Rotating Shift Work_ Sleep_and Hospital Nurses by ghkgkyyt

VIEWS: 17 PAGES: 4

									Public Health Briefs


Rotating Shift Work, Sleep, and
Accidents Related to Sleepiness in
Hospital Nurses
Diane R. Gold, MD, MPH, Suzanne Rogacz, MD, Naomi Bock, MD, MS,
Tor D. Tosteson, DSc, Timothy M. Baum, MS, Frank E. Speizer, MD, and
Charles A. Czeisler, PhD, MD

Introduction                                  dents in the past year. The variable "any
                                              accident or error" included automobile
     Concem for iatrogenic risks to pa-       accidents, medication errors, on-the-job
tients caused by hospital staff experienc-    procedural errors, and on-the-job per-
ing sleep disruption has led to a reexami-    sonal injuries that the nurse reported had
nation of the work schedules and sleep        occurred because of sleepiness.
pattems of physicians."> This cross-sec-           We used prior hypotheses based on
tional study was designed to examine the      physiologic and epidemiologic data re-
impact of work schedule on the sleep          garding shift work and circadian disrup-
schedule, sleepiness, and accident rates of   tion7-17 in the definition of shift categories
female nurses in a Massachusetts hospital.    (Table 1), which were created from shift
                                              schedules for the current month. Day/
                                              evening shift work was grouped as a single
Methods                                       category because shifting from days to
                                              evenings has not been demonstrated to
     A self-administered questionnaire        disrupt circadian rhythms.
was distributed June through September              Sleep and wake times were used to
1986 to 878 registered nurses, licensed       determine whether a nurse obtained "an-
practical nurses, and other ancillary staff   chor sleep'- t least 4 hours of sleep ob-
in the hospital. The nurse was asked to       tained regularly during the same clock
record, for the current week, the previous    hours every night, both during work days
2 weeks, and the following week, the num-     and days off. Studies by Minors and Wa-
ber of shifts worked for each work-shift      terhouse18 suggest that loss of anchor
category (i.e., day, evening, night) at the   sleep may be a surrogate for circadian
hospital and at any other job. The nurse      rhythm disruption.
was also asked, "Does your job involve a            The relationship between shift and
variable work shift? That is, do you work     outcomes such as sleepiness and acci-
the day shift sometimes and the night shift   dents was explored first through univari-
at other times?," which is a National Cen-    ate and then through multivariate analy-
ter for Health Statistics (NCHS) question
previously used to identify shift workers
in the US population.5'6 In addition, the     Diane R. Gold, Naomi Bock, Tor D. Tosteson,
                                              Timothy M. Baum, and Frank E. Speizer are
nurse recorded her sleep and wake times       with the Channing Laboratory, and Suzanne
when she worked the day shift, the            Rogacz and Charles A. Czeisler are with the
evening shift, the night shift, and on days   Center for Circadian and Sleep Disorders Med-
off.                                          icine, Department of Medicine, Harvard Med-
      Information was collected regarding:    ical School, Brigham and Women's Hospital,
                                              Boston.
quality of sleep; the use of alcohol (per           Requests for reprints should be sent to
month), prescription or nonprescription       Diane R. Gold, MD, MPH, Channing Labora-
medication, sleeping aids, or other kinds     tory, Brigham and Women's Hospital, Harvard
of drugs to get to sleep; nodding off at      Medical School, 180 Longwood Avenue, Bos-
                                              ton, MA 02115.
work (per week); nodding offwhile driving           This paper was submitted to the Journal
to or from work in the past year; and ac-     March 6, 1991, and accepted with revisions No-
cidents, errors, and "near-miss" acci-        vember 12, 1991.


                                                   American Journal of Public Health 1011
       Public Health Briefs

                                                           ses, using EPISTAT and SAS (proc                      pared to day/evening nurses, more rota-
                                                           logistic). We attempted to validate the re-           tors were 35 years old or younger (77.8%
                                                           porting of medication errors, but this was            versus 64.5%). By contract, the option of
                                                           not possible because written accident re-             not rotating and not working the night shift
                                                           ports submitted to the hospital's risk man-           was dependent on seniority. Conse-
                                                           agement office usually documented the                 quently, whereas 23.7% of rotators and
                                                           nursing supervisor rather than the nurse              22.1% of night nurses had worked at the
                                                           associated with the error.                            hospital for 1 year or less, only 9.3% of
                                                                                                                 day/evening nurses had done so.
