ArticlePDF Available

The Working Hours Of Hospital Staff Nurses And Patient Safety

Authors:

Abstract

The use of extended work shifts and overtime has escalated as hospitals cope with a shortage of registered nurses (RNs). Little is known, however, about the prevalence of these extended work periods and their effects on patient safety. Logbooks completed by 393 hospital staff nurses revealed that participants usually worked longer than scheduled and that approximately 40 percent of the 5,317 work shifts they logged exceeded twelve hours. The risks of making an error were significantly increased when work shifts were longer than twelve hours, when nurses worked overtime, or when they worked more than forty hours per week.
The Working Hours Of
Hospital Staff Nurses And
P a tient Safety
Both errors and near errors are more likely to occur when hospital
staff nurses work twelve or more hours at a stretch.
by Ann E. Rogers, Wei-Ting Hwang, Linda D. Scott, Linda H. Aiken, and
David F. Dinges
ABSTRACT: The use of extended work shifts and overtime has escalated as hospitals cope
with a shortage of registered nurses (RNs). Little is known, however, about the prevalence
of these extended work periods and their effects on patient safety. Logbooks completed by
393 hospital staff nurses revealed that participants usually worked longer than scheduled
and that approximately 40 percent of the 5,317 work shifts they logged exceeded twelve
hours. The risks of making an error were significantly increased when work shifts were lon-
ger than twelve hours, when nurses worked overtime, or when they worked more than forty
hours per week.
S
everal trends in hospital use and staffing patterns have converged
to create potentially hazardous conditions for patient safety. High patient
acuity levels, coupled with rapid admission and discharge cycles and a short-
age of nurses, pose serious challenges for the delivery of safe and effective nursing
care for hospitalized patients.
1
While systematic national data on trends in the
number of hours worked per day by nurses are lacking, anecdotal reports suggest
that hospital staff nurses are working longer hours with few breaks and often little
time for recovery between shifts.
2
Scheduled shifts may be eight, twelve, or even
sixteen hours long and may not follow the traditional pattern of day, evening, and
night shifts. Although twelve-hour shifts usually start at 7 p.m. and end at 7 a.m.,
some start at 3 a.m. and end at 3 p.m. Nurses working on specialized units such as
202 July/August 2004
DataWatch
DOI 10.1377/hlthaff.23.4.202 ©2004 Project HOPE–The People-to-People Health Foundation, Inc.
Ann Rogers (aerogers@nursing.upenn.edu) is an associate professor in the School of Nursing and in the Center for
Sleep and Respiratory Neurobiology, School of Medicine, University of Pennsylvania, in Philadelphia. Wei-Ting
Hwang is an assistant professor in the Department of Biostatistics and Epidemiology, Center for Clinical
Epidemiology and Biostatistics, University of Pennsylvania School of Medicine. Linda Scott is an associate
professor in the Kirkhof College of Nursing, Grand Valley State University, in Grand Rapids, Michigan. Linda
Aiken is the Claire M. Fagin Leadership Professor of Nursing and a professor of sociology at the University of
Pennsylvania. David Dinges is a professor in the Division of Sleep and Chronobiology, Department of Psychiatry,
Center for Sleep and Respiratory Neurobiology, University of Pennsylvania School of Medicine.
surgery, dialysis, and intensive care are often required to be available to work extra
hours (on call), in addition to working their regularly scheduled shifts. Twenty-
four-hour shifts are becoming more common, particularly in emergency rooms and
on units where nurses self-schedule.
No state or federal regulations restrict the number of hours a nurse may volun
-
tarily work in twenty-four hours or in a seven-day period.
3
Even though state leg
-
islatures in approximately nineteen states have considered bans on mandatory
overtime for nurses and other health care professionals, bills prohibiting manda
-
tory overtime for nurses have passed only in California, Maine, New Jersey, and
Oregon. No measure, either proposed or enacted, addresses how long nurses may
work voluntarily.
4
The recent Institute of Medicine (IOM) report, Keeping Patients
Safe, explicitly recommends that voluntary overtime also be limited.
5
The well-documented hazards associated with sleep-deprived resident physi
-
cians have influenced changes in house staff rotation policies.
6
In contrast, al
-
though shift-working nurses have been the focus of numerous studies, it is not
known if the long hours they work have an adverse effect on patient safety in hos
-
pitals.
7
The purpose of this paper is to examine the work patterns of hospital staff
nurses and to determine if there is a relationship between hours worked and the
frequency of errors.
Study Data And Methods
n
Sample. A cover letter explaining the study and eligibility criteria was mailed
to a random nationwide sample of 4,320 members of the American Nurses Associa-
tion (ANA) during the winter of 2002; 1,725 nurses expressed interest by returning
their completed demographic questionnaire to the Survey Research Institute at
Temple University in Philadelphia. Two logbooks covering a two-week period each,
instructions for completing the logbooks, and postage-paid envelopes were mailed
to 891 eligible subjects (unit-based hospital staff nurses working full time). Three
hundred sixty-two subjects returned both logbooks, and thirty-one completed only
one of the two logbooks, for a return rate of approximately 40 percent. The Institu
-
tional Review Board at the University of Pennsylvania approved this study, and sub
-
jects were paid $140 for their participation.
n
Subjects. The sample of 393 registered nurses (RNs) was predominantly fe
-
male (92 percent), Caucasian (79 percent), middle-aged (mean age 44.8 ±8.8 years,
range 22–66), and experienced (mean 17.2 ±10.0 years). Only 26.3 percent of the par
-
ticipants reported less than ten years’ experience, while 41.9 percent reported
twenty or more years. All participants worked full time (at least thirty-six hours per
week) as hospital staff nurses. Half reported working in hospitals with more than
300 beds; only 11 percent reported working in a hospital with less than 100 beds. The
majority of participants were employed at hospitals located in urban (56 percent) or
suburban (19 percent) areas. The remaining participants worked in hospitals lo
-
cated in small towns (18 percent) or rural areas (7 percent). The characteristics of
Patient Safety
HEALTH AFFAIRS ~ Volume 23, Number 4 203
nurses in the study sample did not differ significantly from those of nurses in the
2000 National Sample Survey of Registered Nurses (NSSRN) in terms of sex, age,
marital status, and work environment (hospital size, urban/rural location, and type
of hospital unit).
