Skip to main content

Table 2 Sample description for imputed genotype chip cohorts

From: Whole-genome association analyses of sleep-disordered breathing phenotypes in the NHLBI TOPMed program

Population Cohort N Age Percent female BMI Apnea-hypopnea index 3% AHI (percent < 5, 5–15, ≥ 15) Average desaturation Average SpO2 Minimum SpO2 Percent sleep under 90% SpO2
African-American CFS* 225 35.46 (20.32) 56.4 29.97 (10.09) 3.99 (10.55) 55.1, 23.1, 21.8 2.90 (1.09) 94.65 (4.01) 88.17 (9.60) 5.20 (16.01)
European-American, Australian ARIC 631 62.74 (5.72) 49.4 29.15 (5.23) 9.15 (15.02) 29.3, 37.9, 32.8 2.50 (1.73) 94.32 (2.15) 85.17 (6.17) 4.12 (11.76)
CFS* 218 37.57 (18.66) 56.9 28.76 (8.11) 3.4 (10.59) 57.8, 22.5, 19.7 2.30 (1.11) 94.09 (3.35) 88.81 (7.80) 3.26 (12.79)
CHS 365 77.44 (4.65) 64.9 27.10 (4.41) 10.50 (15.14) 25.8, 39.2, 35.1 2.63 (1.57) 94.41 (1.91) 84.87 (5.96) 3.93 (11.89)
FHS* 192 57.45 (9.68) 51.0 28.87 (5.16) 7.30 (14.38) 38.0, 31.8, 30.2 2.42 (1.51) 94.73 (1.80) 85.76 (5.46) 2.82 (8.38)
MrOS 2181 76.65 (5.60) 0.0 27.21 (3.75) 13.00 (18.00) 18.9, 36.1, 45.0 3.54 (1.48) 93.85 (1.73) 84.39 (5.88) 4.40 (9.95)
WASHS 1508 52.29 (13.71) 40.9 31.84 (7.93) 7.24 (15.37) 40.1, 31.1, 28.8 3.56 (2.00) 94.56 (2.38) 84.61 (7.86) 5.44 (13.82)
Hispanic, Latino-American HCHS, SOL 7155 46.10 (13.81) 57.8 29.68 (5.86) 2.00 (6.15) 69.1, 19.3, 11.6 N, A 96.46 (0.95) 87.06 (6.11) 0.83 (2.99)
Starr 782 52.34 (11.29) 71.9 32.15 (6.78) 10.35 (17.18) 31.5, 31.5, 37.1 N, A 94.65 (2.09) 85.78 (7.50) 2.83 (8.79)
  1. Eight studies contributed 13,257 individuals with genomic data imputed with a TOPMed Freeze 5b reference panel and objectively measured phenotypes (225 African-Americans, 5095 European-Americans, 7937 Hispanic/Latino-Americans). ARIC, CFS, CHS, FHS, and HCHS/SOL imputed genomic data reflect individuals without available sequencing in TOPMed Freeze 6. The overall sample had a mean age of 53.7 and was 46.9% female. Values are displayed as mean (SD), except for the skewed apnea-hypopnea Index, which is displayed as median (IQR). Sample size N reflects individuals with non-missing AHI and covariate values. *Family cohort