Premium
Estimating distributions of usual 24‐hour sodium excretion from timed‐spot urines in adults 18‐39 years (632.11)
Author(s) -
Wang ChiaYih,
Carriquiry Alicia,
Chen TeChing,
Loria Catherine,
Pfeiffer Christine,
Liu Kiang,
Sempos Christopher,
Perrine Cria,
Swanson Christine,
Cogswell Mary
Publication year - 2014
Publication title -
the faseb journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.709
H-Index - 277
eISSN - 1530-6860
pISSN - 0892-6638
DOI - 10.1096/fasebj.28.1_supplement.632.11
Subject(s) - percentile , urine , evening , morning , statistics , mathematics , population , zoology , medicine , chemistry , physics , biology , environmental health , astronomy
We developed and tested calibration equations from spot urine specimens to estimate the distribution of usual 24h sodium excretion (24UNa). A convenience sample of 407 adults, 18‐39 y, (54% female, 48% black) collected each void in a separate container for 24h; 133 collected a second 24h urine 4‐11 d later. We selected 4 timed voids (morning [M], afternoon [A], evening [E], overnight [O]) from each 24h collection. Participants were randomly assigned to test (n=247) or validation (n=160) groups. Total sodium excreted in each of the 4 one‐spot and 6 combined two‐spot (e.g., M+A) urines from the test groups were used with covariates to develop sex‐specific equations for predicting 24UNa using Fuller’s error‐in‐the‐equation measurement error model. Equations applied to the validation data were used to estimate usual 24UNa distributions with the Iowa State University method. Mean bias in predicting the median of 24UNa with one‐spot urines ranged from ‐69 to 1617 mg for men and ‐72 to ‐9 mg for women, and with two‐spot urines from ‐124 to 115 mg and ‐50 to 9 mg, respectively. Mean bias in predicting percentiles (5 th , 25 th , 50 th , 75 th , and 95 th ) did not differ from 0 (p=0.06‐0.90) for men or women using an O void or combined E+O urine. Our method to estimate population distribution of 24UNa, which uses timed‐spot urine specimens and accounts for day‐to‐day variation and covariance between measurement errors, merits further investigation.