Statistical analyses

The following information is reproduced from our previous work [7]. The KNHANES participants were not randomly sampled. The survey was designed using a complex, stratified, multistage probability sampling model; thus, individual participants were not equally representative of the Korean population. To obtain representative prevalence rates from the dataset, it was necessary to consider the power of each participant (sample weight) as representative of the Korean population. After approval was obtained from the Korea Centers for Disease Control and Prevention, we received a survey dataset that included information about the survey location; strata by age, sex, and various other factors; and the sample weight for each participant. The survey sample weights, calculated while considering the sampling rate, response rate, and age/sex proportions of the reference population (2005 Korean National Census Registry), were used in all analyses to provide representative estimates of the non-institutionalized Korean civilian population.

Statistical analyses were performed using SPSS software (ver. 21.0 for Windows; SPSS, Chicago, IL, USA). To compare the weighted prevalence of anemia according to age and sex, the chi-square test was performed. We compared age- and sex-adjusted clinical characteristics according to the presence of anemia for each sex using a general linear model (GLM). Weighted and adjusted hemoglobin level, weighted and adjusted prevalence of anemia, and weighted and adjusted prevalence of IDA and non-IDA were compared among the 5 eGFR groups using the GLM. All tests were two-sided, and P-values of < .05 were considered to indicate statistical significance.