Supplementary data: Association of liver biomarker values beyond current thresholds and negative clinical outcomes in primary biliary cholangitis: analysis of a real-world healthcare claims database

<p dir="ltr"><b>These are peer-reviewed supplementary materials for the article</b><b> </b><b>'</b><b>Association of liver biomarker values beyond current thresholds and negative clinical outcomes in primary biliary cholangitis: analysis of a real-world healthcare claims database</b><b>'</b><b> </b><b>published in the</b><b> </b><b><i>Journal of Comparative Effectiveness Research</i></b><b>.</b></p><ul><li><b>Supplemental Table 1: </b>Diagnosis, Procedure, and Drug Codes</li><li><b>Supplemental Table 2: </b>Original Cox Proportional Model With Individual Uncontrolled ALP and Uncontrolled Bilirubin (n = 2370)</li><li><b>Supplemental Table 3: </b>Sensitivity Analysis #1 (Cox Proportional Model Among Only Patients With a Confirmed PBC Diagnosis Based on ICD-10 Codes [n = 1532])</li><li><b>Supplemental Table 4: </b>Sensitivity Analysis #1 (Cox Proportional Model With Individual Uncontrolled ALP and Uncontrolled Bilirubin Among Only Patients With a Confirmed PBC Diagnosis Based on ICD-10 Codes [n = 1532])</li><li><b>Supplemental Table 5:</b> Sensitivity Analysis #2 (Cox Proportional Model Among Patients Without NASH or Cirrhosis [n = 1977])</li><li><b>Supplemental Table 6: </b>Sensitivity Analysis #2 (Cox Proportional Model With Individual Uncontrolled ALP and Uncontrolled Bilirubin Among Patients Without NASH or Cirrhosis [n = 1977])</li></ul><p dir="ltr"><b>Aim: </b>To assess the predictive effect of threshold deviations for multiple liver biomarkers on negative clinical outcomes, including hepatic decompensation, liver transplantation, and death in patients with primary biliary cholangitis (PBC) using longitudinal data from a large administrative claims database. <b>Materials & methods:</b> A time-dependent Cox proportional hazards model with time-dependent covariates assessed time to first occurrence of hospitalization for hepatic decompensation, liver transplantation, or death in patients with PBC in the Optum Clinformatics Data Mart™ database. Separate models analyzed proportion of time outside prespecified biomarker thresholds (defined as multiples of upper limit of normal [ULN] for alkaline phosphatase [ALP], total bilirubin [TB], aspartate aminotransferase [AST], and alanine aminotransferase [ALT]; lower limit of normal for albumin). Another model evaluated both ALP and TB with the lowest relevant threshold applied for each biomarker. <b>Results:</b> Overall, 2402 patients were included; 85.3% were female, mean age was 63.3 years, median follow-up was 2.2 years (interquartile range: 1.1–3.9 years). On average, patients had approximately five reported measurements for each biomarker evaluated. Each 10% increase in time outside thresholds was associated with a 6.5%, 9.2%, 9.9%, 7.3% and 23.3% increase in risk of negative outcomes for ALP, TB, AST, ALT and albumin, respectively. <b>Conclusion: </b>In patients with PBC, values outside predefined thresholds for biomarkers including ALP, TB, AST, ALT and albumin strongly predicted negative clinical outcomes. These findings highlight the importance of managing biomarkers beyond ALP in monitoring and the treatment of patients with PBC.</p>