8
Our sample has slightly more nurses who identified their ethnicity
as Asian (10.7 percent) than among participants in the NSSRN (3.8 percent).
n
Instruments. Spiral-bound logbooks were used to collect information about
hours worked (both scheduled and actual hours), time of day worked, overtime,
days off, and sleep/wake patterns. Subjects completed seventeen to forty items per
day; all forty questions were completed only on days the nurses worked. Questions
regarding errors and near errors were included, and space was provided for nurses
to describe any errors or near errors that might have occurred during their work pe
-
riods. On days off, nurses were asked to complete the first seventeen questions
about their sleep/wake patterns, mood, and caffeine intake. All items in the logbook
and the logbook format itself were pilot-tested before this study began.
Logbooks (both paper and electronic) have been used to collect data during
field studies of pilots’ cockpit alertness for more than ten years, and from various
other groups of subjects including air traffic controllers, flight controllers during
space shuttle missions, and emergency room physicians.
9
Data recorded about
sleep patterns in these logbooks compare well with data recorded using objective
measures such as wrist actigraphy or ambulatory polysomnography.
10
Although logbooks are not often used to collect information about medical er-
rors, there is some evidence that daily, anonymous, end-of-shift reporting of errors
in a logbook is a valid approach to ascertaining the nature and prevalence of nurs-
ing errors. During a one-month study period of medication errors at a large mili-
tary hospital, nurses completed formal incident reports on only 6 percent of the
medication errors and 15 percent of the near errors that they reported using daily,
anonymous coupons.
11
Another study found that resident physicians also were
more likely to report potential injuries to patients using a confidential e-mail sys
-
tem with daily prompts about reporting than they were to complete traditional
incident reports.
12
n
Analysis. Data from demographic questionnaires and logbooks were summa
-
rized using descriptive statistics and frequency tables. The duration of scheduled
and actual work hours per shift was calculated and aggregated per nurse and per
week. Cutpoints for classifying shift durations were chosen as 8.5 hours and 12.5
hours because “eight-hour” and twelve-hour” shifts are usually scheduled to allow
for a half-hour handover period at the end of the shift. A work shift was classified as
an overtime shift if the actual work hours were longer than the scheduled hours or if
the nurse reported that the shift was “scheduled overtime.”
A binary response for making an error during a worked shift was used as the
primary outcome in analyses. When a nurse caught him/herself before making an
error during a shift, a binary near-error variable was reported and treated as the
secondary outcome. Errors and near errors were codified into categories by study
204 July/August 2004
DataWatch
investigators, based on the descriptions provided in logbooks (for example, medi
-
cation administration, procedural, transcription). The univariate associations be
-
tween the risk of making an error or a near error and (1) the actual duration of the
shift, and (2) overtime were estimated separately using logistic regression models.
The effect of overtime was also examined by stratifying shifts by their expected
duration. Since multiple work shifts from the same nurse contributed to this anal
-
ysis, procedures based on Generalized Estimating Equation (GEE) were used to
determine the odds ratio (OR) while accounting for the nonindependence be
-
tween repeated measurements.
13
Significance tests were two-sided with alpha =
.05. Multivariate analyses also were conducted to evaluate the adjusted associa
-
tions between errors (or near errors), work hours, and overtime, while controlling
for other variables including age, hospital size, and type of hospital unit. For the
week-level data, logistic regression models were performed to assess if working
more than forty hours or fifty hours would increase the probability of making one
or more errors (or near errors) in a week.
Study Results
Data collected on 5,317 work shifts revealed that hospital staff nurses worked
longer than scheduled daily, and generally worked more than forty hours per
week. Half of the shifts worked exceeded ten and a half hours. Although 31 per-
cent of the scheduled shifts were scheduled for durations greater than or equal to
12.5 hours, there were 2,057 shifts (39 percent) where nurses worked at least 12.5
consecutive hours (Exhibit 1). Fourteen percent of the respondents reported
working sixteen or more consecutive hours at least once during the four-week pe-
Patient Safety
HEALTH AFFAIRS ~ Volume 23, Number 4 205
EXHIBIT 1
Description Of Work Patterns Of Full-Time Hospital Staff Nurses, 2002
Variable Number of shifts Percent
Number of shifts 5,317 100.0
Scheduled shifts
a
Up to 8.5 hours
8.5–12.5 hours
12.5 or more hours
2,452
1,183
1,623
46.6
22.5
30.9
Actual shifts
b
Up to 8.5 hours
8.5–12.5 hours
12.5 or more hours
771
2,484
2,057
14.5
46.8
38.7
Number of overtime shifts
Number of mandatory overtime shifts
4,292
360
81.4
6.8
SOURCE: Authors’ analysis of survey results.
a
Scheduled shift hours were missing from 59 shifts. Mean length (hours): 10.3 (standard deviation, ±2.3); range: 1.0–22.5
hours.
b
Actual work hours were missing from 5 shifts. Mean length (hours): 10.8 (SD, ±2.5); range: 1.2–23.7 hours.
riod. The longest shift worked was twenty-three hours, forty minutes.
Nurses reported leaving work at the end of their scheduled shift less than 20
percent of the time during the study period. Although overtime was reported at
the end of all types of shifts, the proportion of shifts involving overtime was signif
-
icantly higher (p = .0001) when eight-hour shifts (85 percent) were compared to
shifts scheduled for eight to twelve hours (79 percent) and twelve hours or longer
(78 percent). Overall, our participants worked, on average, fifty-five minutes lon
-
ger than scheduled each day, and all participants worked beyond their scheduled
work shift (overtime) at least once during the twenty-eight-day data-gathering
period. Almost two-thirds of the nurses worked overtime ten or more times dur
-
ing that period, and a third reported working overtime each day they worked dur
-
ing that period. There were 360 shifts where nurses reported being mandated to
work overtime and another 143 shifts where they described being “coerced to
work voluntary overtime. Even though nurses worked approximately four days
per week, averaging 40.2 (±12.9) hours per week (range 8–97.2 hours per week),
one-quarter worked more than fifty hours per week for two or more weeks of the
four-week period.
There were 199 errors and 213 near errors reported during the data-gathering
period. More than half of the errors (58 percent) and near errors (56 percent) in-
volved medication administration. Other errors included procedural errors (18
percent), charting errors (12 percent), and transcription errors (7 percent). Ap-
proximately 6 percent of the errors and 29 percent of the near errors reported
lacked sufficient information for categorization. Thirty percent of the nurses re-
ported making at least one error, and 32 percent reported at least one near error.
