A record-linkage study of the development of hepatocellular carcinoma in persons with hepatitis C infection in Scotland

Abstract

We investigated trends in first-time hospital admissions and deaths attributable to hepatocellular carcinoma (HCC) in a large population-based cohort of 22 073 individuals diagnosed with hepatitis C viral (HCV) infection through laboratory testing in Scotland in 1991 -2006. We identified new cases of HCC through record-linkage to the national inpatient hospital discharge database and deaths registry. A total of 172 persons diagnosed with HCV were admitted to hospital or died with first-time mention of HCC. Hepatocellular carcinoma incidence increased between 1996 and 2006 (average annual change of 6.1, 95% confidence interval (CI): 0.9 -11.6%, P ¼ 0.021). The adjusted relative risk of HCC was greater for males (hazard ratio ¼ 2.7, 95% CI: 1.7 -4.2), for those aged 60 years or older (hazard ratio ¼ 2.7, 95% CI: 1.9 -4.1) compared with 50 -59 years, and for those with a previous alcohol-related hospital admission (hazard ratio ¼ 2.5, 95% CI: 1.7 -3.7). The risk of individuals diagnosed with HCV developing HCC was greatly increased compared with the general Scottish population (standardised incidence ratio ¼ 127, 95% CI: 102 -156). Owing to the advancing age of the Scottish HCV-diagnosed population, the annual number of HCC cases is projected to increase, with a consequent increasing burden on the public healthcare system. British Journal of Cancer (2008) The aetiology of hepatocellular carcinoma (HCC) includes excessive alcohol consumption and chronic infection with the hepatitis B or C virus. The incidence of HCC has increased in developed countries worldwide (Deuffic et al, 1999;El-Serag and Mason, 1999;Law et al, 2000), but data are limited on its cause. However, modelling initiatives have projected increasing numbers of HCC deaths attributable to hepatitis B and/or hepatitis C virus (HCV) infection (Deuffic-Burban et al, 2004Sweeting et al, 2007).In Scotland, HCV is responsible for a significant burden on healthcare (Hutchinson et al, 2006). Meeting the public health challenge of providing sufficient resources for treatment and care of patients with HCV requires up-to-date data on morbidity and mortality. However, national surveillance systems for detecting advanced HCV-related liver disease are currently lacking in the United Kingdom and elsewhere. The existence of high-quality national HCV diagnosis and hospital discharge databases provided the opportunity to use record-linkage techniques to investigate the annual number of new diagnoses of, and deaths from, HCV-related HCC.Our principal goal was to investigate trends in the annual incidence of HCC in the entire HCV-diagnosed population in Scotland through record linkage to hospital admission and deaths records. We also aimed to confirm previously reported associations between risk factors and progression to HCC. Up-to-date information regarding factors affecting the rate of development of HCC in individuals diagnosed with HCV is needed for informing clinical management and care, and for refining guidelines for early detection. MATERIALS AND METHODSThe desi...

Hospital-related morbidity in people notified with hepatitis C: A population-based record linkage study in New South Wales, Australia

“…Previous population-based linkage studies have examined risk of death from alcoholic liver disease and show a similar excess overall (standardized mortality ratios [8][9][10][11][12][13][14][15] [2,5] and a higher excess in females. The trends in our cohort are of concern, as high levels of alcohol intake are associated with an increased risk of fibrosis progression and liver cancer in people with chronic HCV infection [6,23].…”

“…Eligible cases were also excluded if age or sex were missing. All hospital admissions in the cohort before or beginning within 14 days of diagnosis were excluded to reduce the bias towards higher rates of admission around the time of diagnosis [6]. In both the cohort and NSW population, duplicate and nested admissions (i.e.…”

“…Previous studies have reported a 3-to 5.5-fold increased risk of death and a higher incidence of liver cancer and immunoproliferative malignancies [1][2][3][4][5][6]. HCV is also now the most commonly reported primary indication for liver transplant in many industrialised countries including Australia [7,8].…”

Evidence of continued injecting drug use after attaining sustained treatment-induced clearance of the hepatitis C virus: Implications for reinfection

“…This probabilistic linkage technique is estimated to have a rate of false positives or false negatives under 5% (Kendrick & Clarke, 1993); the probabilistic linkage has also been previously described (McDonald et al, 2008).…”

Hospitalization rates associated with hepatitis B and HIV co-infection, age and sex in a population-based cohort of people diagnosed with hepatitis C

“…All hospital admissions before or beginning within 14 days of diagnosis were excluded to reduce the bias towards higher rates of admission around the time of diagnosis [6]. Duplicate and nested admissions (i.e.…”