Skip to main content
. 2022 Feb 3;17(2):e0263069. doi: 10.1371/journal.pone.0263069

Table 3. 25(OH)D level stratified by COVID-19 disease severity category before and after cosinor correction models for the annual mean as well as each patient’s date of hospitalization.

No. (%) p value 2-sided (between groups) p value 2-sided (severe/critical vs. mild/moderate disease)
Characteristics COVID-19 disease severity category (WHO)
Mild Moderate Severe Critical
Pre-infection 25(OH)D level (ng/mL) evaluated 14–730 days before positive COVID-19 test
median 34.2 18.9 11.3 10.0
<20 7 (8.8) 50 (58.1) 64 (86.5) 12 (92.3) p < 0.001 § p < 0.001
20–29.9 8 (10) 24 (27.9) 3 (4.1) 1 (7.7)
30–39.9 37 (46.2) 3 (3.5) 4 (5.4) 0 (0.0)
≥40 28 (35.0) 9 (10.5) 3 (4.0) 0 (0.0)
Predicted 25(OH)D level (ng/mL) annual mean correction based on the cosinor model [Eq 1]
median 36.2 18.1 11.1 10.9
<20 7 (8.8) 46 (53.5) 64 (86.5) 11 (84.6) p < 0.001 § p < 0.001
20–29.9 11 (13.8) 26 (30.2) 3 (4.1) 2 (15.4)
30–39.9 35 (43.8) 5 (5.8) 4 (5.4) 0 (0.0)
≥40 27 (33.8) 9 (10.5) 3 (4.1) 0 (0.0)
Predicted 25(OH)D level (ng/mL) corrected to the hospitalization date based on the cosinor model [Eq 3]
median 36.1 19.4 13.0 11.8
<20 6 (7.5) 52 (60.5) 63 (85.1) 11 (84.6) p < 0.001 § p < 0.001
20–29.9 12 (15) 18 (20.9) 4 (5.4) 2 (15.4
30–39.9 33 (41.3) 8 (9.3) 4 (5.4) 0 (0.0)
≥40 29 (36.3) 8 (9.3) 3 (4.1) 0 (0.0)

§ P values were determined using Kruskal-Wallis Test.

† P values were determined using Mann-Whitney Test.