highly significantp = 0.002
These trends in IS and SWS were not significant, but as per above this was due to the low TST in the first hour of night: when this hour was omitted, there was a highly significant linear decrease in both IS (slope = -4.90 s/hour of night, r 2 = 0.732, p = 0.002) and SWS (slope = −7.66 s/hour of night, r 2 = 0.753, p = 0.001).