One nurse reported eight errors, while another nurse reported nine near errors.
Our analysis showed that work duration, overtime, and number of hours
worked per week had significant effects on errors. The likelihood of making an er
-
ror increased with longer work hours and was three times higher when nurses
worked shifts lasting of 12.5 hours or more (odds ratio = 3.29, p = .001) (Exhibit 2).
Working overtime increased the odds of making at least one error, regardless of
how long the shift was originally scheduled (OR = 2.06, p = .0005). Our data also
206 July/August 2004
DataWatch
EXHIBIT 2
Association Of Errors Or Near Errors With Nurses’ Work Duration, 2002
Work
duration (hours)
Number
of shifts
Shifts with one or more errors Shifts with one or more near errors
Number Percent OR (p value) Number Percent OR (p value)
Up to 8.5
8.5–12.5
12.5 or more
Total
771
2,484
2,057
5,312
12
77
103
192
1.6
3.1
5.0
3.5
1.00
1.85 (.06)
3.29 (.001)
20
94
97
211
2.6
3.8
4.7
4.0
1.00
1.44 (.18)
1.80 (.04)
SOURCE: Authors’ analysis of survey results.
NOTES: Five shifts with four errors cannot be classified because of missing work durations. OR is odds ratio.
suggest that there is a trend for increasing risks when nurses work overtime after
longer shifts (OR = 1.34, 1.53, and 3.26 for scheduled eight-hour, eight-to-twelve-
hour, and twelve-hour shifts, respectively), with the risks being significantly ele
-
vated for overtime following a twelve-hour shift (p = .005) (Exhibit 3). Although
the effects of working prolonged shifts were clearly associated with errors, there
was no interaction between scheduled shift duration and overtime (p = .17).
Finally, working more than forty hours per week and more than fifty hours per
week significantly increased the risk of making an error (Exhibit 4). Results were
somewhat similar for near errors (Exhibits 2–4).
Nurse and employment characteristics were also examined as potential con
-
founders in the multivariate models. Our results suggest that the relationships of
errors or near errors and work hours and overtime were not affected by age, hospi
-
tal size, or type of hospital unit.
Discussion
This study represents one of the first nationwide efforts to quantify hospital
staff nurse work hours and work patterns, and to determine whether extended
staff nurse work hours contribute to errors and near errors. Our findings confirm
that the work schedules of hospital staff nurses are unpredictably prolonged. All
nurses reported working longer than scheduled at least once, and the majority re-
ported working longer than scheduled ten times or more in a twenty-eight-day
period, as well as working more than forty hours per week. Almost one-sixth of
the sample reported working sixteen or more consecutive hours at least once dur-
ing the period, which suggests that double shifts (or longer) are not confined to
rare emergencies. Mean daily overtime durations were slightly higher than those
Patient Safety
HEALTH AFFAIRS ~ Volume 23, Number 4 207
EXHIBIT 3
Association Of Errors Or Near Errors With Nurses’ Scheduled Work Duration And
Overtime, 2002
Scheduled work
duration (hours)
Number
of shifts
Shifts with one or more errors Shifts with one or more near errors
Number Percent OR (p value) Number Percent OR (p value)
Up to 8.5
No OT
OT
377
2,075
8
65
2.1
3.1
1.00
1.34 (.42)
15
76
4.0
3.7
1.00
0.90 (.74)
8.5–12.5
No OT
OT
246
937
6
36
2.4
3.8
1.00
1.53 (.36)
3
42
1.2
4.5
1.00
2.32 (.08)
12.5 or more
No OT
OT
360
1,263
6
70
1.7
5.5
1.00
3.26 (.005)
8
67
2.2
5.3
1.00
2.34 (.03)
Total 5,258 191 3.6 211 4.0
SOURCE: Authors’ analysis of survey results.
NOTES: Fifty-nine shifts with five errors and two near errors cannot be classified because of missing scheduled work durations.
OR is odds ratio. OT is overtime.
reported in two small observational studies (fifty-five minutes, compared with
forty-two and forty-five minutes, respectively).
14
Although the occurrence of errors did not increase significantly until shift du-
rations exceeded 12.5 hours per day, risks began to increase when shift durations
exceeded 8.5 hours. Since errors are relatively rare, it is possible that this study
lacked sufficient power to detect the effects of work hours or overtime on errors
when nurses were scheduled to work shorter shifts (less than 12.5 hours). Cer-
tainly the trend toward increasing errors with longer work durations is consistent
with other studies that have demonstrated that extended work periods are associ-
ated with increased accidents and neuropsychological deficits among nurses and
have contributed to at least two hospitalwide epidemics of Staphylococcus
aurous.
15
Investigations of these epidemics showed that nurses, who were fatigued
and stressed by high patient caseloads and understaffing, made frequent mistakes
and procedural errors. Despite the lack of information about accident rates involv
-
ing nurses, probed performance tests reveal that nurses working twelve-hour sim
-
ulated shifts make more frequent errors on grammatical reasoning tasks and med
-
ical record reviewing.
16
There are already hints that the fatigue associated with working twelve-hour
shifts is contributing to absenteeism and job dissatisfaction among RNs. Fatigue
related to length of shift or the potential of overtime at end of shift, or both, was
identified as the cause of approximately 12 percent of the absences reported by a
random sample of Canadian hospital staff nurses. Not only did RNs report an un
-
usually high number of sick days year (7.4 days, compared with 3.2 for other work
-
ers), but also nurses working twelve-hour shifts reported significantly higher ab
-
senteeism rates than nurses working traditional eight-hour shifts. Nurses who
worked twelve-hour shifts also expressed lower levels of job satisfaction than
nurses working eight-hour shifts.
17
Inasmuch as the probability of making an error because of long work hours or
208 July/August 2004
DataWatch
EXHIBIT 4
Association Of Errors Or Near Errors With The Number Of Hours Worked Per Week By
Nurses, 2002
Hours worked
Number
of weeks
Weeks with one or more errors Weeks with one or more near errors
Number Percent OR (p value) Number Percent OR (p value)
More than 40
No
Yes
Total
743
681
1,424
64
101
165
8.6
14.8
11.6
1.00
1.96 (<.0001)
75
92
167
10.1
13.5
11.7
1.00
1.42 (.03)
More than 50
No
Yes
Total
1,110
314
1,424
112
53
165
10.1
16.9
11.6
1.00
1.92 (.0001)
120
47
167
10.8
15.0
11.7
1.00
1.46 (.03)
SOURCE: Authors’ analysis of survey results.
