0710 REPRODUCIBILITY OF CIRCADIAN VARIABLES (MEQ, MSF, DLMO, PHASE ANGLE, AND CIRCADIAN PERIOD) OVER MONTHS

Abstract

We conducted two studies that were similar for the first 10 days (nature.com/articles/srep08381, nature.com/articles/srep36716), and determined each subject’s morningness-eveningness score (MEQ), Mid-sleep on Free Days (MSF) from the MCTQ, baseline dim light melatonin onset (DLMO), phase angle of entrainment, and free-running circadian period. Ten African-Americans (6 women, 4 men) and 8 European-Americans (2 women, 6 men) participated in both studies separated by 9 to 33 months (mean ± SD = 16 ± 7). They were 32.2 ± 6.7 years old at the first study. The purpose of this report is to examine the reproducibility of these circadian variables. Subjects slept in the lab on a fixed 8-h sleep schedule similar to their usual sleep schedule for 4 days, followed by a circadian phase assessment with 30 min saliva samples to calculate baseline DLMO. Phase angle was the interval from DLMO to bedtime. There were also 3 days of ultradian LD cycles producing forced desynchrony. Circadian period was determined from phase assessments before and after the days of ultradian LD cycles. For each circadian variable, we made scatter plots with identical x and y axes. Lines of unity showed when the variable was exactly the same in both studies. Lines parallel to and on each side of the line of unit showed how much the variable differed between the studies. We also calculated Pearson correlations for each variable. The MEQ score differed by less than 10 points between the two studies; MSF by 1 h or less, except for 2 subjects, baseline DLMO by 1 h or less except for 3 subjects, phase angle by 2 h or less, and circadian period by 0.3 h or less except for 2 subjects. All correlations were significant (p≤.0001), MEQ r=.85, MSF r=.78, baseline DLMO r=.81, phase angle r=.80, circadian period r=.78. A longer time between the two studies did not make the variables more different. In this small sample, there were no differences between men and women or between European and African-Americans in the stability of these variables. Circadian parameters were relatively stable over months. NIHR01NR007677.