Abstract
Serum albumin on admission has been investigated among inpatients with COVID-19. However, studies on the lowest level of serum albumin during hospitalization and adverse outcomes are limited. This research aimed to explore association between them in older adults with COVID-19. A retrospective study was conducted with 300 patients aged 60 or older with first confirmed COVID-19 from January to February 2023. An adverse outcome was defined as development of acute respiratory failure, shock, or death. Data on demographics, comorbidities, laboratory parameters, the initial phase of COVID-19, coinfection, sepsis, receipt of antiviral treatment and outcomes were gathered from the electronic medical records. The association between the lowest level of serum albumin and adverse outcomes was analyzed using univariate and multivariate regression models, along with generalized additive models. After adjusting potential confounders, nonlinear relationship with an inflection point of 29.1 g/L was detected between the lowest level of serum albumin and adverse outcomes in the elderly. The effect sizes and the confidence intervals on the left and right sides of the inflection point were 0.667 (0.520, 0.856) and 1.171 (0.875, 1.568), respectively. This demonstrated that the lowest level of serum albumin was negatively correlated with adverse outcomes when albumin was <29.1 g/L. A rise of 1 unit in the lowest level of albumin equated to a 33.3% decrease in the risk of adverse outcomes. The correlation between the lowest level of serum albumin and adverse outcomes of COVID-19 is a nonlinear. this study indicates that serum albumin levels should be sustained above the critical inflection point identified to reduce the risk of adverse outcomes.