S1285 Association Between Distance to a Radiology Center and HCC Surveillance Among Patients With Cirrhosis in Florida

Abstract

Introduction: Hepatocellular carcinoma (HCC) is a leading cause of morbidity and mortality in patients with cirrhosis, and HCC survival is directly correlated with stage at diagnosis. Current guidelines recommend HCC surveillance with an abdominal ultrasound every 6 months, but unfortunately many patients do not meet this threshold.1 Our aim was to determine whether distance to a radiology center, measured in travel time, was associated with HCC surveillance rates. Methods: We included data on adult patients with cirrhosis within the OneFlorida Clinical Research Consortium from October 1, 2015- December 31, 2019. The primary outcome was a continuous measure of the percentage of time up to date with HCC surveillance (PTUDS) based on abdominal ultrasound (US), triple phase CT, and/or MRI with contrast. Travel time was calculated using ArcGIS geomodelling software as the estimated minimum travel time between the geographic centroid of patient’s zip code and the nearest American College of Radiology-accredited center for US, CT, and/or MRI. Linear regression models were fit with PTUDS as the outcome; all covariates with a p< 0.05 were included in the final multivariable model. (Figure) Results: Among 25,299 patients with cirrhosis (median follow-up=4.1 years), the median PTUDS was 10.0% (interquartile range 0-29.9%). Variables found to have a statistically significant association with PTUDS are displayed in Table. Travel time, hepatic encephalopathy at baseline, and ascites at baseline were associated with increased PTUDS. Patients with alcohol-related liver disease, nonalcoholic steatohepatitis, and cryptogenic cirrhosis had lower PTUDS compared with patients with HCV. Conclusion: Travel time to the nearest radiology center is not associated with lower HCC surveillance rates while race, etiology of liver disease, and disease severity do appear to variably influence surveillance. By establishing factors associated with currently suboptimal surveillance rates, we can create targeted interventions to improve surveillance and, ultimately, patient outcomes.Figure 1.: Median PTUDS by County Table 1. - Multivariable Linear Regression Model of Factors Associated with HCC Surveillance Variable Beta Coefficient, 95% CI P-Value Travel Time 0.0016 (0.0012-0.0020) <0.001 Liver Disease Etiology Hepatitis C Virus Reference - Hepatitis B Virus 0.0246 (0.0019-0.0473) 0.034 Wilson's Disease 0.0137 (-0.571-0.0845) 0.704 Hemochromatosis 0.0067 (-0.0220-0.0354) 0.647 a1-Antitrypsin Deficiency 0.0673 (0.0200-0.1145) 0.005 Alcohol-Related Liver Disease -0.0720 (-0.0801- -0.0639) <0.001 Primary Biliary Cholangitis -0.0037 (-0.0261- 0.0186) 0.745 Autoimmune Hepatitis -0.0193 (-0.0406- 0.0019) 0.075 Primary Sclerosing Cholangitis 0.0123 (-0.0462- 0.0708) 0.680 Nonalcoholic Steatohepatitis -0.0835 (-0.0917- -0.0752) <0.001 Unknown/Cryptogenic -0.01566 (-0.1664- -0.1467) <0.001 Race White/Caucasian Reference - American Indian/Alaskan -0.0207 (-0.0871- 0.0457) 0.541 Asian 0.0531 (0.0258- 0.0805) <0.001 Black 0.0171 (0.0082- 0.0260) <0.001 Native Hawaiian/Pacific Islander -0.0075 (-0.1109- 0.0959) 0.887 Multiple Race -0.0474 (-0.0781- -0.0168) 0.002 Refuse to answer 0.0904 (0.0212- 0.1596) 0.010 No information 0.0575 (0.0283- 0.0866) <0.001 Other 0.0050 (-0.0048- 0.0148) 0.318 Unknown -0.0591 (-0.0837- -0.0346) <0.001 Disease Severity Hepatic Encephalopathy at Baseline 0.0431 (0.0328- 0.0534) <0.001 Ascites at Baseline 0.0638 (0.0551- 0.0724) <0.001