Readmission Requires Survival and Access: Reframing Global Quality Metrics in Cirrhosis

1.

Idilman et al. studied 4208 patients with cirrhosis across 37 countries and reported a striking outcome pattern. High‐income countries had more 90‐day readmissions than low‐ and lower‐middle‐income countries, 62.4% versus 44.7%, and more liver transplantation, 14.4% versus 4.4%, but lower mortality, 21.1% versus 32.8% [1]. This pattern cautions against treating readmission as a universally adverse measure of care quality.

The comparison of 30‐ and 90‐day readmission needs careful interpretation. The 90‐day endpoint contains every 30‐day event, and 1170 of 1709 readmissions had already occurred by Day 30 [1]. Similar associations in the two logistic models may therefore reflect overlapping cumulative outcomes rather than shared mechanisms. To identify late first readmission, the risk set would need to begin on Day 31 among patients who remained alive, under observation, and free of earlier readmission. The current results establish similarity between cumulative associations, but not equivalence between early events and those arising during Days 31–90.

The cross‐income comparison raises a related issue. Readmission is conditional on both survival and the ability to return to hospital. Death removes that possibility, while transplantation changes the clinical state and the meaning of later hospital use. Separate binary models, based on different outcome denominators, do not retain this event order. Where mortality is higher, lower observed readmission can coexist with greater unmet need because some patients die or lose access before returning. Competing death can distort binary readmission comparisons [2], and hierarchical competing‐risk methods have already been applied to readmission in cirrhosis [3].

This distinction is clinically important. In a large cirrhosis cohort, early outpatient follow‐up was associated with slightly higher readmission but substantially lower mortality [4]. Greater hospital use may therefore accompany timely detection and treatment escalation rather than failed care. Conversely, no readmission may indicate recovery, death before return, inability to regain access, or planned end‐of‐life care. These states should not receive the same quality meaning.

The CLEARED findings do not show that more readmission is beneficial. They show that readmission rates are not directly portable across health systems with different survival, transplantation, and access. Days alive and at home integrate mortality with healthcare use and offer a more patient‐centered outcome [5]. Global benchmarking should therefore define early and late readmission using distinct risk sets and interpret hospital return together with death and transplantation.

Conflicts of Interest

The authors declare no conflicts of interest.

Linked Articles

This article is linked to Idilman et al. paper. To view this article, visit https://doi.org/10.1111/liv.70804.

Data Availability Statement

Data sharing not applicable to this article as no datasets were generated or analysed during the current study.