NOTE: OR is odds ratio.
overtime was not altered significantly by the age or experience of the nurses, or by
the type of unit or hospital size, other factors may be important. More specifically,
physiological factors such as fatigue, system variables such as increased work in
-
tensity, or a combination of fatigue and increased work intensity may contribute
to the errors and near errors we observed. It is also possible that heavy workloads
themselves may increase the risk of making an error.
The use of mandatory overtime to cover staffing vacancies is a controversial and
potentially dangerous practice.
18
More than one-quarter of nurse participants
(28.7 percent) reported working mandatory overtime at least once during the
data-gathering period, a percentage that is quite similar to that reported in two
surveys of more than 47,000 nurses and in a “Quick Poll” posted on the American
Association of Critical Care Nurses Web site.
19
Mandatory overtime is generally defined as nurses’ being told that they could be
fired, be subjected to disciplinary proceedings, or lose their nursing license if they
refused to stay beyond their regularly scheduled shift or come in to work on their
day off.
20
Although not actually threatened with job loss or disciplinary proceed-
ings, many nurses also report feeling that there will be repercussions if they refuse
to work extra hours or that overtime is voluntary but feels like it is required.”
21
Perhaps that is why approximately 60 percent of the participants in the American
Nurses Association Staffing Survey (N = 4,258) reported being “forced to work
voluntary overtime.”
22
Our data are derived from the self-reports of a relatively small number of hospi-
tal staff nurses and may not be representative of the work schedules and clinical
practices of other U.S. hospital nurses. However, the demographic characteristics
of our nurse sample and our findings about hours worked are consistent with data
reported by hospital staff nurses in the NSSRN, a probability-based sample.
23
In
addition, the percentage of staff nurses who identified twelve-hour shifts as their
usual shift pattern (60.6 percent) is quite similar to Marlene Kramer and Claudia
Schmalenberg’s report that almost two-thirds of the 279 staff nurses they inter
-
viewed worked twelve-hour shifts.
24
Although our response rate was lower than that usually reported for surveys of
nurses, this study required more effort than the usual survey; subjects were asked
to respond to between seventeen and forty items every day for twenty-eight
days.
25
Given the subject burden, it is possible that responders were more invested
than nonresponders were in documenting a relationship between the hours they
worked and effects on patient safety. However, the amounts of overtime reported
varied, with some nurses indicating minimal overtime and others reporting ex
-
tremely long shift durations or working more than fifty hours per week, or both.
Patient Safety
HEALTH AFFAIRS ~ Volume 23, Number 4 209
“The long and unpredictable hours documented here suggest a link
between poor working conditions and threats to patient safety.”
Perhaps more important, the major unit of analysis for this study was the actual
work shift (N = 5,317) rather than the nurse (N = 393).
The definition of error was not specified in the survey instrument. Nevertheless,
all incidents described by participants were obvious deviations from current stan
-
dards of practice. Reported medication errors clearly fell into the categories famil
-
iar to all nurses: wrong patient, wrong medication, wrong dose, wrong route
(such as intravenous, oral), wrong time, and errors of omission.
26
Nurses were
asked whether they made an error, not to assess whether it led to harm.
By not collecting data that could identify where participants worked, we re
-
duced the fears usually associated with reporting errors. Studies have shown that
nurses typically underreport errors because they fear repercussions, including
disciplinary action by employers and regulatory agencies. As a result, only those
errors considered potentially life-threatening, or approximately 5 percent of sig
-
nificant errors, are usually reported.
27
Errors that are considered minor” or are in
-
tercepted before reaching the patient are almost never reported.
28
In fact, near er
-
rors are now considered nonreportable events by the Joint Commission on
Accreditation of Healthcare Organizations (JCAHO).
29
The errors nurses reported in this study occurred in the context of well-docu-
mented deficiencies in nurses’ practice conditions in U.S. hospitals, deficiencies
that nurses have been reporting for well over a decade.
30
The long and unpredict-
able hours documented here suggest a link between poor working conditions and
threats to patient safety. As advocated by the IOM report on medical errors, safer
patient care is more likely to result from changes in the environment in which
health care is provided than from blaming health care professionals, who may be
providing the best care possible under poor circumstances.
31
Hospital staff nurses’ long hours may have adverse effects on patient care; we
found that both errors and near errors are more likely to occur when hospital staff
nurses work twelve or more hours. Because more than three-fourths of the shifts
scheduled for twelve hours exceeded that time frame, routine use of twelve-hour
shifts should be curtailed, and overtime—especially that associated with twelve-
hour shifts—should be eliminated. Additional research with larger samples, in
-
clusion of other variables such as workload and patient acuity, and more precise
measurements of error is suggested.
Financial support for this study was provided by the Agency for Healthcare Research and Quality (R01 HS11963-
01) and a Robert Wood Johnson Foundation Investigator Award in Health Policy Research (Linda Aiken).
Christina Gaughan and Douglas M. Sloane provided valuable statistical consultation.
210 July/August 2004
DataWatch
NOTES
1. L.H. Aiken, J. Sochalski, and G.F. Anderson, “Downsizing the Hospital Workforce,” Health Affairs 15, no. 4
(1996): 88–92; L. Unruh, “Nursing Staff Reductions in Pennsylvania Hospitals: Exploring the Discrepancy
between Perceptions and Data,” Medical Care Research and Review 59, no. 2 (2002): 197–214; and Joint Com
-
mission on Accreditation of Healthcare Organizations, Healthcare at the Crossroads: Strategies for Addressing the
Evolving Nursing Crisis (Oakbrook Terrace, Ill.: JCAHO, 2002).
2. American Nurses Association, Analysis of the American Nurses Association Staffing Survey (Warwick, R.I.: Cor
-
nerstone Communications Group, 2001); California Nurses Association, “Mandatory Overtime Is Detri
-
mental to Patient Care and the Health of Nurses,” 20 April 2001, www.calnurse.org/cna/patient/
nursespeak.html (21 April 2004); and Nurse Week/American Association of Nurse Executives Institute for
Patient Care Research and Education, National Survey of Registered Nurses, 2002, www.nurseweek.com/
survey (8 March 2004).
