Sleepiness and days of recovery

https://doi.org/10.1016/S1369-8478(01)00009-2Get rights and content

Abstract

In the literature recovery after work is taken for granted – one has 16 h off between work bouts and one has 36 h off each weekend. However, the situation for those working irregular work hours may be quite different because of night work, long shifts, or long sequences of working days. Strictly speaking we don't have any scientific support for theories on how recovery days should be patterned. This paper, therefore, brings together data from a series of our own studies that involve irregular work hours, with the specific purpose of looking at the recovery process. The results show that for the average normal office week worker two days of recovery are normally sufficient. For those who work long shifts in long sequences three days are needed for normalization, whereas 12 h shifts in 2–3 day sequences seem not to cause accumulated fatigue. Interestingly, fatigue/sleepiness is often at its peak during the first day of recovery – not the last day of the working week. Air crew and oil rig workers take a longer time to recover, probably because of too much adjustment of the biological clock. As a rule, long haul air crew is usually much more fatigued than short haul crew during their days off, despite the fact that long haul flying is voluntary and opted for by those who fly it. Also train drivers are affected during their days off by their irregular work hours – in particular backwards rotating schedules seem to cause accumulation of fatigue. The results suggest that one day of recovery never is sufficient, two days usually is, whereas 3–4 days are necessary after periods of severely disturbed circadian rhythmicity.

Introduction

The speed of recovery after a week of work is apparently assumed to be 1–2 days since custom and legislation in most countries prescribes two days off after five days of work. However, scientific support for this notion is difficult to find.

One might also assume that long work shifts, a higher weekly work hour load, night work, or early morning work would demand a longer recuperative time. Thus, Totterdell, Spelten, Smith, Barton, and Folkard (1995) have shown that one day off is insufficient for shift working nurses and that often three days may be required. Meijman (1981) showed that three days are required after a series of seven night shifts. On the other hand, Rosa and Colligan (1988) showed that two days of rest are sufficient for normalizing most psychological functions after a 60 h week. Fischer et al. (1993) found that two days off between a series of night shifts resulted in more naps and rest behavior during the working week than one with 4 days off. Patkai and Dahlgren (1981) showed that the satisfaction was higher with 3–5 days off and few long periods off, than a system with two days off but more frequent long periods off. Nicholson, Jackson, and Howes (1978) showed that workers on a 6 days on – 2 days off system often added days of vacation or sick leave to create a longer series of days off. Using experimental night work, Knauth, Rutenfranz, Herrmann, and Pöppel (1978) showed that it took two days for the body temperature rhythm to readjust after two night shifts, but 3–4 days after 21 night shifts. Our own studies have indicated a readjustment time of three days for three-shift workers after a sequence of night shifts (Kecklund & Åkerstedt, 1994).

The studies cited seem to suggest that frequently more than two days of recovery are necessary after irregular work hours. However, the results have been obtained rather unsystematically and with very different methods. The present paper attempts to pull together results from our own studies, using the same outcome variable throughout. This will make a comparison possible across very different work hour systems. We will draw upon data from studies of regular day work, rapidly rotating night work in train drivers and process operators, slowly rotating night work, 84 h working weeks and from studies of air crews on flights crossing many time zones.

Section snippets

The measurement method

Before analyzing, however, one needs to decide on how to define “recovery”. We suggest that for most people, “to be recovered” means feeling alert and fresh during the day. One could also conceive of additional aspects, such as a positive mood, high performance capacity and other variables, but for the present purpose we will restrict ourselves to alertness. The latter variable is in our studies measured via the “Karolinska Sleepiness Scale” (Åkerstedt & Gillberg, 1990). It ranges from 1 to 9,

The traditional day week with weekends off

The “baseline” study to use for comparison involves data from an ongoing study of working individuals on different types of regular day work (8 a.m. to 5 a.m.) for five days of day work and two days off. The data were obtained from 25 subjects in the age range 25–64 who were recruited for a study of day-to-day variation in sleep, stress, and health. Each participant carried a wrist-worn activity monitor (Actiwatch™) and filled out a sleep diary each day for 6 weeks. The latter included the KSS.

Discussion

Considering the overall impression of irregular work and subjective sleepiness it is obvious that night and early morning work are characterized by severe subjective sleepiness regardless of occupation. However, the rest between shifts within a “working week” are often sufficient if a proper night sleep is allowed – otherwise not.

On the other hand, more than one day of recovery may be required for some individuals, however, notably those with a less than optimal sleep behavior. A second day of

Trying to predict sleepiness during work and recovery

Given the clear effects of work/rest pattern on sleepiness demonstrated above, it should be possible to predict sleepiness based on the work schedule. Indeed, during the recent years there has been a rapid development of developing general models for predicting sleep and alertness in different situations. Thus, the two-process model of Borbély (1982) predicts the behavior of sleep length and sleep stages in connection with displaced sleep.

Inspired by the Borbély model we developed a

Comments

It should be emphasized that the present comparative analysis is focused on sleepiness, with particular emphasis on safety issues. Clearly, there is also a component of emotional fatigue to be considered, that is, being essentially, “fed up” with working at the same task for long periods of time, not getting sufficient time off to carry out one's own preferred activities. There is also a number of other parameters that may ultimately have to be considered when discussing recovery, for example,

References (25)

  • M.E. Jewett et al.

    Refinement of a limit cycle oscillator model of the effects of light on the human circadian pacemaker

    Journal of Theoretical Biology

    (1998)
  • L. Torsvall et al.

    Sleepiness on the job: Continuously measured EEG changes in train drivers

    Electroencephalography and Clinical Neurophysiology

    (1987)
  • Åkerstedt, T., & Folkard, S. (1994). Prediction of Intentional and Unintentional Sleep Onset. In R. D. Ogilvie & J. R....
  • T. Åkerstedt et al.

    Validation of the S and C components of the three-process model of alertness regulation

    Sleep

    (1995)
  • T. Åkerstedt et al.

    Predicting duration of sleep from the three-process model of alertness regulation

    Occupational and Environmental Medicine

    (1996)
  • T. Åkerstedt et al.

    Predicting sleep latency from the three-process model of alertness regulation

    Psychophysiology

    (1996)
  • T. Åkerstedt et al.

    The three-process model of alertness and its extension to performance, sleep latency, and sleep length

    Chronobiology International

    (1997)
  • T. Åkerstedt et al.

    Subjective and objective sleepiness in the active individual

    International Journal of Neuroscience

    (1990)
  • B. Bjorvatn et al.

    Rapid adaptation to night work at an oil platform, but slow readaptation following return home

    Journal of Occupational and Environmental Medicine

    (1998)
  • A.A. Borbély

    A two-process model of sleep regulation

    Human Neurobiology

    (1982)
  • C.A. Czeisler et al.

    Exposure to bright light and darkness to treat physiologic maladaptation to night work

    New England Journal of Medicine

    (1990)
  • F.M. Fischer et al.

    Day- and shiftworkers' leisure time

    Ergonomics

    (1993)
  • Cited by (0)

    View full text