The purpose of this study was to examine the circadian temperature rhythm in clockwise (CW) and counter-clockwise (CCW) rapidly rotating shift schedules. Arguments against the CCW rotation of shifts are that they result in shortened sleep and promote greater disruption of circadian rhythms. The 3-week study included a week of day shifts (0800-1600) and 2 weeks of shiftwork. The CW 2-2-1 schedule rotated from two early mornings (0600-1400) to two evenings (1400-2200) to one midnight shift (2200-0600) allowing 24 hours off at each shift rotation and a 48-hour weekend. The CCW schedule rotated from two evenings to two early mornings to one midnight shifts allowing only 8 hours off at each shift rotation and an 80-hour weekend. Analysis of the 72-hr periods at the end of each workweek, including the midnight shifts and recovery periods during weeks 2 and 3 were compared to the same 72-hour period at the end of week 1 (baseline). A cosine function that fit the temperature curves by minimizing the sums of squares produced parameters that underwent analysis of covariance procedures. Significant differences were found between rotation conditions for amplitude and acrophase. An attenuation of amplitude and a delay in the acrophase was the found for the counter-clockwise condition. Features inherent in this schedule might explain these effects, particularly, the increased opportunity for "sleeping in" at the beginning of the week and an expanded (2-shift) workday at the end of the week.
