Fig 1.
Flow charts of the study populations.
Fig 2.
Trajectories of RDW during follow-up visits.
We identified four distinct trajectories of Red Cell Distribution Width (RDW) changes during the 28-day follow-up period through group-based trajectory modeling (GBTM). The solid lines represent the average predicted levels of specific categories, serving as a function of follow-up estimated by the best-fitting model, and the shaded areas around the solid lines indicate the confidence intervals for the calculated trajectories.
Table 1.
Baseline characteristics according to RDW trajectories.
Table 2.
Cox proportional hazards analyses for 28-day all-cause mortality according to RDW trajectories.
Table 3.
Multivariable regression analysis (Cox or linear) for secondary outcomes according to RDW trajectories.
Fig 3.
Kaplan‒Meier survival analysis curves for all-cause mortality based on trajectories of the RDW.
Fig 4.
Adjusted HR (95% CIs) for 28-day all-cause mortality according to RDW trajectories in subgroups analyses*.
*Results were adjusted for age, gender, race, BMI, MAP, SOFA, APS Ⅲ, OASIS, CCI, hypertension, diabetes mellitus, heart failure, malignant tumors, COPD, cirrhosis, WBC, hemoglobin, hematocrit, PLT, albumin, sodium, potassium, total calcium, chloride, pH, lactate, total bilirubin, creatinine, and bicarbonate, and stratified variables were not included in the relevant models.