3. A.E. Rogers, “Work Hour Regulation in Safety-Sensitive Industries,” in Keeping Patients Safe: Transforming the
Work Environment of Nurses, ed. A. Page (Washington: National Academies Press, 2004), 314–358.
4. Ibid.
5. Page, ed., Keeping Patients Safe.
6. D.M. Gaba and S.K. Howard, “Fatigue among Clinicians and the Safety of Patients,” New England Journal of
Medicine 347, no. 16 (2002): 1249–1255; M.B. Weinger and S. Ancoli-Israel, “Sleep Deprivation and Clinical
Performance,” Journal of the American Medical Association 287, no. 8 (2002): 955–957; S.K. Howard et al., “Stim
-
ulation Study of Rested versus Sleep-Deprived Anesthesiologists,” Anesthesiology 98, no. 6 (2003): 1345–
1355; I.R. Holzman and S.H. Barnett, “The Bell Commission: Ethical Implications for the Training of Physi
-
cians,” Mt. Sinai Journal of Medicine 67, no. 2 (2000): 136–139; and Association of American Medical Colleges,
AAMC Policy Guidance on Graduate Medical Education, 2002, www.aamc.org/hlthcare/gmepolicy/start.htm (12
May 2002).
7. See, for example, J. Barton and S. Folkard, The Response of Day and Night Nurses to Their Work Sched-
ules,” Occupational Psychology 64, no. 3 (1991): 207–218; G. Clissold et al., A Study of Female Nurses Com-
bining Partner and Parent Roles with Working a Continuous Three-Shift Roster: The Impact of Sleep, Fa-
tigue, and Stress,” Contemporary Nurse 12, no. 3 (2002): 294–302; N. Kurumatani et al., The Effects of Fre-
quently Rotating Shiftwork on Sleep and the Family Life of Hospital Nurses,” Ergonomics 37, no. 6 (1994):
995–1007; and P. Totterdell et al., “Recovery from Work Shifts: How Long Does It Take?” Journal of Applied
Psychology 80, no. 1 (1995): 43–57.
8. E. Spratley et al., The Registered Nurse Population: National Sample Survey of Registered Nurses, March 2000 (Wash-
ington: Health Resources and Services Administration, 2001).
9. P.F. Gander et al., “Flight Crew Fatigue 1: Objectives and Methods,” Aviation, Space,and Environmental Medicine
69, no. 9 (Suppl.) (1998): B1–B7; M.R. Rosekind et al., “NASA Airlog: An Electronic Sleep/Wake Diary,”
Sleep Research 25 (1996): 525; T.D. Luna, J. French, and J.L. Mitcha, “A Study of USAF Air Traffic Controller
Shiftwork: Sleep, Fatigue, Activity, and Mood Analyses,” Aviation, Space, and Environmental Medicine 68, no. 1
(1997): 18–23; S.M. Kelly et al., “Flight Controller Alertness and Performance during MOD Shiftwork Op
-
erations,” in Seventh Annual Workshop in Space Operations, Applications, and Research (Houston: Space Technology
Interdependency Group, 1993), 405–416; and R. Smith-Coggins et al., “Rotating Shiftwork Schedules: Can
We Enhance Physician Adaptation to Night Shifts?” Academic and Emergency Medicine 4, no. 10 (1997):
951–961.
10. Gander et al., “Flight Crew Fatigue 1”; and Luna et al., A Study of USAF Air Traffic Controller Shiftwork.”
11. P.A. Patrician, L.R. Brosch, and J.A. Williams, “Medication Errors and Nursing Staffing: What’s the Con
-
nection?” (Paper presented at the AcademyHealth Annual Research Meeting, Nashville, Tennessee, 29
June 2003).
12. A.C. O’Neil et al., “Physician Reporting Compared to Medical-Record Review to Identify Adverse Events,”
Annals of Internal Medicine 119, no. 5 (1993): 370–376.
13. K.-Y. Liang and S.L. Zeger, “Longitudinal Data Analysis using Generalized Linear Models,” Biometrika 73,
no. 1 (1986): 13–22.
14. A.L. Tucker and A.C. Edmondson, “Managing Routine Exceptions: A Model of Nurse Problem Solving Be
-
havior,” in Advances in Health Care Management, ed. G.T. Savage, J.D. Blair, and M.D. Fottler (Greenwich, Conn.:
JAI Press, 2002), 87–113; and A.L. Tucker and A.C. Edmondson, Why Hospitals Don’t Learn from Fail
-
ures: Organizational and Psychological Dynamics That Inhibit System Change,” California Management Re
-
view 45, no. 2 (2003): 55–72.
Patient Safety
HEALTH AFFAIRS ~ Volume 23, Number 4 211
15. R.R. Rosa, “Extended Workshifts and Excessive Fatigue,” Journal of Sleep Research 4, Suppl. 2 (1995): 51–56;
K. Reid and D. Dawson, “Comparing Performance on Simulated Twelve Hour Shift Rotation in Young and
Older Subjects,” Occupational and Environmental Medicine 58, no. 1 (2001): 58–62; K. Hanecke et al., “Accident
Risk as a Function of Hour at Work and Time of Day as Determined from Accident Data and Exposure
Models for the German Working Population,” Scandinavian Journal of Work and Environmental Health 24, Suppl.
3 (1998): 43–48; T. Akerstedt, “Work Injuries and Time of Day—National Data” (Proceedings of a Consen
-
sus Development Symposium, Work Hours, Sleepiness, and Accidents,” Stockholm, Sweden, 8–10 Sep
-
tember 1994), 106; B. Russell et al., “An Outbreak of Staphylococcus Aureus Surgical Wound Infection As
-
sociated with Excess Overtime Employment of Operating Room Personnel,” American Journal of Infection
Control 11, no. 2 (1983): 63–67; and P.M. Arnow et al., “Control of Methicillin-Resistant Staphlococcus
Aureus in a Burn Unit: Role of Nurse Staffing,” Journal of Trauma 22, no. 11 (1982): 954–959.
16. M.E. Mills, B. Arnold, and C.M. Wood, “Core 12: A Controlled Study of the Impact of Twelve-Hour Sched
-
uling,” Nursing Research 32, no. 6 (1983): 356–361.
