Matsumoto M, Miyagishi T, Sack R L, Hughes R J, Blood M L, Lewy A J
Department of Psychiatry and Neurology, Asahikawa Medical College, Nishikagura, Japan.
Psychiatry Clin Neurosci. 1998 Apr;52(2):160-1. doi: 10.1111/j.1440-1819.1998.tb01005.x.
This study evaluated the Actillume instrument and the modified Action 3 sleep-wake scoring algorithm, in which the scoring factor (P) was set at 0.10, 0.14, 0.20, 0.30, 0.40 and 0.50. Fifteen subjects, each of whom underwent polysomnography with simultaneous wrist actigraphy four times, yielded a total of 60 sleep studies. The sleep data from each subject were divided into four groups. In the high sleep efficiency index groups of the calibration and validation samples, the accuracy of the algorithm significantly differed within six P-values and was highest at P=0.14. In the low sleep efficiency index groups of both samples, however, there were no significant differences in the accuracy. Thus, these results indicate that P=0.14 should be most appropriate for this actigraph and algorithm.
