Night and rotating shifts are the schedules most consistently linked to short sleep and poor sleep quality
Across large surveys of real-world schedules, the pattern is fairly clear: sleep is shortest and worst around night shifts. In a survey of more than 6,000 workers across eight European countries, sleep duration was shortest (about 6 hours on average) and sleep quality poorest between night shifts, compared with day and evening shifts [4]. Workers who spent more than 90% of their time on night shifts, or who rotated between shifts, were also more likely to report chronic insomnia than day workers [4]. That survey relied entirely on self-report, so it describes how workers experience their sleep rather than what a sleep lab would record.
Smaller studies that combined self-report with objective testing point in the same direction. In healthcare workers, certain schedule patterns were identified as higher risk because they limited sleep, and consecutive night work was linked to impaired alertness and performance [3]. In underground miners working extended hours, researchers compared daytime and nighttime sleep using both self-report scales and objective tests of sleepiness and wakefulness [1]. The abstract does not report the direction of every result, but the design shows that the same workers can be assessed both ways within one study.
Self-report and objective tests can disagree because they measure different things
A self-report questionnaire asks a worker to rate how sleepy or how poor their sleep felt. An objective test such as the maintenance of wakefulness test asks a person to sit quietly and try to stay awake under standardized conditions, and it records whether they actually doze off [1]. These are related but distinct: someone can feel very sleepy yet still hold wakefulness in a short lab test, or feel only mildly sleepy yet perform poorly on a vigilance task. The miner study used both approaches in the same workers, which is the kind of design needed to see where the two measures line up and where they part ways [1].
Self-report can also be influenced by a worker's preferences. The miner study's abstract specifically notes the possibility of bias in self-report items arising from a preference for one shift length [1]. That is a caution about interpretation: a worker who likes a particular schedule may rate their sleep more favorably than an objective test would justify. The evidence does not establish how large that bias is, only that it is a plausible reason self-report and objective results may not match perfectly.
Changing the schedule changes sleep and sleepiness, but the evidence is strongest for limits on very long shifts
The most direct evidence on schedule changes comes from a systematic review of 11 studies with over 2,000 participants, which looked at whether adapting shift schedules improves sleep quality, sleep duration, or sleepiness [2]. The clearest result concerned very long shifts: when workweeks were capped at 16-hour daily shifts instead of having no daily limit, workers slept about 0.73 hours more per day and reported less sleepiness during shifts [2]. That finding comes from two randomized trials, and the review rated the sleepiness evidence as moderate certainty, stronger than most other comparisons in the review [2].
Other schedule changes produced weaker or mixed results. Forward rotation compared with backward rotation showed no clear difference in sleep quality or duration, though one small trial suggested forward rotation reduced sleepiness during shifts, on very low-certainty evidence [2]. Faster rotation was linked to about 0.38 hours less sleep per day, again on very low-certainty evidence, and one study suggested it reduced sleepiness during shifts [2]. Shorter shifts and more compressed workweeks showed no consistent effect on sleep quality or duration, and the review rated that evidence as very low certainty [2]. Importantly, every outcome in this review was self-reported; it did not compare self-report against objective measures [2].
Whether a worker likes their schedule may shape both their sleep and how they report it
One study asked whether dissatisfaction with a shift system relates to altered daily patterns of sleep and wakefulness, following workers through an entire shift cycle and comparing self-reported measures with objective sleep metrics [5]. The abstract does not state the outcome, but the question itself is useful: two workers on the same schedule may sleep differently, and report differently, depending on how well that schedule suits them. This is a reminder that schedule pattern is not the only variable. The same night shift can be tolerable for one person and disruptive for another, and self-report may capture that personal dimension in a way a single objective test cannot [5].
Taken together, the evidence supports a practical conclusion: night and rotating schedules are consistently associated with shorter, poorer sleep and more sleepiness, especially when measured by self-report in large surveys [4]. Objective tests confirm that alertness can be impaired during consecutive night work [3], and studies that use both methods in the same workers offer the best chance of understanding where the two views diverge [1]. But the direct comparisons remain limited, and the systematic review of schedule changes relied entirely on self-report, so it cannot tell us whether objective alertness would improve in the same way that reported sleepiness does [2].
Sources used in this answer
Subjective and objective evaluation of sleep and performance in daytime versus nighttime sleep in extended-hours shift-workers at an underground mine
In underground miners on extended-hours shifts, sleep and performance were assessed with both self-report scales and objective tests of sleepiness and wakefulness, and the authors noted that self-report items may be biased by preference for a particular shift length.
Adapting shift work schedules for sleep quality, sleep duration, and sleepiness in shift workers
A systematic review of 11 studies found that limiting daily shifts to 16 hours increased self-reported sleep duration by about 0.73 hours per day and reduced self-reported sleepiness, while other schedule changes such as forward rotation, faster rotation, shorter shifts, and compressed workweeks had uncertain or no clear effects on self-reported sleep outcomes.
The impact of shift work on sleep, alertness and performance in healthcare workers
In healthcare workers, higher-risk schedule patterns were associated with limited sleep, and consecutive night work was linked to impaired alertness and performance using both objective and self-reported measures.
Shift work schedule patterns and sleep outcomes: results from the SHIFT2HEALTH survey across eight European countries.
In a survey of over 6,000 European workers, self-reported sleep duration was shortest and sleep quality poorest between night shifts, and rotating shifts and predominantly night work were associated with higher odds of chronic insomnia compared with day work.
Tolerance to shift work—how does it relate to sleep and wakefulness?
A study examined whether dissatisfaction with a shift schedule was related to altered daily patterns of sleep and wakefulness across a full shift cycle, comparing self-reported measures with objective sleep metrics.