Table 4.
Associations between changes in daytime napping and global cognition among participants whose napping durations were 30–90 min at baselinea.
| Unstandardized beta Co-efficient (95% confidence interval) | |||||
|---|---|---|---|---|---|
| Type of changes in napping b | |||||
| Decreased by >0.5 h | Decreased by 0–0.5 h | No change | Increased by 0–0.5 h | Increased by >0.5 h | |
| Global cognition in wave 2b | |||||
| Model 1 | −1.587 (−2.583, −0.590)** | −1.023 (−2.223, 0.177) | Ref. | −1.164 (−2.457, 0.129) | −0.996 (−2.183, 0.191) |
| Model 2 | −1.108 (−2.072, −0.144)* | −0.688 (−1.842, 0.466) | Ref. | −0.968 (−2.209, 0.273) | −0.625 (−1.767, 0.517) |
| Model 3 | −0.769 (−1.600, 0.062) | −0.531 (−1.525, 0.463) | Ref. | −0.824 (−1.893, 0.245) | −0.253 (−1.237, 0.732) |
| Global cognition in wave 3c | |||||
| Model 1 | −0.824 (−1.767, 0.120) | −0.226 (−1.372, 0.920) | Ref. | 0.070 (−1.118, 1.257) | −0.927 (−2.021, 0.166) |
| Model 2 | −0.445 (−1.364, 0.474) | −0.022 (−1.137, 1.093) | Ref. | 0.301 (−0.849, 1.452) | −0.396 (−1.464, 0.672) |
| Model 3 | −0.128 (−0.911, 0.655) | −0.098 (−1.047, 0.852) | Ref. | 0.364 (−0.616, 1.343) | −0.019 (−0.930, 0.891) |
| Global cognition in wave 4d | |||||
| Model 1 | −1.210 (−2.298, −0.121)*** | 1.328 (−0.040, 2.696) | Ref. | 0.268 (−1.152, 1.688) | −1.952 (−3.183, −0.722)** |
| Model 2 | −0.897 (−1.952, 0.158) | 1.198 (−0.114, 2.510) | Ref. | 0.385 (−0.980, 1.751) | −1.558 (−2.738, −0.379)* |
| Model 3 | −0.147 (−1.060, 0.766) | 1.002 (−0.129, 2.134) | Ref. | 0.303 (−0.874, 1.480) | −1.123 (−2.141, −0.105)* |
Using generalized linear models (GLMs).
P < 0.05,
P < 0.01,
P < 0.001.
Model 1: Adjusted for age and gender.
Model 2: Adjusted for Model 1+ educational levels, physical activities, BMI, depression, use of tranquilizers, smoking, alcohol consumption, night sleep duration at baseline, medical history including hypertension, dyslipidemia, heart disease, and other 11 chronic diseases.
Model 3: Adjusted for Model 2 + baseline global cognition score.
Change in napping presented the variations between napping in wave 2 and napping at baseline, being divided into five groups: decreased >0.5 h/day, decrease 0–0.5 h/day, no change, increased 0–0.5 h/day, and increased >0.5 h/day, the GLMs showed the associations between this change in napping and global cognition scores in wave 2.
Change in napping presented the variations between napping in wave 3 and napping at baseline, being divided into five groups: decreased >0.5 h/day, decrease 0–0.5 h/day, no change, increased 0–0.5 h/day, and increased >0.5 h/day, the GLMs showed the associations between this change in napping and global cognition scores in wave 3.
Change in napping presented the variations between napping in wave 4 and napping at baseline, being divided into five groups: decreased >0.5 h/day, decrease 0–0.5 h/day, no change, increased 0–0.5 h/day, and increased >0.5 h/day, the GLMs showed the associations between this change in napping and global cognition scores in wave 4.