Cirrhosis is a chronic liver disease that often results from harmful alcohol consumption or infection with viral hepatitis. Patients with cirrhosis have a poor prognosis, in part because they have an increased risk of many cancers. We aim to examine the risk of cancer in patients with cirrhosis, and to contrast this risk with that for people without cirrhosis.
Not all patients with cirrhosis share the same risk of cancer. We are primarily interested in the impact of socioeconomic status. We speculate that deprived patients have a longer diagnostic delay, i.e., the time from suspicion of cancer to imaging to diagnosis of cancer is longer, and therefore they have more advanced cancers when cancers are diagnosed. We also believe that deprived patients develop more cancers, but that is because of their unhealthier lifestyle, not because of their socioeconomic status.
Among patients with cancer, survival is likely to be worse for those who also have cirrhosis. It is less clear whether, among patients with cancer and cirrhosis, socioeconomic status affects survival. For example, they may be less able to speak up for themselves if something is wrong. We are interested in determining the impact of socioeconomic status on survival of patients with cancer, as well as determining whether any impact is different among patients with cirrhosis than among patients without cirrhosis.
Using linked data we will be able to identify socioeconomic discrepancies in the healthcare management of patients with cirrhosis, and we can do so in far greater depth than has previously been achieved.
CANCER RISK
Patients with cirrhosis have an increased risk of many cancers, and they have a high absolute risk of hepatocellular carcinoma and head and neck cancer. Due to the high risk they might benefit from screening examinations, but such decisions require more accurate information than what is currently available. We aim to provide that information using linked data from hospital episode statistics, diagnostic imaging data, cancer registry data, and individual socioeconomic status. We will contrast absolute risk between cirrhosis patients and matched controls from the general population.
Among patients with cirrhosis, we will examine risk factors for developing cancer. Our primary interest is in the effect of socioeconomic status on cancer incidence and cancer stage at diagnosis. We will use regression modelling to compare cancer incidence/stage and diagnostic routes across categories of socioeconomic status.
CANCER SURVIVAL
We will compare all-cause and cancer-related mortality from cancer diagnosis between patients with cirrhosis and the population controls. Depending on the number of patients with a specific cancer type, we will conduct separate analyses for the most prevalent cancer types. We will use Cox regression to estimate the effect of cirrhosis on mortality following cancer diagnosis, controlling for confounding.
Within the cohort of patients with cirrhosis we will focus on the effect of individual socioeconomic status, examining the hypothesis that deprived patients are disadvantaged. We will also examine whether any discrepancies in survival between deprived and affluent patients are increasing over calendar time.
Finally, we will examine the interaction between the effects of cirrhosis and low socioeconomic status on survival time among patients with colorectal cancer. This will be based on a Cox regression model including an interaction term between cirrhosis and socioeconomic status, and confounders.
Incidence rate and cumulative risk of cancer: A) for patients with cirrhosis compared to a general population sample without cirrhosis, and B) among patients with cirrhosis of differing socioeconomic status.
Survival time and cause-specific mortality after cancer diagnosis: A) for patients with cirrhosis compared to a general population sample without cirrhosis, and B) among patients with cirrhosis of differing socioeconomic status.
Joe West - Chief Investigator - University of Nottingham
Peter Jepsen - Corresponding Applicant - Aarhus University Hospital
Colin Crooks - Collaborator - University of Nottingham
Gro Askgaard - Collaborator - Aarhus University Hospital
Joanne Morling - Collaborator - University of Nottingham
Kate Fleming - Collaborator - University of Liverpool
Morten Daniel Jensen - Collaborator - Aarhus University Hospital
Timothy Card - Collaborator - University of Nottingham
Trevor Hill - Collaborator - University of Nottingham
HES Admitted Patient Care;HES Diagnostic Imaging Dataset;HES Outpatient;NCRAS Cancer Registration Data;NCRAS National Radiotherapy Dataset (RTDS) data;NCRAS Systemic Anti-Cancer Treatment (SACT) data;ONS Death Registration Data;Patient Level Index of Multiple Deprivation