P17 The challenges of risk stratifying patients with non-alcoholic fatty liver disease (NAFLD) in primary care: important lessons for policy makers and integrated care boards (ICBs)

<h3>Introduction</h3> Non-alcoholic fatty liver disease (NAFLD) is a major cause of cirrhosis and hepatocellular carcinoma in the UK. Prevalence is estimated at 25% and risk stratification is imperative in primary care, especially given the lengthy asymptomatic phase and emergence of disease modifying therapies. <h3>Aims and Objectives</h3> In this retrospective study, we compared the management of NAFLD at our GP practice with guidelines published by the British Society of Gastroenterology (BSG) and Integrated Care Board (ICB); to identify those at risk of advanced liver fibrosis, the BSG recommend FIB-4 calculation at the time of NAFLD diagnosis, and subsequently every three years. <h3>Methods</h3> We collected anonymised patient data from our electronic medical record system (EMIS Health) and identified all adults with a coded diagnosis of NAFLD. Demographics, ethnicity, primary language, blood results and other recent assessments were noted. <h3>Results</h3> 7148 adult patients were registered at our GP practice and 586 (8.2%) were coded as having NAFLD. Three were excluded as they left or died prior to data analysis. The median age was 58 years (range 18–91) and there was a range of ethnicities: White (56.1%), Asian/Asian British (24.9%), Black African/Caribbean/British (10.1%), mixed/multiple ethnicity (4.1%), other groups (1.9%) and no data recorded (2.9%). Cardiovascular monitoring was performed inconsistently over the previous 12 months: blood pressure measurement (74.6%), HbA1c (68.3%), BMI (46.1%) and lipid profile (25.4%). For 39.8% (232/583) of our population, appropriate blood tests were available to calculate the FIB-4 score (AST, ALT, platelet count); however, the score was only recorded in 13.4% of cases (78/583). Of these, 20 patients (25.6%) were stratified as medium or high-risk and 8 patients had an ELF test to further gauge fibrosis risk (figure 1). 16.5% (96/583) had abnormal LFTs within the past 12 months but the FIB-4 score could not be calculated due to missing data; 8.4% (49/583) of patients with abnormal LFTs did not have a FIB-4 score despite having the appropriate blood tests. In a multivariable logistic regression model, female sex (OR 1.87, 95% CI 1.30–2.69) and non-White ethnicity (OR 1.62, 1.11–2.38) were predictive of a recent liver function test (AUROC 0.64, p value &lt;0.001). <h3>Conclusion</h3> In the primary care setting, non-invasive assessment of liver fibrosis was performed less frequently than cardiovascular profiling. Moreover, the prevalence of NAFLD was lower than expected and suggests a significant proportion of undiagnosed individuals. Understanding the barriers will encourage more robust clinical pathways and linkage to specialist centres.