Skip to main content
. 2014 Sep 11;9(9):e107380. doi: 10.1371/journal.pone.0107380

Table 2. Baseline data used in age regrouping.

Category Number of subjects BMI WHR Smoker Drinker Presence of hypertension Presence of diabetes AHI
Male (N = 1,975)
Age groups ∼40 952 (48.2) 27.18 (25.10, 29.41) 0.95 (0.92,0.99) 451 (47.4) 273 (28.7) 207 (21.7) 30 (3.2) 41.1 (17.5, 64.7)
41∼ 1023 (51.8) 26.65 (25.10, 29.04) 0.96 (0.93,0.99) 517 (50.5) 314 (30.7) 282 (27.6) 105 (10.3) 45.1 (21.2, 63.5)
Female (N = 378)
Age groups ∼44 99 (26.2) 25.24 (22.03, 28.93) 0.90 (0.86, 0.95) 10 (10.1) 12 (13.33) 19 (19.2) 8 (8.1) 19.5 (9.2, 47.7)
45∼53 85 (22.5) 25.95 (24.03, 27.68) 0.90 (0.89, 0.94) 2 (2.4) 6 (7.79) 24 (28.2) 4 (4.7) 21.3 (12.6, 38.9)
54∼ 194 (51.3) 26.13 (23.58, 28.74) 0.92 (0.9, 0.96) 2 (1.0) 8 (4.55) 47 (24.2) 16 (8.2) 26.5 (13.9, 49.6)

To explore centralized covariant tendencies between OSA and potential risk factors, the variable distribution characteristics were assessed by serial descriptive statistical approaches. The BMI and WHR values of both male and female populations displayed approximately satisfactory linear correlation trends as OSA severity increased. Although the plot of age against OSA severity was complex, further restricted cubic spline analysis revealed four (male) and three (female) age groups showing correlations between age and OSA severity. Then, the study populations were regrouped by age group.

Abbreviations: BMI, Body mass index; WHR, waist circumference/hip circumference ratio; AHI, apnea-hypopnea index. Skewed data were presented as the median (interquartile range), and categorical data were presented as the number (percentage). All ages age shown in years.