JAMA Netw Open. 2026 Jul 1;9(7):e2622711. doi: 10.1001/jamanetworkopen.2026.22711.
ABSTRACT
IMPORTANCE: Evidence on sex differences in the natural history of cirrhosis and cause-specific mortality remains limited.
OBJECTIVE: To compare long-term all-cause and cause-specific mortality between male and female patients with cirrhosis.
DESIGN, SETTING, AND PARTICIPANTS: This population-based retrospective cohort study included hospitalized adults with cirrhosis identified from the California Department of Healthcare Access and Information database from January 2005 to December 2019. Propensity score matching (PSM) was used to balance baseline characteristics between male and female patients. Data were analyzed from August to December 2025.
EXPOSURE: Sex (females compared with males).
MAIN OUTCOMES AND MEASURES: Ten-year all-cause and cause-specific (liver-related [hepatocellular carcinoma (HCC) and non-HCC], non-liver-related [cardiovascular and other]) mortality were assessed. Hazard ratios (HRs) of all-cause death were estimated using Cox models, and subdistribution hazard ratios (sHRs) of cause-specific death and receipt of liver transplantation (LT) were estimated using Fine-Gray competing risk models, stratified by age, race, ethnicity, social deprivation level, cirrhosis cause, and cirrhosis severity.
RESULTS: Among 316 957 patients with cirrhosis (124 228 females [39.2%]; mean [SD] age, 61.4 [14.4] years; 192 729 males [60.8%]; mean [SD] age, 58.8 [13.1] years), PSM yielded 110 044 male-female pairs with balanced characteristics. Compared with males, females had lower estimated 10-year cumulative all-cause mortality (51 703 patients [58.2%] vs 53 310 patients [59.8%]), cardiovascular mortality (9215 patients [10.5%] vs 10 531 patients [11.9%]), and HCC-related mortality (3255 patients [3.6%] vs 4745 patients [5.4%]), but higher non-HCC liver-related mortality (18 572 patients [20.9%] vs 16 993 patients [19.0%]). In Cox models, females had a lower hazard of all-cause death (HR, 0.94; 95% CI, 0.93-0.96). In Fine-Gray models, females had lower hazards of cardiovascular-related (sHR, 0.86; 95% CI, 0.84-0.89) and HCC-related (sHR, 0.68; 95% CI, 0.65-0.71) death, but a higher hazard of non-HCC liver-related death (sHR, 1.09; 95% CI, 1.07-1.12). Stratified analyses suggested that females younger than 50 years (sHR, 1.06; 95% CI, 1.02-1.11) and those with alcohol-associated cirrhosis (sHR, 1.12; 95% CI, 1.08-1.17) had higher hazard of liver-related mortality, and females were less likely to undergo LT (sHR, 0.77; 95% CI, 0.66-0.89).
CONCLUSIONS AND RELEVANCE: In this cohort study of patients with cirrhosis, males had slightly higher all-cause mortality largely due to higher cardiovascular and HCC-related mortality, while females had higher non-HCC liver-related mortality; females younger than 50 years and those with alcohol-associated cirrhosis had higher liver-related mortality but lower rates of LT. These sex disparities warrant specific consideration in future guidelines for disease monitoring and transplant allocation.
PMID:42424078 | DOI:10.1001/jamanetworkopen.2026.22711