Risk stratification of chronic liver disease in the community

Introduction There is an appreciable undiagnosed burden of Chronic liver disease within our society with mortality rates increasing over recent decades. The reasons for this include the rise in lifestyle related risk factors (e.g. hazardous alcohol use, type 2 diabetes and obesity), the asymptomatic nature of the disease which ensures patients do not present to a clinician at an early stage, and the ineffectiveness of current diagnostic tests available to primary care physicians. By transforming the diagnostic pathways we may be able to start proactively identifying and risk stratifying patients with liver disease. Implementation of a risk stratification pathway into a primary healthcare setting would allow a large number of patients to be stratified and identify those who would benefit from a referral to specialist services or further follow up in secondary care. The studies within this thesis have aimed to focus on determining the optimal approach to achieve this and whether this may be cost effective. Methods The first part of the thesis was to understand and document the current evidence on liver disease diagnosed by a non-invasive test in a community setting focusing on those patients at risk of non-alcoholic fatty liver disease and alcoholic liver disease. In addition to the prevalence of disease within the studied populations the aim was to identify which non-invasive tests for liver fibrosis have been used to stratify patients at risk and to evaluate the difference between unselected and targeted populations. A systematic review of the literature was performed to address these aims. The second part of the thesis was to improve our understanding of which risk factors were associated with significant liver disease. A prospective diagnostic study using transient elastography was performed to investigate this. The focus was on a raised body mass index but the interaction with alcohol and type 2 diabetes was also investigated along with the performance of the portable transient elastography device within an overweight subgroup of the study cohort. The third part of the thesis was to gain an understanding of the cost effectiveness of implementing a new pathway for liver disease in a community setting. A health economic evaluation of this alternative approach compared with current standard care from an NHS England perspective was completed using Markov modelling to estimate long-term health and economic effects. Results From the systematic review, transient elastography and Fibrotest, were identified to be the most frequently used tests for liver fibrosis within a community setting and had their results compared against histological findings. Subsequently these were the most validated non-invasive tests in a general population setting. The prevalence of advanced liver fibrosis and cirrhosis in the general population was identified to be between 0.9%-2% and 0.1%-1.7% respectively with studies which targeted patients with risk factors for liver disease (e.g. hazardous alcohol use or type 2 diabetes) reporting a higher prevalence of disease (advanced liver fibrosis (0%-27.9%), cirrhosis (2.4%-4%)). In our prospective study, implementation of a risk stratification pathway based on the risk factors of hazardous alcohol use and/or type 2 diabetes and/or a raised body mass index saw 703 patients attend of which 82 (11.7%) had an elevated transient elastography reading consistent with clinically significant liver disease. Seventy seven percent of the cohort had a single risk factor, whilst 21.3% had a combination of two and 1.4% had all three risk factors. An elevated transient elastography reading was approximately as common in patients with just a raised body mass index (≥27.5kg/m2) (8%) as it was in patients with more recognised solitary risk factors (Type 2 diabetes, 10%; Hazardous alcohol use 3.5%). A raised body mass index in combination with other risk factors further increased the proportion of patients with an elevated transient elastography reading and therefore demonstrated that a raised body mass index as a single or combined risk factor for chronic liver disease is important. Of the patients who attended the pathway 477 had a body mass index ≥ 28kg/m2 and had a transient elastography reading with both the M and XL probes. Twenty one percent of these patients had no valid measurements with the M probe. The XL probe significantly increased the number of valid (M vs XL probe: 66.2% vs 90.2%, p