                                                           Resudt                                                      During their work days, rotators and
                                                                                                                 night nurses reported fewer hours of sleep
                                                                Of 878 hospital employees contacted              than day/evening nurses (Figure 1). Of the
                                                           by the research assistants, 687 (78.3%) re-           day/evening nurses, 92.2% obtained an-
                                                           turned the questionnaire, 36 (4.1%) re-               chor sleep regularly throughout the
                                                           fused to participate, and 155 (17.7%) failed          month. In contrast, only 6.3% of night
                                                           to return the questionnaire. The study co-            nurses, none of the rotators, and none of
                                                           hort included 593 female registered nurses            the nurses in work categories 4 through 6
                                                           and 42 female licensed practical nurses.              (Table 1) obtained anchor sleep regularly
                                                                The mean age of the nurses was 33.9              throughout the month.
                                                           years (range 21 to 65 years). When com-                     Of the nurses from all work categories
                                                                                                                 who responded "yes" to the NCHS ques-
                                                                                                                 tion on variable work schedules, only
                                         IIORXDAYS                                                               53.6% experienced anchor sleep disrup-
                                                                                                                 tion. The group responding "yes" included
                                                                                                                 49.4% of the day/evening nurses, 94.1% of
              as 1                                                                                               rotators, and 2.9% of night nurses.
                                                                                                                       In comparison to day/evening nurses,
                                    ~~~
                                     70 ~                            ~     ~~~~~~~
                                                                            _          _
                                                                                       (=                        night workers had 1.8 times the odds and
                                                                                       ~~~~~~~~~~~~~slsep/24 h   rotators had 2.8 times the odds of report-
                    R
                                                                     ~~~*>5-7 hi                                 ing poor quality sleep (Table 2). Night
                           ~~ ~
                         E ~         ~      ~   ~     ~     ~    ~
                                                                                                                 nurses and rotators had twice the odds of
          T                                                                          U >L h                      using medications to get to sleep.
                                                                                                                       Nodding off on the night shift oc-
                                                                                                                 curred at least once per week in 35.3% of
              20-
              is                                                                                                 rotators, 32.4% of night nurses, and 20.7%
               a
                                                                                                                 of day/evening nurses who worked occa-
                           Day/     might       Rotator          Occasional                                      sional nights. On the other hand, neither
                        Evening                                      Nights                                      day/evening nurses nor rotating nurses re-
                                         DAYS OFF                                                                ported significant problems with nodding
                                                                                                                 off on the day or evening shift (rates for
              lee                                                                                                nodding off: 2.8% and 2.7%). When com-
                                                                                                                 pared to day/evening nurses, rotators had
              79-                                                                                                3.9 times the odds and night nurses had
                                                                                                                 3.6 times the odds of nodding off while
                                                                                                                 driving to or from work in the preceding
~~~~     fie.                                                                                                    year (Table 2).
                           n                                          Ngt
                                                                                                                       In univariate analyses two confound-
               . Onl .% of dyeni n. e s                   5 _o                                                   ers of the relationship between shift and
                                                                                                                 accidents were identified: working at the
                            Ca      Nigt            Roao         O-casina                                        hospital 1 year or less was associated with
                                                                                                                 medication errors, and an age of 35 years
                                                                                                                 or younger predicted automobile near-
                                                                                                                  miss accidents. The use ofalcohol to get to
                           Day-o    might           Rtotator -Occeasional                                         sleep was an independent predictor of all
                        EvPen ing                              N1ights                                            categories of accidents.
           Ak. Only 396 of day/evo nurse sbM :r 5 hors2 hours on wordays, as ooxnparSed                                Adjusting for these factors, the odds
           to 8% of rotars and 20% of rnght nurses. s tan 50% of nurses fom each of the work                      of reporting any accident or error were
           scedusspt >7 hour*24 how onworkdays. On daysofgreaterthan75% ofnumsesfrcrf                             twice as high for rotators as for day/
           each of Fe work sd e slpt >7 hours/24 hours.
