Skip to main content

Table 1 Summary of main data sources for the model

From: Modelling the burden of hepatitis C infection among people who inject drugs in Norway, 1973–2030

 

Source

Used to hardcode parts of the model

Infectivity of PWIDs

The infectivity of PWIDs was modulated by a) the proportion of PWIDs who were infectious, b) the dispersion of the injecting epidemic throughout Norway (the more disperse, the less infectious) (dispersion over time is shown in Additional file 1: Figure S1), and c) the coverage of needle and syringe programs (Additional file 1: Figure S1).

People who inject drugs

Norwegian Institute for Alcohol and Drug Research (SIRUS) has non-internally consistent numbers of new PWIDs and prevalence of active PWIDs, former PWIDs who will relapse, and former PWIDs who will never relapse for each year from 1973 to 2013. We subsequently estimated internally consistent PWID numbers from 1973 to 2030 (Additional file 1: Figure S2).

Age of injecting debut

Age of injecting debut was taken from SIRUS estimates in 1975, 1985, and 1995, and the life quality report from 2003 to 2012. We predicted mean age of injecting debut from 1973 to 2030 using linear regression (Additional file 1: Figure S3).

Model Targets

PWIDs with HCV

The proportion of PWIDs with HCV RNA was based on data collected through cross sectional health studies among PWIDs attending low threshold harm reduction-based health care centers in Oslo targeted towards drug users [12].

Cirrhosis

We received data from one hospital (Akershus universitetssykehus HF) on the number of PWIDs (current and former) treated with cirrhosis associated with HCV in 2013. We extrapolated the total number of cases treated in Norway by dividing by the hospital’s catchment area (10%).

HCC disease

The number of liver cancer or HCC cases were obtained from the Norwegian cancer registry (NCR) using ICD10 code C22 (Malignant neoplasm of liver and intrahepatic bile ducts, which includes HCC). Numbers were adjusted by the proportion of HCC among those with ICD10 code C22 (77%) and for disease attributable to HCV in Norway (26%), resulting in 20% of the extracted data estimated to be HCV associated HCC [25].

Liver transplants

From the Nordic Liver Transplant Registry (NLTR) [21] we used the number of all liver transplants performed from 2000 to 2013 with antibodies against HCV as targets for liver transplants in the model. Mortality within the first year after the LTX is considered as HCV related mortality. After the first year the individuals who received a LTX have the same risk as the rest of the population in the model.

Cirrhosis mortality

Aggregate data on death entries (2000–2013) were obtained from the “Norwegian Cause of Death Registry”. We included individuals with an underlying cause of cirrhosis (ICD10 codes K74.3, K74.4, K74.5 and K74.6). The mortality numbers were then adjusted to the attributable risk of hepatitis C (14%) [26].

HCC mortality

Aggregate data on death entries (2000–2013) were obtained from the “Cause of Death Registry”. We included individuals with an underlying cause of HCC (ICD10 code C22). The mortality numbers were adjusted for HCC among those with ICD10 code C22 (77%) and for to the attributable risk of hepatitis C (26%) [25]

Treatment rates

The Norwegian prescription database (NorPD) was used to estimate the treatment rates for HCV in Norway. Genotype distribution of HCV was based on data from the NIPH [12]. We estimated treatment success for each year, combining the absolute number of treatments from NorPD with treatment response and duration of treatment per genotype.

Used to calculate health estimates from output

Disability weightings

Using the Global Burden of Disease study 2010 disability weights [23], we assigned disability weights to the various health stages in the model.