. It was further assumed that the introduction of NSPs had no effect on HCV transmissions through routes other than injecting drug use.

Estimates of the numbers of people living with HCV by disease stage if NSPs had not been introduced were then derived using the same progression rate distributions described above.
Top of page

3.3.4 Estimated effect of NSPs in reducing numbers of injecting drug users with hCV by disease stage

To allow costing of the effect of NSPs in reducing the number of IDUs living with HCV, estimates of the reduction in the number of people living with HCV by disease stage were obtained by subtracting the estimates obtained with NSPs from the corresponding estimates without NSPs. In these analyses, HCV incidence due to injecting drug use was assumed to cease from 2001 onwards, and estimates were projected forward until all people infected with HCV were estimated to have died, either from HCV-related or unrelated mortality.