                                                                                                                  evening nurses (Table 3). Rotators had 2.5
           FIGURE 1 -Hours of sleep per 24 hours on workdays and on days off by catory of                         times the odds of reporting near-miss ac-
                     work schedule.                                                                               cidents. After adjustment the effect of ro-
                                                                                          i


        1012 American Joumal of Public Health                                                                                        July 1992, Vol. 82, No. 7
                                                                                                                                                                                                                                                                                                                                                                                       Pubik Health Briefs




                              M
                                                                                                                   ......................

                                                                                                                                                                                                                                                                                                             ...........



     ..................................                                                                           .........................
                                                                            X.x.
                                                                                                                                                                                                                                       ............................I
                                                                                                                                                                                                                                                                                ------------                                                                                                                        x
                                                                                                                                                                                                                                                                                                  ............                                                                                                      ...
                                                                                                                                                                                                                                                                                                                         ................

                                                                                                                                                                                                               ...................
                                                                                                                                                                                                                                                                        P -0      ...........    --------------------------

                                                                                                                                                                        X
                                                                                                                                                                                                                                                                                                                                             -N.
                                                                                                                                                                                                                   ...             ...............
                                                                                                                                                                                                                         .......             ....x




                                                                                                                                                                                                                                             ............X
                                                                                   x ...




                                                                                                                                                                                                                                                                                                                                                                                ;..............
                                                                                                                                                                                       .....................................................




                                                                                                                                                    :.X,X




                                                                                           :X
                                                                                                ...........                                                                                                                                                                                                                                                ............
                                                                                                                                                                                                                                                                                                                                                                                                         ...


                                                                                                                                                                                                                ...........            ...........................


                                                                                                                                                                                                                                                                          X.:
                                                                                                                                                                                                                                                                                                                                   ---------------                                        ..........................

                                                                                                                                                                                                                                  ..................
                                                                                                                                                                                                                  ...........M--.I-       ..-M-;-:"                                                                                                           X
                                                                                                                                                                                                   ..........................        cm.-C.'e.....'ee
                                                                                                                                                                                                            ..................                          ......
                                                                                                                                      .......

                                                                                                                                                                                                                                                                                 ..........................
                                                                                                                                                                                                                                                                                                                                      ............................... xv- -..

                                                                                                                                                                                                                                                                                                                                       ............................
                                                                                                                                                            .......................................................




                                                                                                                                                                                                                                                                                                                .............




                                                                                                                                                                                                                                                                                                                                                            ..........


                                                                                                                                                                                                                                                                                                                                                              ..         ..


                                                                                                                                                                                                                    .........
                                                                                                                                                                                                                                                                                                                                                 ..............

                                                                                                              R
                                                                                                                                                                                                                                                                                                                         ...........K
                                                                                                                                                                                                                                               X
                                                                                                                                                                                                                                                                                                                                 X:
                                          ..................




                       ...........
                                                                                                                              -1   ......
                                                                                                                                                                                                                                                                                                  ........
                                                                                                                                                                                                                                                                                                                                                                                           ...........




                                                                                                                                                                                                                                                   ..........
                                                  ...........................




                                                                                                                                                                                                                                                                                                                                 X
                                                                                                                                                                                                                                     ...                                                                                                    e,
                                                                                                                                       ..................




                                                                                                                                                                                                                                                                                                                                                                                                                           -X


 o2o2.4                    ..........




                                                                                                                                                                                                                                                             ...............




                                                                                                                                                                              ..X


                      67                                                                                                                                     ::X
                                                                                                                                                                                                                                                                                                                   NO.
                                                                                                                                                                                                                                                                                                                                                                                                               .........




tating on medication errors was reduced                                                                                                         sectional study. The hypotheses gener-                                                                                                           grants NICHHD 5-RO1-HD20174 and NHLBI
from 2.2 (95% CI: 1.0, 4.9) to 1.8 (Table 3).                                                                                                   ated might be best tested through a case-                                                                                                        5-T32-HL07427 (Gold and Tosteson) and
                                                                                                                                                                                                                                                                                                 5-MOI-RRO2635 (Rogacz).
