Fig 1.
Study design.
Table 1.
Demographic and physical characteristics of the sample population.
Table 2.
Blood hemoglobin concentrations of participant groups as determined by various methods.
Table 3.
Factors influencing the bias of assessed testing methods, as evaluated by univariate and multiple regressions.
Table 4.
Factors influencing the ability of the Pronto® device to generate readings.
Fig 2.
Bland-Altman plots for three hemoglobin measurement methods.
Bland-Altman plots for all measurement methods. Y axes: bias produced by each method; X axes: the average of Hb results from each method and the Sysmex KN21. The three plots depict the HemoCue wicking method bias (A), HemoCue gravity method bias (B), and the Pronto device method bias (C).
Table 5.
Sensitivity, specificity, and correction classification rates for the three methods using a cutoff of 11.5 g/dl*.