Sex difference in the near-24-hour intrinsic period of the human circadian timing system

Abstract

The circadian rhythms of melatonin and body temperature are set to an earlier hour in women than in men, even when the women and men maintain nearly identical and consistent bedtimes and wake times. Moreover, women tend to wake up earlier than men and exhibit a greater preference for morning activities than men. Although the neurobiological mechanism underlying this sex difference in circadian alignment is unknown, multiple studies in nonhuman animals have demonstrated a sex difference in circadian period that could account for such a difference in circadian alignment between women and men. Whether a sex difference in intrinsic circadian period in humans underlies the difference in circadian alignment between men and women is unknown. We analyzed precise estimates of intrinsic circadian period collected from 157 individuals (52 women, 105 men; aged 18-74 y) studied in a month-long inpatient protocol designed to minimize confounding influences on circadian period estimation. Overall, the average intrinsic period of the melatonin and temperature rhythms in this population was very close to 24 h [24.15 ± 0.2 h (24 h 9 min ± 12 min)]. We further found that the intrinsic circadian period was significantly shorter in women [24.09 ± 0.2 h (24 h 5 min ± 12 min)] than in men [24.19 ± 0.2 h (24 h 11 min ± 12 min); P < 0.01] and that a significantly greater proportion of women have intrinsic circadian periods shorter than 24.0 h (35% vs. 14%; P < 0.01). The shorter average intrinsic circadian period observed in women may have implications for understanding sex differences in habitual sleep duration and insomnia prevalence.

Fig. 1.

Histogram of circadian periods as assessed using core body temperature (Top) or melatonin (Middle) in a group of adults studied in FD protocols. (Bottom) Relationship between temperature and melatonin period for each of the 129 subjects in whom both measures were determined. The slope of the dashed line, which represents a linear regression fit through these data with a forced intercept at 0, is 0.99988 ± 0.00024.

Fig. 2.

Histogram of circadian periods as assessed using core body temperature in men (Upper) and women (Lower).

Fig. 3.

Scatter plots of temperature and melatonin periods with respect to the age of the participants. (Upper) Age (in years) vs. temperature period (in hours) for all 157 participants, with data from women in filled circles (●) and data from men in open circles (○). (Lower) Age vs. melatonin period for the subset of 129 subjects in whom melatonin data were also available, with data from women in filled triangles (▲) and data from men in open triangles (△).

Fig. 4.

Scatter plot of relationship between circadian period and phase angle of entrainment in women and men studied in FD protocols. Phase angle of entrainment refers to the interval between circadian phase (fitted core body temperature minimum or melatonin maximum projected from the entire FD dataset to the beginning of the FD segment) and wake time/lights-on. (Upper) Relationship in all 157 subjects as assessed using circadian phase of the core body temperature minimum; linear regression analysis yields a slope of 3.6 ± 0.5 h. (Lower) Relationship as determined using circadian phase of the melatonin maximum in the subset of 129 subjects in whom melatonin data were available; linear regression analysis yields a slope of 4.6 ± 0.5 h. Filled symbols represent women, and open symbols represent men.