                                                                                                                                                control study design, with validation of                                                                                                               We would like to acknowledge the enthu-
Disussion                                                                                                                                       reported accidents.                                                                                                                              siastic participation ofthe hospital nurses, nurs-
                                                                                                                                                     The data present a potential di-                                                                                                            ing administrators, and Department of Risk
     The results of this study are consistent                                                                                                   lemma for hospital policymakers. Even if                                                                                                         Management and Loss Control personnel in
                                                                                                                                                rotating leads to circadian rhythm disrup-                                                                                                       this research project. We are grateful to Mr.
with laboratory investigations that have                                                                                                                                                                                                                                                         Joseph Ronda for assistance in data manage-
demonstrated that sleep deprivation and                                                                                                         tion and accidents, nurses and hospital                                                                                                          ment and to Ms. Darlene Bramble for secre-
misalignment of circadian phase as experi-                                                                                                      administrators will have to determine                                                                                                            tarial assistance.
enced during rotating shift work are each                                                                                                       their priorities: the family responsibilities
associated with frequent lapses of attention                                                                                                    of nurses and the staffing requirements of                                                                                                        References
                                                                                                                                                hospitals may limit the potential for al-                                                                                                             1. Friedman RC, Bigger JT, Komfield DS.
and increased reaction time, leading to in-                                                                                                                                                                                                                                                              The intern and sleep loss. N Engli Med
                                                                                                                                                                                                                                                                                                                                          J
creased error rates on performance                                                                                                              tering hospital schedules to improve the                                                                                                                          1971;285:201-203.
tasks.7-9 These data also suggest that a                                                                                                        health of nurses and the safety of pa-                                                                                                                2. Lurie N, Rank B, Parenti C, Woolley T,
record of a representative work schedule                                                                                                        tients.1920 El                                                                                                                                           Snoke W. How do house officers spend
may be of greater use than the NCHS vari-                                                                                                                                                                                                                                                                their nights? A time study of internal med-
                                                                                                                                                                                                                                                                                                         icine house staff on call. N Engi J Med
able shift question in identifying workers                                                                                                                                                                                                                                                               1989;320:1673-1677.
with disrupted sleep/wake patterns.                                                                                                             Acknowledgments                                                                                                                                       3. Orton DI, Gruzelier JH. Adverse changes
     Reporting bias and selection bias may                                                                                                      This research was supported in part by Na-                                                                                                               in mood and cognitive performance of
have influenced the findings in this cross-                                                                                                     tional Institutes of Health research and training                                                                                                        house officers after night duty. Br Med J.



July 1992, Vol. 82, No. 7                                                                                                                                                                                                                                                                                            American Journal of Public Health 1013
Public Health Briiefs

    1989;298:21-23.                                    Efficieny. London, England: English Uni-           work in full and part-time night nurses. J
 4. Durnford S. Junior hospital doctors: tired          versities Press, Ltd.; 1972:101-113.              PhysioL 1977;267:23P.
    and tested. BrMed J. 1988;297:931-932.           9. Williams HL, Lubin A, Goodnow JJ. Im-         16. Adams J, Folkard S, Young M. Coping
 5. Danchik KM, Schoenbom CA, Elinson J.                paired performance with acute sleep loss.         strategies used by nurses on night duty.
    H ighlght from Wave I of the National               Psychological Monogr. 1959;73:1-26.               Ergonomics. 1986;29:185-196.
    Swvey of Personal Health Practces and           10. Czeisler CA, Weitzman ED, Moore-Ede           17. Gadbois C. Women on night shift: interde-
    Consequences: United States. Vital and              MC, Zimmerman JC, Knauer RS. Human                pendence of sleep and off-the-job activities.
    health statistics; series 15, no. 1. Washing-       sleep: its duration and organization depend       In: Reinberg A, Vieux N, Andauer P, eds.
