Objectives: Recent data suggest decreasing in-patient mortality in patients hospitalized with cirrhosis. We sought to determine if improvements in short-term outcomes for patients with cirrhosis are associated with changes in longer-term outcomes.
Methods: We examined temporal trends in 30 days and 1-year postdischarge mortality among patients hospitalized with cirrhosis at any of the 126 Veterans Administration hospitals from 2004 and 2013. We adjusted for a range of demographic, liver disease severity, and comorbidity-related factors to account for differences in patient cohorts over time.
Results: We identified 109,358 unique patients who were hospitalized with cirrhosis between 2004 and 2013. In-hospital mortality decreased from 11.4 to 7.6%, whereas 1-year mortality decreased from 34.5 to 33.2%. Over a third of out-of-hospital deaths occurred within the first 30 days after discharge; 30-day mortality increased from 9.3 to 10.1%. After adjusting for patient factors, the odds of in-hospital mortality in 2013 were 30% lower (adjusted odds ratio (OR)=0.70, 95% confidence interval (CI), 0.64-0.78), 1-year mortality were 13% lower (adjusted OR=0.87, 95% CI=0.82-0.93), whereas the 30-day mortality were 10% higher than 2004 (adjusted OR=1.10, 95% CI=0.99-1.21), although the latter did not reach statistical significance.
Conclusions: In patients admitted with cirrhosis, reduction in in-hospital mortality was associated with less marked reduction in 1-year mortality, and an unchanged, if not higher, 30-day mortality following discharge. Our data suggest that some of the burden of mortality in cirrhosis has shifted from in-hospital to the immediate postdischarge period.