#541 Secular trends in CKD and ESKD mortality rates in the United States, 2011–2021

Abstract

Abstract Background and Aims Patients with chronic kidney disease (CKD) and end-stage kidney disease (ESKD) have a considerably higher mortality than the general population. Nevertheless, in the United States, mortality rates in the CKD and ESKD populations have declined steadily over the last decade until the onset of the COVID-19 pandemic [1]. Through a quantitative analysis of the age- and time-dependence of mortality rates in renal patient populations in the US, we aimed at quantifying trends in life expectancy between 2011 and 2021. Method We analysed CKD and ESKD mortality rates as reported in the US Renal Data System's (USRDS) Annual Data Report 2023 for the years 2011–2021 [1]. There, CKD mortality rates were determined from a Medicare 5% random sample (CKD stages 4 and 5) for patients aged 65 and above (“CKD 65+”). ESKD mortality rates were obtained from the USRDS database; we analysed data for patients aged 45 and above (“ESKD 45+”). We used fits of analytical functions to the data to hosphorzed their dependence on age and time. This technique enables us to extract quantitative trends in time by tracking fit parameters. Results We found that for each year between 2011 and 2021, mortality rates in the adult CKD and ESKD populations in the US largely conformed to an exponential dependence on patient age, $m = {{m}_0}( t ){{2}^{a/{{\Delta }}a}}$, that is, mortality rates doubled over constant age increments ${{\Delta }}a$. While overall, mortality rates were declining between 2011 and 2019, the age increments over which mortality rates doubled were nearly identical for each year (CKD 65+: ${{\Delta }}a$ = 10.5 ± 0.4 y, ESKD 45+: ${{\Delta }}a$ = 15.1 ± 0.9 y; mean ± SD). Exploiting this simple relationship, we found that the changes in mortality rates over time could largely be reduced to a steady right-shift of an otherwise unchanged mortality curve over time (shown for the ESKD 45+ group in Fig. 1a). By quantifying this right-shift over time, we estimated the gain in 1-year survival probability for different ages (Fig. 1b,c). This gain in survival does not necessarily imply a prolonged time on renal replacement therapy (RRT), e.g., when patients concomitantly start RRT at a later age. The onset of the COVID-19 pandemic in 2020 coincides with an abrupt cessation of these positive trends, reflected by a setback of mortality curves close to those in the year 2016 for the CKD population and to before 2011 for the ESKD population. Conclusion Public health data on CKD and ESKD show systematic dependencies of mortality rates on age and time, which can be revealed using analytical parameterisations of the data. Using this technique, we can capture trends in mortality rates and 1-year survival probability in adult kidney patient populations in the US during the years 2011–2019 and can quantify the abrupt cessation of these trends during the onset of the COVID-19 pandemic. In the CKD population we observed a setback of mortality rates close to those in the year 2016 with the onset of the pandemic. Changes in mortality rates in the ESKD population were more distinct. Specifically in elderly patients (aged 65 and above) 1-year survival dropped abruptly to levels lower than those observed in the previous decade.