Epidemiology of Alcohol-Related Hepatitis, Alcohol-Related Cirrhosis, and Non-Alcohol-Related Cirrhosis in Alberta, Canada: An Administrative Data Validation and Population-Based Study

Date
2024-07-22
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Background: Population-based studies evaluating the epidemiology of chronic liver diseases rely on accurate case definitions. This thesis aimed to develop new coding algorithms for alcohol- (AC) and non-alcohol-related cirrhosis (NAC) to use alongside a previously validated alcohol-related hepatitis (AH) coding algorithm to examine population-level AC, NAC, and AH hospitalization rate trends from 2012-2022, in Alberta, Canada. Methods: Using a randomly selected cohort of 700 admissions with diagnostic codes for alcohol- and cirrhosis-related conditions from the 2008-2022 Calgary Discharge Abstract Database (DAD), we validated (using electronic medical record review) newly developed and commonly used AC/NAC case definitions. The newly validated coding algorithms were used to select all adult AC/NAC/AH hospitalizations in the Alberta DAD from 2012-2022. All admissions were postal code matched to the 2016 Alberta Census data aggregated to the DA level. Temporal trends in annual sex/age-standardized hospitalization rates per 100,000 Alberta population were examined using Joinpoint analysis stratified by sex, age, area-level income quartile, and rural/urban residency. Results: Our new AC algorithm selecting codes for AC, alcohol-related hepatic failure, or alcohol use disorder (AUD) with a decompensated cirrhosis condition or NAC code provided higher accuracy than previous methods (AUROC 0.88 vs. 0.50-0.81, p<0.001). Our new NAC algorithm, excluding AUD codes and selecting for a NAC or a primary decompensated cirrhosis condition code provided higher accuracy than previous approaches (AUORC 0.87 vs. 0.58-0.76, p<0.001). Using these new algorithms, we determined that between 2012-2022, AC hospitalization rates increased only for urban females, those <35 years of age, and for those living in regions with the highest area-level income. AH hospitalization rates increased predominantly for people <35 years and for females. During the COVID-19 pandemic AC/AH hospitalization rates significantly increased for those from rural and low-income areas. NAC hospitalization rates were not impacted by the COVID-19 pandemic, only increasing in people ≥65 years. Conclusions: The new case definitions exhibit enhanced AC/NAC identification accuracy compared to previous methods. Over the past decade, AC/AH hospitalizations increased in younger and female populations, and NAC hospitalizations steadily increased in older populations.
Description
Keywords
Alcohol-related liver disease, Cirrhosis, Hospitalization, ICD-10, Administrative database, COVID-19
Citation
Swain, L. A. (2024). Epidemiology of alcohol-related hepatitis, alcohol-related cirrhosis, and non-alcohol-related cirrhosis in Alberta, Canada: an administrative data validation and population-based study (Master's thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca.