(2): Effects on Subjective Symptoms, Physical Activity,

Transcription

1 Industrial Health 1999, 37, Original Article Shift Work-Related Problems in 16-h Night Shim Nurses (2): Effects on Subjective Symptoms, Physical Activity, Heart Rate, and Sleep Masaya TAKAHASHII*, Hideki FUKUDA1, Keiichi MIKI1, Takashi HARATANII, Lumie KURABAYASHI1, Naomi HISANAGAI, Heihachiro ARIT01, Hideko TAKAHASHI2, Makiko EGOSHI2 and Misuzu SAKURAI2 ' National Institute of Industrial Health, Department of Nursing, Juntendo Hospital,, Nagao 6 chome, Tama-ku, Kawasaki , Japan 1-3, Hongo 3 chome, Bunkyo-ku, Tokyo , Japan Received December 10, 1998 and accepted February 24, 1999 Abstract: We compared the shift work-related problems between 16-h night shift and 8-h evening/ night shifts among nurses in a university hospital with respect to subjective symptoms, physical activity, heart rate (HR), and sleep. The nurses of one group (n=20) worked a 16-h night shift under a rotating two-shift system, while those of the other group (n=20) worked an 8-h evening or night shift under a rotating three-shift system. The 16-h night shift was staffed by three or four nurses who alternately took a 2-h nap during the shift, and had at least one day off after each shift. Subjective symptoms and daily behavior were measured every 30 min by the nurses before, during, after each shift as well as during days off using a time-budget method. Also, physical activity, heart rate (HR), and posture were recorded during shifts. The results showed similar or lower levels of sleepiness, difficulty in concentration, fatigue, physical activity, and HR during the 16-h shift compared to the 8-h shifts. No differences in subjective symptoms between the two shift schedules were observed before or after the shifts or during days off. The main sleep was longer after the shifts and during days off in the 16-h shift group than in the 8-h shift group. Our results suggest that the work-related problems in 16-h night shift nurses may not be excessively greater than those in 8-h evening/night shift nurses, as long as appropriate countermeasures are taken during and after the extended shift. Key words: Shift work, l6-h shift, Nurse, Fatigue, Sleep Introduction A two-shift system has been rapidly introduced as a work schedule for hospital nurses in Japan instead of the existing three-shift system (percentage of two-shift systems in Japan: 26.6% in 1993 and 43.9% in 1996)1). The main reason for this move to a two-shift system is the advantages it offers, such as lengthened days off resulting from longer intervals between shifts, fewer days of night shifts, and reduced commuting time2_4~. However, one drawback to this system *To whom correspondence should be addressed. is that night shifts are longer than 8 h. The number of hospitals in Japan that have a 16-h night shift is expected to increase significantly in the future, as legislation has recently been passed allowing night shifts as long as 16 h to be implemented in national hospitals5~. Although previous studies in 12-h shift nurses have had contrasting findings, they have nonetheless revealed a number of negative consequences associated with increased shift length, including an increase in the number of health and fatigue complaints6' 7) and decreased quality and quantity of nursing care8-10~. In a related finding, studies involving industrial workers on 12- and 14-h night shifts have showed

2 NURSES' STRESS ASSOCIATED WITH 16-H NIGHT SHIFT 229 Table 1. Age and shift work experience of nurses by ward under two- and three-shift systems an increased level of fatigue and sleepiness, and poor performance, especially during the final few hours of the shift1-14~. Hence, the 16-h night shift may exacerbate shift work-related problems in the nurses working it15~ In the present study, we compared the 16-h night shift with the 8-h evening and night shifts among nurses in a university hospital with respect to subjective symptoms, physical activity, heart rate, and sleep. This hospital has taken several measures for minimizing the potential disadvantages of the 16-h night shift. Such measures have included 1) increasing the number of nurses working each 16-h shift, 2) allowing the nurses a 2-h nap period during each 16-h shift, 3) reducing nursing duties during shift; for example, reducing the number of written records when appropriate, and 4) scheduling at least one day off after each 16-h shift. Particular attention was given to the following issues: differences in shift work-related problems between wards'' 16), time courses of the subjective and objective parameters, and sleep"' 'g). Parts of the present results have been reported previously19>, Methods Participants The present study was conducted in a private university hospital located in an urban area of Tokyo, Japan. The numbers of nurses working under rotating two- and threeshift systems were 53 and 62, respectively. The nurses were chosen to be participants if they were in twenties, worked shift for two or more years, were single, and lived alone in a dormitory or an apartment. Thirty and twenty-seven nurses met these selection criteria under the two- and three-shift systems, respectively. Then, 20 nurses were randomly selected from each group (Table 1). The wards studied included one surgical and two mixed wards under the twoshift system, while two surgical, one internal, and one mixed wards under the three-shift system. All the participants were given explanations concerning the details of this research and recording methods for subjective symptoms and physiological measures prior to the start of the study, and gave written informed consent. The study protocol was approved by the Ethic Committee of our institute. Shift schedules Under the two-shift system, the nurses worked three 8-h day shifts (8:00-16:10) and one 16-h night shift (15:50-8:10) followed by at least one day off a week. Three nurses worked each 16-h night shift on the surgical and one mixed wards, and four nurses on the other mixed ward. The 16-h night shift nurses were allowed to alternately take a 2-h nap between 22:00 and 6:00 on a bed prepared in a resting room within a nurse station (Fig. 1). The mean numbers of the 8- h day shifts, l6-h night shifts, and days off were 11.3 ± 2.2 SD, 4.9 ± 2.2, and 9.3 ± 0.9 times in one month during the present study, respectively. The two-shift system was introduced four to nine months before the study. Under the three-shift system, the nurses in principle worked a day shift (8:00-16:10) followed by one day off, one to two night shifts (23:30-8:10), one to two evening shifts (15:50-0:00), and one day off, though the shift schedules varied between the

3 230 M TAKAHASHI et al. catecholamines, and cortisol in both urine and saliva. on these measures will be reported elsewhere. Results Fig. 1. The number of nurses working 16-h night, 8-h evening and night shifts. four wards. Two nurses worked each 8-h night shift and were not allowed to take a nap. One or more additional nurses worked during the evening shift and the end of the night shift (Fig. 1). The mean numbers of the 8-h day, evening, night shifts, and days off were 11.5 ± 1.7, 4.9 ± 0.9, 4.1 ± 1.3, and 10.4 ± 2.2 times in one month during the present study, respectively. Study design Based on the results of a pilot study made in the same hospital, and as a result of other more practical considerations such as a necessity of completing the data collection by April 1997, we designed to investigate only the first shift of 16-h night, the first shifts of 8-h evening and night, and the day off under each schedule. By design, each day of data collection was preceded by three days of day shifts and/or days off in order to minimize any possible carry-over effects of a prior night shift. Measures The methods used for data collection in this study have been described in detail in a companion article20~. Briefly, subjective symptoms and daily behavior were measured every 30 min and recorded by the nurses on a self assessment form ("Check Sheet") using a time-budget technique. The measurement was made during two (for the 8-h shifts) or three (for the 16-h shift) consecutive days starting at 0:00 on the day of each shift. The same measurement was made during day off for each schedule. Additionally, physical activity, HR, and posture were continuously recorded every 30 sec during each shift with an ambulatory monitor worn on the nurses' waist (ACTIVTRACER AC-300, GMS, Japan). The following information was also collected: subjective fatigue feeling21~, subjective assessment of sleep quality22~, morningness-eveningness type23~, urinary Data analysis Data for the 16-h night, 8-h evening and night shifts were obtained from 20, 20 and 19 nurses, respectively. Data during days off under the 16-h and 8-h shift schedules were obtained from 19 and 17 nurses, respectively. The subjective symptoms, daily behavior, physical activity, and HR were analyzed using the 4-h averaged values. This was done to prevent any missing values from the 2-h nap during the 16-h shift from affecting the results and to obtain data from equally segmented periods of time for analysis of variance (ANOVA). For each participant, the percentages of time with each subjective symptom and time spent in each daily behavior within each 4-h period were calculated before, during, after the shifts and during days off. Then, the resultant percentages were averaged over the participants. Thirty-sec values of physical activity and HR during shifts were averaged for each 4-h period, while those data recorded during napping on the 16-h shift were excluded from the calculation of an average score. We examined whether the subjective symptoms, physical activity, and HR recorded during the shifts differed between the surgical, internal, and mixed wards by the Kruskal-Wallis test. The results indicated that those measures were generally higher on the surgical wards than on the internal or mixed wards. Therefore, the subjective symptoms, physical activity, and HR recorded during the shifts were compared between the 16-h and 8-h shifts on the surgical or on the internal/ mixed wards separately. The timing and length of sleep and naps were examined on each shift day and day off. Napping was defined according to the criteria of Rosa24~; that is, any extra period of sleep lasting over 30 min with a time interval of more than 1 h from the main (i.e. longest) sleep period. Statistical analysis The subjective symptom data for each 4-h period during shifts were compared between the 16-h and 8-h shifts on all wards by the Mann-Whitney U-test. The data also were compared between the two schedules on the surgical wards or on the internal/mixed wards separately by the Mann- Whitney U-test. The 4-h averaged physical activity and HR during shifts were examined on all wards by two-way repeated measures ANOVA with the Greenhouse-Geisser correction25~. For the ANOVA, shift schedule (16-h, 8-h) was a between-subject factor, and time of day (16:00-20:00, Industrial Health 1999, 37,

4 NURSES' STRESS ASSOCIATED WITH 16-H NIGHT SHIFT 231 Table 2. Average percentages of time with each subjective symptom in each 4-h period during shifts on all wards and on surgical and internal/mixed wards 20:00-0:00, 0:00-4:00, 4:00-8:00) was a within-subject factor. Also, the physical activity and HR data were analyzed on the surgical wards or on the internal/mixed wards separately by the shift schedule x time of day ANOVA. The subjective symptom data before and after shifts and during days off were compared between the two schedules on all wards by the Mann-Whitney U-test. The sleep/nap data were compared between the two schedules on all wards by unpaired t-test. All statistical analyses were performed using SAS software (Release 6.12, SAS Institute, USA). This paper reports the results of 5 subjective measures (sleepiness, difficulty in concentration, fatigue, physical dullness, and busy at work), 2 objective measures (physical activity and HR), and sleep/nap data. between 0:00 and 4:00, and was significantly higher than during the 8-h night shift (p<0.05). Yet, those symptoms lessened subsequently. As a result, their peak levels during the 16-h shift were almost equal or lower than during the 8- h evening or night shift. Physical dullness increased with the elapsed time of each shift, and thereby was significantly higher between 0:00 and 4:00 during the 16-h shift than during the 8-h night shift (p<0.05). Overall, the number of times the nurses reported being busy at work was fewer during the 16-h shift than during the 8-h shifts. Similar results to those mentioned above were obtained from both the surgical and on the internal/mixed wards, but the differences between the 16-h and 8-h shifts were not significant because of a high level of variances in the scores (Table 2). Results 1. Subjective symptoms during shifts On the subjective symptoms variables collapsed across all wards under each schedule (Table 2), sleepiness, difficulty in concentration, and fatigue during the 16-h shift increased 2. Physical activity during shifts Significant reductions in physical activity were observed during the 16-h shift compared to the 8-h shifts on all wards, consistently between 0:00 and 8:00 (the effect of shift schedule: F=17.38, df=1,37, p<0.01; Fig. 2). The effect of time of day was significant (F=30.03, df=3,111, p<0.01),

5 232 M TAKAHASHI et al. Fig. 2. Time courses for physical activity during both 16-h night ( 0 ) and 8-h evening surgical and internal/mixed wards. Values are means and SD. ) and night A ) shifts on all wards, on Fig. 3. Time courses for heart rate during both 16-h night (0) and 8-h evening ( A) and night ( A) shifts on all wards, on surgical and internal/mixed wards. Values are means and SD. but the interaction between the two factors was not statistically significant. These results were confirmed by a separate ANOVA with two factors of shift schedule and time of day with two levels ([16:00-20:00, 20:00-0:00] or [0:00-4:00, 4:00-8:00]). In either the surgical ward (F=9.34, df=1,12, p<0.01) or the internal/mixed wards (F=4.23, df=1,23, p=0.051), lowered physical activity was observed during the 16-h shift than during the 8-h shifts, especially between 0:00 and 8:00 (Fig. 2). 3. HR during shifts The HR was significantly lower during the 16-h shift than during the 8-h shifts on all wards (the effect of shift schedule: F=6.58, df=1,37, p<0.05; Fig. 3). A significant effect for time of day was found (F=37.67, df=3,111, p<0.01) with a non-significant interaction between the two factors. The results were supported by a separate ANOVA with two factors of shift schedule and time of day with two levels. Figure 3 shows the decreased HR during the 16-h shift compared to the 8-h shifts both on the surgical and on the internal/mixed wards, with a greater decrease apparent between 0:00 and 8:00. However, neither the effect of shift schedule nor the interaction of shift schedule by time of day was statistically significant on each ward. Only the effect of time of day Industrial Health 1999, 37,

6 NURSES' STRESS ASSOCIATED WITH 16-H NIGHT SHIFT 233 was significant on the surgical (F=16.57, df=3,36, p<0.01) and internal/mixed wards (F=17.66, df=3,69, p<0.01). The above mentioned results may make it reasonable to conclude that the differences in the dependent measures (subjective symptoms, physical activity, and HR) observed during the shifts were due to the shift schedule, rather than the different specialities of the various wards. 4. Subjective symptoms before/after shifts and during days off None of the subjective symptoms in any of each 4-h period before and after the shifts were significantly different between the 16-h and 8-h shifts. Additionally, the measures taken during days off did not differ significantly between the 16- h and 8-h shifts, except for increased sleepiness observed from 16:00 to 20:00 for the 16-h shift nurses compared to the 8-h shift nurses (10.7 ± 15.2 vs. 3.7 ± 15.2%, p<0.05). 5. Timing and length of sleep/nap on both workdays and days off Figure 4 depicts the temporal distributions of the main sleep and naps under the 16-h and 8-h shift schedules on both workdays and days off. Table 3 summarizes the data for timing and length of the main sleep and naps before, during, after shifts and during days off. A longer main sleep occurred before the 16-h shift than before the 8-h evening shift (p<0.05). A 3.2-h nap was taken before the 8-h night shift by 16 out of 19 nurses. During the 16-h shift, an average of 1.50 ± 1.30 h of nap was taken by almost all nurses. The time for the beginning of nap was distributed from 22:00 to 4:00. After the 16-h shift, 4.2-h nap was obtained from 14:00 to 18:00, and this time period coincided with the main sleep following the 8-h night shift. The main sleep following the 16-h shift started at midnight of the following day, and was significantly longer (9.13 ± 2.00 h) than that following either the 8-h evening shift (7.28 ± 1.92 h, p<0.01) or the 8-h night shift (5.90 ± 2.02 h, p<0.01). The main sleep during days off significantly increased by 1.4 h for the 16-h shift nurses compared to the 8-h shift nurses (8.70 ± 1.33 vs ± 1.77 h, p<0.05, Fig. 4). This appeared to be due to both an earlier bedtime (p=0.14) and a later wake-up time (p=0.08) for the 16-h shift nurses (Table 3). Data on the naps during days off did not significantly differ between the 16-h and 8-h shift nurses. Discussion The present results indicate that the 16-h night shift produced similar or lower levels of sleepiness, difficulty in Fig. 4. Temporal distributions of main sleep and naps on workdays and days off among 16-h and 8-h shift nurses. 16-N, 16-h night shift; 8-h E, 8-h evening shift; 8-h N, 8-h night shift. The ordinate means the percentage of nurses who took a main sleep and nap for each 30- min time period. concentration, fatigue, physical activity, and HR during the shift compared to the 8-h shifts. Although these symptoms were elevated 12 h after the start of the 16-h shift, they decreased afterwards. An analysis by wards suggested that these results were due to the shift schedule, not to the specialties of the various wards. Also, similar levels of subjective symptoms before and after the shifts or during days off were observed between the 16-h and 8-h shift nurses. The sleep duration after the shifts and during days off was longer for the 16-h shift nurses than for the 8-h shift nurses. Our findings are consistent with those of previous studies showing similar or reduced fatigue among the 12-h shift nurses compared to the 8-h shift nurses16, 26, 27), Studies involving other types of shift workers have suggested that fatigue and sleepiness may be comparable between the extended (12-14 h) and 8-h shifts28-32), In terms of the subjective symptoms and sleep associated with the 8-h shifts, our data agree with those of previous reports involving the

7 234 M TAKAHASHI et al. Table 3. Main sleep and naps taken before, during, after each shift and during days off 8-h shift nurses in Japan33-35). The lack of excessively adverse consequences in connection with the 16-h night shift may be attributed to the number of measures taken, which were outlined in the Introduction. In particular, the increased number of nurses working each 16-h night shift may be a crucial factor in that it allows the nurses to take turns napping during the shift. A minimum of two nurses works the shift between 22:00 and 6:00, and thus a nurse can take a 2-h nap throughout. Indeed, the nurses spent an average of 1.5 h in bed during the nap opportunity (Table 3). This may maximize the beneficial effects of napping36-39). We previously suggested that despite the effects of sleep inertia immediately after napping, the nap taken during the 16-h shift may prevent such symptoms as fatigue, sleepiness, and physical dullness from increasing further40~. Moreover, reduction of the nursing duties during the 16-h night shift may have been an important contributing factor. The nurses working the 16-h shift can pace their duties over the course of the shift, and in this way reduce any feelings of stress related to a lack of time. This may be related to nurses working the 16-h shift being less busy at work than those working the 8-h shifts (Table 2). Scheduled days off following the 16-h night shift may be essential for the institution of such a long shift. In the hospital studied, the 16-h shift nurses have 48 h of rest by the next day shift, whereas the 8-h shift nurses have only 16 h of rest by the second day of each shift. The longer period of non-work time resulted in 13.4 ( ) h of sleep following the 16-h night shift, which included a nap, as opposed to 5.9 h of sleep following the 8-h night shift (Table 3). The increased amount of sleep may allow the nurses to recuperate from their prior extended shift15,18, 41) Our results also showed that the main sleep during days off was longer among the 16-h shift nurses than among the 8-h shift nurses (Table 3), suggesting the added advantage of taking more sleep on their days off for the 16-h shift nurses. Alternatively, the 16-h shift nurses, in order to recover completely, seem unable to avoid taking the extra sleep during their days off. In any case, it is reasonably assumed that if the measures were not taken as in this hospital, the 16-h night shift would produce a substantial increase in occupational health and safety risk among the nurses. Besides the effects of appropriate measures, the characteristics of a given participant may have contributed Industrial Health 1999, 37,

8 NURSES' STRESS ASSOCIATED WITH 16-H NIGHT SHIFT 235 to our findings. Aging has been known to decrease the ability to adjust to night work42' 43) In addition, nurses with small children are not able to set aside enough time to get a sufficient amount of sleep or leisure compared to those without children34, 44), The fact that the 16-h shift nurses in our study had mean age of 25.2 years, were single, and lived alone may thus have made the 16-h night shift less problematic. We did not compare objective levels of sleepiness and performance using electroencephalogram and behavioral tasks or the job content between the 16-h and 8-h shift nurses. Night shifts have been shown to increase the risk of medication error and needle stick injury45' 46). Some studies have revealed poor and less patient care during the 12-h shiftss-10). Further study is thus needed to examine the 16- h shift effects on those variables. In summary, the present study shows that the work-related problems in 16-h night shift nurses may not be excessively greater than those in 8-h evening and night shift nurses. Our findings suggest the importance of the countermeasures taken for minimizing the fatigue/stress resulting from a 16-h night shift. References 1) Hospital report (1996) The Japan Ministry of Health and Welfare. 2) Tepas DI (1985) Flextime, compressed workweeks, and other alternative work schedules. In: Hours of work-temporal factors in work-scheduling. eds. by Folkard S, Monk TH,147-64, John Wiley and Sons, New York. 3) Colligan MJ, Tepas DI (1986) The stress of hours of work. Am Ind Hyg Assoc J 47, ) US Congress, Office of Technology Assessment (1991) Biological rhythms: implications for the worker. OTA- BA-463, U.S Government Printing Office, Washington, DC. 5) Notification No. 322, Hospital Service Policy Division, The Japan Ministry of Health and Welfare, October 17, ) Mills ME, Arnold B, Wood CM (1983) Core-12: a controlled study of the impact of 12-hour scheduling. Nurs Res 32, ) Iskra-Golec I, Folkard S, Marek T, Noworol C (1996) Health, well-being and burnout of ICU nurses on 12- and 8-h shifts. Work Stress 10, ) Todd C, Reid N, Robinson G (1989) The quality of nursing care on wards working eight and twelve hour shifts: a repeated measures study using the MONITOR index of quality of care. Int J Nurs Stud 26, ) Todd C, Reid N, Robinson G (1991) The impact of 12-hour nursing shifts. Nurs Times 87, ) Reid N, Robinson G, Todd C (1993) The quantity of nursing care on wards working 8- and 12-hour shifts. Int J Nurs Stud 30, ) Knauth P, Keller J, Schindele G, Totterdell P (1995) A 14-h night-shift in the control room of a fire brigade. Work Stress 9, ) Rosa RR, Bonnet MH (1993) Performance and alertness on 8-hour and 12-hour rotating shifts at a natural gas utility. Ergonomics 36, ) Rosa RR (1995) Extended workshifts and excessive fatigue. J Sleep Res 4 (suppl 2), ) Tucker P, Barton J, Folkard S (1996) Comparison of eight and 12 hour shifts: impacts on health, wellbeing, and alertness during the shift. Occup Environ Med 53, ) Knauth P (1993) The design of shift systems. Ergonomics 36, ) Fields WL, Loveridge C (1988) Critical thinking and fatigue: how do nurses on 8- & 12-hour shifts compare? Nurs Econ 6, ) Niedhammer I, Lert F, Maree M (1994) Effects of shift work on sleep among French nurses. A longitudinal study. J Occup Med 36, ) Totterdell P, Spelten E, Smith L, Barton J, Folkard S (1995) Recovery from work shifts: how long does it take? J App! Psycho! 80, ) Fukuda H, Takahashi M, Arito H. Nurses' workload associated with 16-h shifts on 2-shift system (1) comparison with 3-shift system. Psychiatry Clin Neurosci (in press). 20) Fukuda H, Takahashi M, Miki K, Haratani T, Kurabayashi L, Hisanaga N, Arito H, Takahashi H, Egoshi M, Sakurai M. Shift work-related problems in 16-h night shift nurses (1) Development of an automated data processing system for questionnaires, heart rate, physical activity and posture. Ind Health, 37, ) Industrial Fatigue Research Committee of Japan Association of Industrial Health (1970) Jpn J Ind Health 12,196-7 (in Japanese). 22) Oguri M, Shirakawa S, Azumi K (1985) Construction of standard rating scale to estimate sleep profile. Seishin Igaku 27, (in Japanese). 23) Ishihara K, Miyasita A, Inugami M, Fukuda K, Miyata Y (1987) Differences in sleep-wake habits and EEG

