P113 Investigating the association between rifaximin prescription rates and socioeconomic factors in Northern Ireland

<h3>Background</h3> The rising incidence of liver cirrhosis (LC) presents a public health challenge in Northern Ireland. The increase is largely attributed to higher rates of alcohol-related liver disease and metabolic dysfunction-associated steatotic liver disease. Rifaximin, indicated for the secondary prevention of recurrent hepatic encephalopathy in patients with LC, has limited therapeutic applications, rendering it a viable proxy for LC prevalence. In Northern Ireland, all prescriptions issued in primary care are publicly available through a centralised, practice-level database. This study introduces a novel data-linkage methodology, integrating census data with central prescribing records, to assess the association between LC and socioeconomic disparities across the region. <h3>Methods</h3> Prescribing data from 2023 was obtained from 310 general practices across Northern Ireland, covering a total of 2,058,672 patients. To account for differences in practice size, rifaximin use was standardised as a milligram equivalent per registered patient per year (MgPPPY). High prescribing was defined as rates greater than or equal to the 75th centile. Each practice was mapped to a super data zone based on 2021 Northern Ireland Census boundaries, enabling linkage to demographic and socioeconomic variables. Logistic regression analysis was performed to estimate odds ratios (ORs) for associations between these factors and high rifaximin prescribing. Data analysis was conducted in RStudio. <h3>Results</h3> 350,023 rifaximin prescriptions were issued. 8166 (2.3%) prescriptions were for 200 mg, the traveller’s diarrhoea dose. The dose was missing for 56,334 (16%) prescriptions and this was imputed as 550 mg. The median MgPPPY was 85mg (IQR: 41–149mg). The outcome threshold was defined as ≥149 mg/patient, which was present in 69 (22%) of practices. A heatmap analysis revealed higher rates of prescribing were concentrated in and around urbanised areas. A lower prevalence of self-reported mental health conditions within a practice’s population was significantly associated with lower rifaximin prescribing rate (OR 0.05; 95% CI: 0.015–0.15, p=&lt;0.005). Similarly, a lower proportion of males was also associated with lower prescribing rates (OR 0.16; 95% CI: 0.05–0.57, p=0.004) (see figure 1). <h3>Conclusion</h3> We have identified substantial differences in geographical variation of rifaximin prescribing in Northern Ireland. Using this as a proxy for LC prevalence through a novel data linkage method, we identified significant associations between prescribing rates and key socioeconomic factors. These findings suggest that LC may disproportionately affect populations with greater social vulnerability and underline the potential of prescribing data as a surveillance tool to inform targeted public health interventions. <h3>References</h3> Armstrong PR, Ring é, MacNicholas R. A decade of rising alcoholic liver disease hospital admissions and deaths in Irish hospitals, 2007–2016: a retrospective cross-sectional analysis. <i>Eur J Gastroenterol Hepatol.</i> 2022 Jun 1;34(6):671–677. National Institute of Health and Care Excellence. Rifaximin for preventing episodes of overt hepatic encephalopathy. 2015. NICE guideline TA337.