Effects of night shift on the cognitive load of physicians and urinary steroid hormone profiles – a randomized crossover trial

Mental and physical stress is common in physicians during night shifts. Neurocognitive effects of sleep deprivation as well as alterations in hormonal and metabolic parameters have previously been described. The aim of this crossover study was to evaluate the effects of night-shift work with partial sleep deprivation on steroid hormone excretion and possible associations with mood, sleep characteristics and cognitive functions in physicians.

In total, 34 physicians (mean age 42 ± 8.5 years, 76.5% male) from different departments of the General Hospital of Vienna, Austria, were randomly assigned to two conditions: a regular day shift (8 h on duty, condition 1) and a continuous day-night shift (24 h on duty, condition 2). In both conditions, physicians collected a 24 h urine sample for steroid hormone concentration analysis and further completed psychological tests, including the sleep questionnaire (SF-A), the questionnaire for mental state (MDBF) and the computer-assisted visual memory test (FVW) before and at the end of their shifts, respectively.

Although mean sleep deprivation during night shift was relatively small (~1.5 h) the impairment in participants’ mental state was high in all three dimensions (mood, vigilance and agitation, p ≤ 0.001). Sleep quality (SQ), feeling of being recovered after sleep and mental balance decreased (p ≤ 0.001), whereas mental exhaustion increased (p < 0.05). Moreover, we could show a nearly linear relationship between most of these self-rating items. Testing visual memory participants made significantly more mistakes after night shift (p = 0.011), however, mostly in incorrectly identified items and not in correctly identified ones (FVW). SQ and false identified items were negatively correlated, whereas SQ and time of reaction were positively associated. It is assumed that after night shift, a tendency exists to make faster wrong decisions. SQ did not influence correctly identified items in FVW. In contrast to previous investigations, we found that only excretion rates for pregnanetriol and androsterone/etiocholanolone ratios (p < 0.05, respectively) were slightly reduced in 24-h urine samples after night shift. A considerable stimulation of the adrenocortical axis could not be affirmed. In general, dehydroepiandrosteron (DHEA) was negatively associated with the sense of recreation after sleep and with the time of reaction and positively correlated with correctly identified items in the FVW test.

These results, on the one hand, are in line with previous findings indicating that stress and sleep deprivation suppress gonadal steroids, but, on the other hand, do not imply significant adrenocortical-axis stimulation (e.g. an increase of cortisol) during the day-night shift.