17. L.R. Zboril-Benson, Why Nurses Are Calling In Sick: The Impact of Health-Care Restructuring,” Cana
-
dian Journal of Nursing Research 33, no. 4 (2002): 89–107.
18. M.S. Bosek, “Mandatory Overtime: Professional Duty, Harms, and Justice,” JONAs Healthcare, Law, Ethics, and
Regulation 3, no. 4 (2001): 99–102; K.L. Capitulo, M.L. Ankner, and J. Miller, “Professional Responsibility
versus Mandatory Overtime,” Journal of Nursing Administration 31, no. 6 (2001): 290–292; and L.L. Curtin,
The Case against Mandatory Overtime,” Seminars for Nurse Managers 10, no. 4 (2002): 274–278.
19. J. Robson, “Nurse Survey Validates Testimony on Mandatory Overtime Bill,” 2 May 2002, www.legis.state
.wi.us/senate/sen15/news/inthenews/pr2002-33.htm (28 February 2004); NurseWeek/American Association
of Nurse Executives Institute for Patient Care Research and Education, National Survey of Registered
Nurses; and American Association of Critical Care Nurses, “AACN Online: Quick Poll Archive,” www
.aacn.org/AACN/Surveys.nsf/parchivelist?OpenForm (12 May 2004).
20. M.P. Campbell, Pennsylvania State Nurses Association, Testimony before the House Labor Relations
Committee, on Mandatory Overtime, 30 October 2003, www.psna.org/HotIssues/testimony/
MOTestimony.htm (28 February 2004); and M. Foley, “Statement for the Committee on Ways and Means
Subcommittee on Health regarding Improving Patient Safety” (Washington: American Nurses Associa-
tion, 24 January 2002).
21. Campbell testimony; and R. Steinbrook, “Nursing in the Crossfire,” New England Journal of Medicine 346, no.
22 (2002): 1757–1766.
22. ANA, Analysis of the American Nurses Association Staffing Survey.
23. Spratley et al., The Registered Nurse Population.
24. M. Kramer and C. Schmalenberg, “Staff Nurses Identify Essentials of Magnetism,” in Magnet Hospitals Re
-
visited: Attraction and Retention of Professional Nurses, ed. M.L. McClure and A.S. Hinshaw (Washington: ANA,
2002), 25–59.
25. D.A. Asch, M.K. Jedrziewski, and N.A. Christakis, “Response Rates to Mail Surveys Published in Medical
Journals,” Journal of Clinical Epidemiology 50, no. 10 (1997): 1129–1136.
26. T.M. Pape, Applying Airline Safety Practices to Medication Administration,” Medsurg Nursing 12, no. 2
(2003): 77–93; and B. Krozier et al., Fundamentals ofNursing (Upper Saddle River, N.J.: Prentice Hall, 2000).
27. L.L. Leape, Out of Darkness: Hospitals Begin to Take Mistakes Seriously,” Health Systems Review 29, no. 6
(1996): 21–24.
28. Ibid.; J. Gladstone, “Drug Administration Errors: A Study into the Factors Underlying the Occurrence and
Reporting of Drug Errors in a District General Hospital,” Journal of Advanced Nursing 22, no. 4 (1995):
628–637; and D.S. Wakefield et al., “Perceived Barriers in Reporting Medication Administration Errors,”
Best Practices and Benchmarking in Healthcare 1, no. 4 (1996): 191–197.
29. JCAHO, Sentinel Event ALERT, 11 May 1998, www.jcaho.org/about+us/news+letters/sentinel+event+alert/
sea_4.htm (12 May 2004).
30. Secretary’s Commission on Nursing, Final Report (Washington: U.S. Department of Health and Human
Services, December 1988); Page, ed., Keeping Patients Safe; and JCAHO, Healthcare at the Crossroads.
31. L.T. Kohn, J.M. Corrigan, and M.S. Donaldson, eds., To Err Is Human: Building a Safer Health System (Washing
-
ton: National Academies Press, 1999).
212 July/August 2004
DataWatch
... Nearly 2 decades ago, recommendations were made to curtail routine use of nursing 12-hour shifts because of safety concerns. 1 A growing body of evidence continues to suggest longer shifts contribute to higher burnout rates in nurses and adversely affect quality and safety. [2][3][4][5][6] Other studies indicate nurses prefer extended shifts and perceive them as beneficial to work-life balance and patient care. ...
... Although specific indicators measured were focused on the mental health setting, the findings were consistent with previous studies that found quality and safety indicators are less favorable in units that use 12-hour shifts compared with those that use 8-hour shift patterns. [1][2][3][4][5][6] The findings suggest that longer shift length may lend to escalating behaviors and more disruptive events. The underlying factors influencing these phenomena were not investigated during this study, but the findings are consistent with previous studies that suggest a positive correlation between longer shift lengths and increased adverse events. ...
... The underlying factors influencing these phenomena were not investigated during this study, but the findings are consistent with previous studies that suggest a positive correlation between longer shift lengths and increased adverse events. 1,4,5,16,17 Disruptive events such as verbal and physical assaults negatively impact milieu and therapeutic recovery for patients. 23 Need for mitigation of negative quality and safety outcomes in units that continue to use 12-hour shifts should be considered. ...
Article
Objective: This study compared outcomes between units that used either 8-hour or 12-hour shifts in acute inpatient mental health units. Background: Most hospitals continue to use 12-hour shifts despite research suggesting safety concerns with longer shifts. There is a gap in the literature on effects of shift lengths on nursing and patient outcomes in acute mental health units. Methods: This study is a retrospective comparative analysis of cross-sectional data between 32 inpatient mental health units that used 8-hour versus 12-hour shifts. Independent samples t test was used to examine differences on several staffing, quality, and safety measures. Results: A moderate effect size was found between the groups in quality and safety measures involving patient disruptive behaviors, with the 8-hour group having more desirable outcomes. Conclusions: Nurse leaders in acute mental health units should consider the impacts of shift length on quality and safety when determining staffing patterns. More research is needed to evaluate correlations or causality.
... With continuous and long-term efforts, the utilization rate of equipment may be increased, which will have an impact on the physical and mental health of staff. The severity of patients' diseases and the short cycle from hospital admission to discharge pose a severe challenge to the provision of safe and effective medical services [3], and the high frequency of medical-patient relationships will inevitably lead to higher medical risks, especially in small hospitals and clinics. It is difficult for a medical institution to take full charge of medical service issues and solve all types of medical service management problems. ...
