Table 2. Variation in sleep patterns recorded in sleep logs, questionnaires, and by actigraphy.
| Overall | Weekdays | Weekend | ||||||||
| Sleep log | Questionnaire | Actigraphy | Sleep log | Questionnairea | Actigraphy | Sleep log | Questionnairea | Actigraphy | ||
| Bedtime | mean | 21:09 | 21:39 | 21:32 | 22:01 | |||||
| 95% CI | 21:01–21:18 | 21:28–21:49 | 21:21–21:42 | 21:45–22:16 | ||||||
| P | <0.001 | <0.001 | <0.001 | |||||||
| Sleep onset time | mean | 21:31 | 21:32 | 21:47 | 21:25 | 21:40 | 21:45 | 22:08 | ||
| 95% CI | 21:22–21:40 | 21:23–21:40 | 21:36–21:58 | 21:16–21:35 | 21:29–21:51 | 21:31–22:00 | 21:53–22:24 | |||
| Pb | <0.001 | ns | 0.004 | ns | <0.001 | 0.001 | ||||
| P | <0.001 | 0.001 | <0.001 | ns | <0.001 | <0.001 | ||||
| Sleep latency | mean | 0:22 | 0:08 | 0:09 | 0:08 | |||||
| 95% CI | 0:18–0:27 | 0:07–0:09 | 0:07–0:10 | 0:06–0:09 | ||||||
| P | <0.001 | <0.001 | <0.001 | |||||||
| Sleep end time | mean | 7:22 | 7:17 | 7:13 | 7:16 | 7:07 | 7:29 | 7:24 | ||
| 95% CI | 7:13–7:30 | 7:10–7:24 | 7:05–7:22 | 7:08–7:25 | 6:59–7:15 | 7:16–7:41 | 7:11–7:37 | |||
| P | <0.001 | ns | <0.001 | <0.001 | ns | ns | ||||
| Sleep period | mean | 9:49 | 9:45 | 9:26 | 9:51 | 9:26 | 9:38 | 9:15 | ||
| 95% CI | 9:40–9:56 | 9:36–9:54 | 9:18–9:34 | 9:42–10:00 | 9:17–9:35 | 9:25–9:51 | 9:02–9:29 | |||
| P | <0.001 | <0.001 | <0.001 | <0.001 | <0.001 | <0.001 | ||||
| Night wakings | mean | 0.05 | 0.23 | 3.47 | 0.04 | 3.52 | 0.09 | 3.33 | ||
| 95% CI | 0.02–0.09 | 0.11–0.36 | 2.90–4.03 | 0.01–0.07 | 2.89–4.15 | 0.01–0.16 | 2.70–3.96 | |||
| P | <0.001 | <0.001 | <0.001 | <0.001 | <0.001 | <0.001 | ||||
Times are presented as hh:mm.
aSleep patterns on the questionnaire for the entire week were also compared with weekday/weekend variables on sleep logs and actigraphy. P-values from paired t-tests are presented without correction in comparisons with actigraphic data in the same column bexcept for comparisons with actigraphic bed time.
A P-value less than 0.008 was considered to indicate statistical significance; Bonferroni correction was used to adjust for multiple comparisons for the 6 sleep variables, but not for other parameters.