9 M 236 M TAKAHASHI et al. sleep variables between active morning and evening subjects. Sleep 10, ) Rosa RR (1993) Napping at home and alertness on the job in rotating shift workers. Sleep 16, ) Vasey MW, Thayer JF (1987) The continuing problem of false positives in repeated measures ANOVA in psychophysiology: a multivariate solution. Psychophysiol 24, ) Washburn MS (1991) Fatigue and critical thinking on eight-and twelve-hour shifts. Nurs Manage 22, 80A-H. 27) Gillespie A, Curzio J (1996) A comparison of a 12-hour and eight-hour shift system. Nurs Times 92, ) Peacock B, Glube R, Miller M, Clune P (1983) Police officers' responses to 8 and 12 hour shift schedules. Ergonomics 26, ) Frese M, Semmer N (1986) Shiftwork, stress, and psychosomatic complaints: a comparison between workers in different shiftwork schedules, nonshiftworkers, and former shiftworkers. Ergonomics 29, ) Williamson AM, Gower CG, Clarke BC (1994) Changing the hours of shiftwork: a comparison of B- and 12-hour shift rosters in a group of computer operators. Ergonomics 37, ) Paley MJ, Price JM, Tepas DI (1998) The impact of a change in rotating shift schedule: A comparison of the effects of 8,10,14 h work shifts. Int J Ind Ergonomics 21, ) Smith L, Folkard 5, Tucker P, Macdonald I (1998) Work shift duration: a review comparing eight hour and 12 hour shift systems. Occup Environ Med 55, ) Matsumoto K, Sasagawa N, Kawamori M (1978) Studies on fatigue of hospital nurses due to shift work. Jpn J Ind Health 20, (in Japanese). 34) Kurumatani N, Koda 5, Nakagiri S, Hisashige A, Sakai K, Saito Y, Aoyama H, Dejima M, Moriyama T (1994) The effects of frequently rotating shiftwork on sleep and the family life of hospital nurses. Ergonomics 37, ) Matsumoto M, Kamata S, Naoe H, Mutoh F, Chiba S (1996) Investigation of the actual conditions of hospital nurses working on three rotating shifts: questionnaire results of shift work schedules, feelings of sleep and fatigue, and depression. Seishin Shinkeigaku Zasshi 98,11-26 (in Japanese). 36) Sakai K, Watanabe A, Onishi N, Shindo H, Kimotsuki K, Saito H, Kogi K (1984) Conditions of night naps effective to facilitate recovery from night work fatigue. J Sci Labour 60, (in Japanese). 37) Matsumoto K, Morita Y (1987) Effects of nighttime nap and age on sleep patterns of shift workers. Sleep 10, ) Sasaki T, Kikuchi Y, Shindo E (1993) The effects of napping in a night duty taken by hospital nurses (1) Changes of arousal level. Jpn J Ergonomics 29, (in Japanese). 39) Matsumoto K, Harada M (1994) The effect of nighttime naps on recovery from fatigue following night work. Ergonomics 37, ) Takahashi M, Arito H, Fukuda H. Nurses' workload associated with 16-h night shifts (2) effects of a nap taken during the shifts. Psychiatry Clin Neurosci (in press). 41) Lowden A, Kecklund G, Axelsson J, Akerstedt T (1998) Change from an 8-hour shift to a 12-hour shift, attitudes, sleep, sleepiness and performance. Scand J Work Environ Health 24 (suppl 3), ) Torsvall L, Akerstedt T, Gillberg M (1981) Age, sleep and irregular workhours: a field study with electroencephalographic recordings, catecholamine excretion and self-ratings. Scand J Work Environ Health 7, ) Harma MI, Hakola T, Akerstedt T, Laitinen JT (1994) Age and adjustment to night work. Occup Environ Med 51, ) Kundi M, Koller M, Stefan H, Lehner L, Kaindlesdorfer S, Rottenbucher S (1995) Attitudes of nurses towards 8-h and 12-h shift systems. Work Stress 9, ) Gold DR, Rogacz S, Bock N, Tosteson TD, Baum TM, Speizer FE, Czeisler CA (1992) Rotating shift work, sleep, and accidents related to sleepiness in hospital nurses. Am J Public Health 82, ) Tan CC (1991) Occupational health problems among nurses. Scand J Work Environ Health 17, Industrial Health 1999, 37,