Risk prediction of iron deficiency for plasmapheresis donors in China: Development and validation of a prediction model.

Abstract

The present study aims to evaluate the iron stores in plasmapheresis donors and develop and validate an iron deficiency (ID) risk prediction model for plasmapheresis donors with potential or existing ID. We assessed plasmapheresis donors' serum ferritin (SF) and haemoglobin (Hb) levels. The candidate factors showing significant differences in the multivariate logistic regression analysis were used to establish a risk prediction scoring system. The participants were divided into a training cohort and an internal validation cohort in a 7:3 ratio. Additional plasmapheresis donors from a different station were recruited for external validation. The SF levels in both male and female donors in the high-frequency group were significantly lower than those of new donors (male: p < 0.001; female: p = 0.008). The prevalence of ID in female regular donors with a high frequency was significantly higher than that in new donors (33.1% vs. 24.6%; odds ratio = 1.209 [95% CI: 1.035-1.412]). Donation frequency, age, Hb, body mass index and being pre-menopausal were identified as independent risk factors for ID (p < 0.05). The developed model exhibited good discrimination ability (area under the receiver operating characteristic curve >0.7) and calibration (p > 0.05) in development, internal validation cohorts and external validation cohorts. A higher donation frequency has been associated with reduced SF levels and an increased risk of ID in women. The developed ID risk prediction model demonstrates moderate discriminative power and good model fitting, suggesting its potential clinical utility.