The liver can be affected by a number of different diseases, such as hepatitis, alcohol abuse and obesity, which progressively damage the liver tissues. Over the long term, this can lead to permanent scarring (cirrhosis), which prevents the liver from performing its many important functions, and ultimately liver failure. The serious effects (complications) of cirrhosis include hepatic encephalopathy (HE): damage to the brain caused by the toxins that the liver can no longer remove from the blood.
In our study, we plan to examine how patients with liver disease experience healthcare – for example, through GP visits, referral to outpatient specialties, visits to accident & emergency. We will consider these patients in four groups of increasing severity: any liver disease, cirrhosis, HE and predicted HE. For patients in the HE groups, we will also examine how drug treatments affect survival. Patients with HE are commonly treated with rifaximin and lactulose, in combination or alone. We will therefore consider four treatment groups: no treatment recorded, rifaximin, lactulose, and combined rifaximin and lactulose.
To carry out the study, we wish to select adult patients from the Clinical Practice Research Datalink (CPRD) with a recorded diagnosis of liver disease, whom we will then categorise into the three severity groups by clinical history and prescription records.
The objective of this study is to better understand the how patients with liver disease experience healthcare by assessing treatment pathways, referral patterns by examining three populations of increasing liver disease severity; liver disease, cirrhosis and hepatic encephalopathy (HE). We plan to describe the characteristics, and comorbidities of these cohorts overall and split by treatment groups; no treatment recorded, rifaximin, lactulose, and combined rifaximin and lactulose. We also aim to identify the impact that Rifaximin and lactulose treatments have on the risk of mortality in the HE group.
Referral patterns will be investigated using the clinical and referral table within Clinical Practice Research Datalink (CPRD) and the linked Hospital Episode Statistics (HES) data to categorise clinical visits, inpatient admissions, outpatient appointments and emergency admissions to investigate patients disease journey within the UK heath service.
Mortality risk in patients will be investigated using time dependent cox regression models adjusting for treatment and associated decompensating events.
Patients within the four severity populations will be identified by:
Liver disease – identified by the first diagnosis of a liver related code within CPRD & HES identified from Read codes, OPCS and ICD-10 codes in CPRD GOLD and Aurum or prescriptions for Rifaximin 550mg
Cirrhosis– identified by the earliest diagnosis code for cirrhosis from within the liver disease population cohort identified by Read code, OPCS, ICD-10 codes in CPRD GOLD and Aurum.
Hepatic encephalopathy - identified from within the liver disease population with a Read code for HE or a Rifaximin 550mg prescription or both a Rifaximin 200mg prescription and a liver disease code.
Predicted Hepatic encephalopathy – (e.g. identified from Tapper et al.9 risk equation (only if validated in UK data)).
Patients will be identified post 2003 and must be of at least 18 years of age and of CPRD acceptable standard.
Health Outcomes to be Measured:
• Decompensating events such as HE
• Treatment pathways
• Referral patterns
HES Accident and Emergency;HES Admitted Patient Care;HES Outpatient;ONS Death Registration Data;Practice Level Index of Multiple Deprivation