Table 1. Sleep Characteristics Across Cohorts.
| Sleep exposure | Data manipulationa | Participants, No (%) (N = 5946)b | ||||
|---|---|---|---|---|---|---|
| ARIC (n = 1791) | CHS (n = 701) | FHS (n = 640) | MrOS (n = 2619) | SOF (n = 195) | ||
| Continuous variables | ||||||
| N1, median (IQR), % | Square root | 4.6 (2.9-7.0) | 4.4 (2.7-6.5) | 4.5 (2.6-7.2) | 5.8 (4.0-8.5) | 4.4 (2.9-6.0) |
| N2, mean (SD), % | NA | 55.6 (11.3) | 57.4 (12.6) | 55.5 (11.1) | 62.6 (9.5) | 55.8 (11.8) |
| N3, median (IQR), % | Square root | 17.9 (9.5-25.7) | 16.5 (7.8-26.1) | 18.2 (11.0-25.8) | 11.0 (3.8-16.7) | 18.1(11.4-25.9) |
| REM, mean (SD), % | NA | 20.7 (5.8) | 19.2 (6.3) | 20.7 (5.7) | 19.5 (6.4) | 19.8 (7.0) |
| Sleep maintenance efficiency, median (IQR), % | NA | 88.3 (82.0-92.6) | 85.4 (77.6-91.2) | 89.9 (83.7-93.6) | 78.5 (70.3-85.1) | 82.6 (73.0-88.3) |
| Wake after sleep onset, median (IQR), min | Natural log | 49.5 (30.5-77.5) | 62.0 (36.0-97.0) | 44.0 (27.3-73.0) | 101.0 (66.0-147.0) | 77.0 (52.0-116.0) |
| Categorical variables | ||||||
| Mild to severe OSA | Dichotomized, apnea hypopnea index <5 (reference) vs ≥5 | 800 (49.6) | 376 (61.2) | 270 (45.2) | 1674 (63.9) | 110 (56.4) |
| Moderate to severe OSA | Dichotomized, apnea hypopnea index <15 (reference) vs ≥15 | 317 (19.6) | 159 (25.9) | 101 (16.9) | 756 (28.9) | 39 (20.0) |
| Sleep time with arterial oxygen saturation <90% | Dichotomized, <1% (reference) vs ≥1% | 530 (29.6) | 274 (39.2) | 181 (28.3) | 1351 (51.6) | 80 (41.03) |
| Total sleep time | Dichotomized, ≤6 h (reference) vs >6-9 hc | 1054 (58.9) | 353 (50.4) | 424 (66.2) | 1334 (50.9) | 102 (52.3) |
Abbreviations: ARIC, Atherosclerosis Risk in Communities study; CHS, Cardiovascular Health Study; FHS, Framingham Heart Study; MrOS, Osteoporotic Fractures in Men Study; N1, stage 1 sleep; N2, stage 2 sleep; N3, stage 3 sleep; NA, not applicable; OSA, obstructive sleep apnea; REM, rapid eye movement; SOF, Study of Osteoporotic Fractures.
a
For data manipulation used in analysis, original units are presented for continuous measures.
b
Denominators varied by specific variable for each study.
c
The first sleep recording in the Sleep Heart Health Study was limited by battery life to 9 hours, meaning sleep durations greater than 9 hours could not be examined across cohorts.