Article
Full-text available
The main purpose of this study is to explore the application of the balanced scorecard (BSC) to service performance measurements of medical institutions using the analytic hierarchy process (AHP) and decision making and trial evaluation laboratory (DEMATEL). According to the concept of BSC, a total of four evaluation dimensions and twenty-two indicators of medical service performance measurements were developed. To collect data, this study delivered expert questionnaires to medical-related professional supervisors, deans, and heads of medical institutions in Taiwan. By combining the AHP and DEMATEL, the priority and causality of service performance standards in medical institutions were obtained. The results of this study show that the customer dimension is the most important service performance measurement dimension for medical institutions. The seven key service performance measurement indicators that are most important for medical institutions, in order, are “complete and comfortable equipment”, “competitiveness of the medical profession”, “continuity of patient-to-hospital treatment”, “classification of medical profession according to customers (VIP system)”, “complete medical service”, “complete salary, remuneration, and policy”, and “medical incomes of institutions”. In terms of causality, provided the complete services of medical institutions are improved, the continuity of patient-to-hospital treatment, the competitiveness of the medical profession, and the medical incomes of institutions would be influenced.
... This situation is thought to explain the negative relationship between weekly working hours and patient safety attitude. Similar to the results of this study, Rogers et al. [15] found that the patient safety level decreased as the working time increased in a study conducted on nurses. ...
Article
Full-text available
Background: Patient safety is a high priority for healthcare systems worldwide. It is considered an indicator of the quality of care. Establishing a patient safety attitude is the first priority in order to create a patient safety culture. Nurses play a critical role in protecting and supporting patients because of the nature of their job. In this way, especially nurses’ attitudes about patient safety will be determined, and threats to patient safety that may arise in the future will be prevented. Methods: In this regard, this study aimed to examine the nurses’ attitudes about patient safety according to certain sociodemographic characteristics. To accomplish this goal, the relevant data of the nurses were obtained by using the Patient Safety Attitude Scale consisting of 6 dimensions and 46 items. The research population consists of 245 nurses working in a University hospital in Ankara. The sample was not calculated, and a questionnaire was distributed to all employees of which 215 nurses completed the questionnaire. Data were collected between 1-30 April 2021. Ethics committee approval was obtained from the hospital. The data obtained were subjected to multivariate regression analysis. Results: The scale used was reliable (r=0.80). The mean of the general patient safety attitude scale is 3.22 with a standard deviation of 0.54. The majority of the participants were found to be between the age groups of 19-26 (38.1%), women (84.7%), and single (52.1%). We also found that the weekly working time of nurses and whether they got patient safety training or no had a statistically significant effect on patient safety attitudes (p<0.05). Therefore, it could be said that as the working time of nurses increased, a decrease in patient safety attitudes were observed, and they exhibited more patient safety attitudes as they got patient safety training. Conclusion: From this point of view, determining the weekly working hours of nurses more appropriately and making them more trained about patient safety may play a key role in creating a higher level of patient safety attitude.
... Their nine-to-five job usually extends to long working hours and a strict dress code. This, along with the constant pressure and caution that comes with being a financial institute, following rules and regulations imposed by the government, and being extra vigilant while handling cash takes a toll on a person's job satisfaction (Rogers et al., 2021). Dealing with customers on a routine basis is just icing on an already pretty stressed cake. ...
Article
An equilibrium state is a pivotal and major aspect of a human being's life. It is effortful to maintain balance in all walks of life including work, family and self. This desire to attain balance made an individual more competitive and ‘fittest for survival’ on the job as well as in family life. All the regulators and ethical practicing organizations sternly emphasize maintaining the balance between work and life. In this research, we aimed to investigate the link between “work-life balance (WLB) and job satisfaction (JS)” in the employees of various banks. The study further aimed to investigate the mediational role of emotional intelligence (EI) with the other two variables; WLB and JS. Three different reliable and valid measures: Work-life Balance (Fields, 2002), Minnesota Job Satisfaction Questionnaire (Spector, 1997), and Emotional Intelligence (Boyatzis, Goleman, & Rhee 2000) were used for data collection. The sample was recruited across Pakistan and consisted of N=198 employees from various banks (Quetta: n=32, Islamabad: n=34, Karachi: n=32, Lahore: n=34, Sahiwal:n=31, and Peshawar: n=35). The sample's age varies from 28 to 58 years with working experience of five years or above in the banking industry. Data analysis revealed a substantial relationship between the balance in work and life, satisfaction with the job, emotional intelligence, and well-being. Results highlighted that both WLB and EI are positive predictors of employee job satisfaction. EI significantly mediates the relationship between WLB and JS. Our findings concluded that employees in banks would be encouraged to develop/enhance their EI. This enhancement in EI would not only influence the WLB but also elevates job satisfaction. Thus, the outcome would be more productivity, relaxed and satisfied employees. Keywords: Employees, Work-life Balance, Job Satisfaction, Emotional Intelligence, Mediational Model and Productivity
Chapter
This chapter shows how the human factors and ergonomics (HFE) discipline can provide concepts, methods, and information for analyzing and redesigning health care systems and processes for the benefits of all involved, e.g., patients, caregivers, physicians, nurses, and other health care workers. It reviews dimensions of quality of care, and dedicate a separate section to the important problem of patient safety. The chapter examines systems approaches to health care and explores the HFE of medical devices and information technology. Special attention is given to the role of HFE in improving health care work systems and processes, and the emerging critical problem of stress and burnout among clinicians. Health information technologies, such as electronic health record technologies, can have negative impacts on patient safety. Several standards and requirements exist to regulate and ensure usability of health technologies.