    ton, DC: US Government Printing Office;             on its circadian phase. Science. 1980;210:        NightandShift Wor*Biological and Social
    1981; US DHHS publication PHS 81-1162.              1264-1267.                                        Aspects. New York, NY: Pergamon Press;
 6. Gordon NP, Cleary PD, Parker CE,                11. Bjerner B, Swensson A. Schichtarbeit and          1981:223-227.
    Czeisler CA. The prevalence and health
                                                        rhythmus. Acta Med Scand SuppL 1953;          18. Minors DS, Waterhouse JM. Anchor sleep
                                                        278:102-107.                                      as a synchronizer of rhythms on abnormal
    impact of shiftwork. Am J Pubic Health.         12. Moore-Ede MC, Czeisler CA, Richardson
    1986;76:1225-1228.                                  GS. Circadian timekeeping in health and           routines. IntJ Chronobiology. 1981;7:165-
 7. Dinges DF, Ome MT, Whitehouse WG,                   disease. NEngIJMedi 1983;309:530-536.             188.
    Orne EC. Temporal placement of a nap for        13. Akerstedt T. Sleepiness as a consequence      19. Czeisler CA, Moore-Ede MC, Coleman
    alertness: contributions of circadian phase         of shift work. Sleep. 1988;11:17-34.              RM. Rotating shift work schedules that dis-
    and prior wakefulness. Sleep. 1987;10:313-      14. Folkard S, Monk TH, Lobban MC. Short              rupt sleep are improved by applying circa-
    329.                                                and long-term adjustment of circadian             dian principles. Science. 1982;217:460-463.
 8. Hamilton P, Wilkinson RT, Edwards RS. A             rhythms in 'permanent' night nurses. Er-      20. Knauth P, Rutenfranz J. Development of
    study of four days partial sleep deprivation.       gonomics. 1978;21:785-799.                        criteria for the design ofshiftwork systems.
    In: Colquhoun WP, ed. Aspects ofHuman           15. Folkard S, Haines SM. Adjustment to night         JHwn ErgoL 1982;11:337-367.




                                                    Birthweight Distributions in Mexico
                                                    City and among US Southwest
                                                    Mexican Americans: The Effect
                                                    of Altitude
                                                    Francis C. Notzon, PhD, Jose Luis Bobadjila, MD, PhD, and
                                                    Irma Cona, MPH

                                                    Inftdudion                                        lower than the rate for US Blacks or Pu-
                                                                                                      erto Ricans (Table 1). This phenomenon,
                                                          Low birthweight rates (birthweight          descnbed as a "public health enigma," is
                                                    <2500 g) vary widely with racial or ethnic        not readily explainable.8 Nevertheless,
                                                    origin.1-3 However, these differences are         the low birthweight rate of Mexican
                                                    attributed primarily to socioeconomic             Americans is so dramatically different
                                                    conditions strongly associated with these         from other US minority groups that a ver-
                                                    groups,4- leading to the assumption that          ification of the accuracy of the Mexican-
                                                    the more socioeconomically disadvan-              American rate is warranted.
                                                                                                           One approach to verifying these data
                                                    taged groups are the most likely to have          is to compare Mexican-American birth-
                                                    low birthweight babies. Low socioeco-             weights with those found in Mexico. If
                                                    nomic status often is associated with fac-
                                                    tors known to impair the rate of fetal
                                                    growth, such as inadequate nutrition and          Francis C. Notzon is with the National Center
                                                                                                      for Health Statistics, Centers for Disease Con-
                                                    smoking.7 Insufficient prenatal care and          trol, US Department of Health and Human
                                                    other factors also may mediate between            Services, Hyattsville, Md. Jose Luis Bobadilla
                                                    low socioeconomic status and low birth-           is with the Population, Health and Nutrition
                                                    weight.4 While the association between            Division, The World Bank, Washington, DC.
                                                                                                      Irma Coria is with the Department of Analysis
                                                    socioeconomic status and ethnicity ap-            and Statistics, National Institute of Perinatol-
                                                    pears to hold for US Blacks and Puerto            ogy, Mexico City, DF, Mexico.
                                                    Ricans, this is not so for Mexican Amer-                 Requests for reprints should be sent to
                                                     icans.                                           Francis C. Notzon, PhD, National Center for
                                                          The low birthweight rate for Mexican        Health Statistics, 6525 Belcrest Road, Hyatts-
                                                                                                      ville, MD 20782.
                                                     Americans is similar to that ofUS Whites,               This paper was submitted to the Journal
                                                     despite low socioeconomic status and             December 4, 1990, and accepted with revisions
                                                     poor access to prenatal care, and is far          November 27, 1991.


                                                                                                                             July 1992, Vol. 82, No. 7

								
To top