Trends in mortality in patients hospitalized with cirrhosis from 2004-2013

Aylin Tansel, The University of Texas School of Public Health


Cirrhosis is a global health problem and is associated with significant morbidity and mortality. This study is a retrospective cohort study using the Veterans Administration (VA) Corporate Data Warehouse that assessed trends in in-hospital and post-discharge mortality rates in a national cohort of veterans with cirrhosis from 2004-2013. From the VA Vital Status File, we ascertained in-hospital mortality, 30-day post-discharge mortality, and one-year post-discharge mortality. We also examined whether these time trends vary by race and cirrhosis etiology. We identified 109,358 patients (mean age 62.1±10.1 years, 97% men). A total of 10,223 (9.3%) patients died during index hospitalization. Age standardized in-hospital mortality declined from 11.5% in 2004 to 7.7% in 2013 (Δ, 4.4 percentage points; 95% CI, 3.9-4.9). Of the patients surviving hospitalization, 32.8% died within one-year of discharge. There was a small decline in age standardized one-year mortality from 34.6% in 2004 to 30.3% in 2013 (Δ, 4.3 percentage points; 95% CI, 3.2-5.4). Older age, history of smoking, prior hospitalization, lower albumin, and higher model for end-stage liver disease score (MELD) were associated with increased risk of both in-hospital and post-discharge mortality. After adjusting for these factors, the odds of in-hospital mortality were 19% lower in 2013 compared to 2004 (adjusted OR=0.81, 95% CI 0.73-0.90). The length of hospitalization significantly decreased from 2004 to 2013 (median length from 6 to 4 days, respectively). The post-discharge mortality rates in the first 30 days increased 15% in 2013 compared to 2004 (adjusted OR= 1.15, 95% CI 1.04-1.27). However, when the follow up period was extended to one-year post discharge that increase in mortality was no longer observed (2013 vs. 2004, adjusted OR=0.93, 95% CI 0.87-0.99). This suggests in-hospital mortality has shifted to the immediate post-discharge period and further studies are needed to evaluate the factors that influence this risk. Additional analysis was done for race and etiology of cirrhosis. No statistically significant racial disparities were noted in crude and adjusted mortality rates between Caucasians and African Americans during the study period. Patients with HCV cirrhosis appeared to have both increasing 30-day and one-year mortality risk over the ten-year study period, which may reflect aging of the HCV birth cohort. Patients with other etiologies of cirrhosis had an increase in 30-day mortality risk and a decrease in one-year post-discharge mortality risk over the study period.

Subject Area


Recommended Citation

Tansel, Aylin, "Trends in mortality in patients hospitalized with cirrhosis from 2004-2013" (2016). Texas Medical Center Dissertations (via ProQuest). AAI10250127.