Article
Full-text available
Çalışanların hata yapma potansiyellerinin azaltılması iş sağlığı ve güvenliğinin artırılması ve dolayısıyla iş veriminin ve kalitesinin artırılması için son derece önemlidir. Hata yapma potansiyeli, çalışanların yerine getirdikleri görevlere, çalışma ortamına, iş yüklerine ve iş koşullarına bağlıdır. Ayrıca her çalışanın görev bazında hata yapma olasılığı kişinin fiziksel ve zihinsel özellikleri bakımından farklılık göstermektedir. Hata yapma olasılıklarına bağlı olarak görevlerin farklı risk düzeyleri mevcuttur. Bu risk düzeylerine göre çalışanlara görev atanması ve iş yüklerinin dengelenmesi iş kazalarını azaltacak ve verimliliğin artmasına neden olacaktır. Bu çalışmada, görev bazında hata yapma olasılıklarına bağlı olarak risk değerlendirilmesi için SWARA ve PCA tabanlı yeni bir HEART yöntemi olan SPC-HEART yöntemi önerilmiştir. Karar vericilerin görüşleri dikkate alınarak hata üreten koşulların önem ağırlıkları SWARA yöntemi ile belirlenirken; her bir çalışan için fiziksel ve zihinsel iş yükü faktörlerinin bileşke etkisi PCA yöntemi ile ortaya çıkarılmıştır. Her bir çalışan için görev bazında hata potansiyellerini hesaplayan proaktif bir risk önleme yaklaşımı elde önerilmiştir. Önerilen yaklaşımın etkinliği hemşireler üzerinde yapılan bir gerçek hayat uygulaması ile gösterilmiştir.
Article
In the absence of effective and adequate vaccines, healthcare decision-makers must rely on non-pharmaceutical interventions (NPIs), such as lockdown, testing, hospital capacity building, and increasing the number of medical staff to control the outbreak of an infectious disease like COVID-19. This manuscript presents a System Dynamics (SD) model to analyze the healthcare system performance under various NPIs during a pandemic. The proposed model, which extends the commonly-used Susceptible-Exposed-Infectious-Recovered (SEIR) model, comprises four sub-models: outbreak, hospital performance, medicine supply, and staff functionality. These sub-models work in harmony to stimulate the impact of NPIs on the disease outbreak pattern and the healthcare system's response to demand surge. The proposed model considers the uncertainty about the nature of the disease, the public's behavior, medicine availability, and medical staff efficiency. The proposed model was applied for the ex-ante evaluation of candidate NPIs adoptable against the COVID-19 outbreak in Iran. Consistent with the reported statistics, the results show that the peak demand can significantly exceed the healthcare system's initial capacity if no action is taken. If simultaneously implemented, lockdown and testing can considerably delay the peak of infections, reduce its magnitude, dampen the hospital demand, and decrease mortality. The proposed model is unique as it determines the extent to which system components (e.g., community, healthcare system, and medicine supply chain) impact the observed outcomes (e.g., morbidity and mortality rates). Its structure is generic and flexible, which facilitates the extension and application of the model to evaluate candidate mitigation policies in various geographical contexts.
Article
Designing the working day is not just a matter of changing the shift system. It also requires reflection on the working conditions and a negotiation of solutions with the stakeholders concerned. This article seeks to show that a participatory approach, built using organisational simulation, provides a framework with which to understand the reality of each profession and co-construct suitable solutions. Our action-research took place in a hospital's pneumology ward. The methodology can be broken down into four phases: diagnosis, sharing of the diagnosis, organisational simulation (the focus of this article) and experimenting with solutions. The results show that the approach gave the stakeholders the opportunity to discover and discuss the rules and constraints of actual work, to compare their different views, and to develop a new and shared view of the work situation. The approach allowed them to co-construct relevant solutions and to appropriate the changes necessary for their success.
Article
Aim To describe the experiences and perceptions of emergency nurses regarding the shortening of night shifts and identify aspects of nurses' preferences for night shifts. Background Shift work can be associated with distinct physical and psychological disadvantages for nurses, especially night nurses. Knowledge regarding the factors influencing their perceptions of night shifts is limited. Methods A qualitative description design. Fifteen nurses from the emergency setting with 6 to 14 years of work experience participated in interviews. Semistructured interviews were conducted between November 2018 and March 2019. A thematic analysis was performed for the data analysis. Findings The following three themes emerged: (1) maintaining quality within quantity, (2) maintaining comfort within busyness, and (3) buffering the gap between ideal and reality. Conclusions Considering work intensity and patient safety, nurses believe that an 8 h night shift is the most suitable length for the emergency department. Long shifts are probably more suitable for other departments with lower night workloads.
Article
Full-text available
Used 4 questionnaires to compare day- and night-shift nurses at a psychiatric hospital to highlight any differences between them on 4 measures. These comprised (1) their satisfaction with their shift schedule; (2) the interference of the shift schedule with their private lives; (3) their reported levels of stress over the past month; and (4) the value they attached to time off work. Findings show an even distribution of reported problems experienced in the private lives of day and night nurses. An important difference between this and previous studies appears to be related to the extent to which individuals have control over the choice of their work hours. The nurses in the present study freely chose to work either day or night shifts. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Article
This paper proposes an extension of generalized linear models to the analysis of longitudinal data. We introduce a class of estimating equations that give consistent estimates of the regression parameters and of their variance under mild assumptions about the time dependence. The estimating equations are derived without specifying the joint distribution of a subject's observations yet they reduce to the score equations for niultivariate Gaussian outcomes. Asymptotic theory is presented for the general class of estimators. Specific cases in which we assume independence, m-dependence and exchangeable correlation structures from each subject are discussed. Efficiency of the pioposecl estimators in two simple situations is considered. The approach is closely related to quasi-likelihood.
Article
The characteristics, education, employment patterns, salaries, job satisfaction, and other characteristics of registered nurses (RNs) across the United States were examined in a national survey. Of the initial sample of approximately 54,000 of the nation's more than 3,066,000 licensed RNs, 35,579 RNs (72%) submitted usable responses. From 1980 to 2000, the RN population increased by more than 1 million with 1996-2000 marking the slowest growth in the RN population during the 20-year period. The percentage of nurses receiving their basic education in diploma programs decreased from 60% to 30%, with the percentage completing associate degree programs increasing from 19% to 40%. Hospitals remained the major employer of nurses although the number of nurses employed in other sectors--especially public and community health, ambulatory care, and other noninstitutional settings--increased. In 1980-2000, full- time RNs actual annual salaries increased from $17,398 to $46,782, whereas their real salaries increased from $17,398 to $23,103. Across the entire sample, just two-thirds of the RNs reported being satisfied with their current position. (Chapter 1 presents information about early RN studies, development of the present study's methodology, the sample of RNs for the present study, and 15 references. Appendixes constituting approximately 80% of the document contain 48 tables, a description of the survey methodology, and the survey questionnaire.) (MN)