Sleep quality, but not quantity, is associated with self-perceivedminor error rates among emergency department nursesAmy L. Weaver MSHA, BSN, RN, CEN (Charge Nurse) a,*, Sonja E. Stutzman PhD (ClinicalResearch Manager) b, Charlene Supnet PhD (Scientific Writer) b,DaiWai M. Olson PhD, RN (Associate Professor and Staff Nurse) b

a The University of Texas Southwestern Medical Center, 6201 Harry Hines Boulevard, Dallas, TX 75390, United Statesb The University of Texas Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas, TX 75390, United States

A R T I C L E I N F O

Article history:Received 12 June 2015Received in revised form 3 August 2015Accepted 5 August 2015

Keywords:FatigueSleep deprivationSleep qualityMedical errorsEmergency nursingNursing careError rates

A B S T R A C T

Introduction: The emergency department (ED) is demanding and high risk. The impact of sleep quantity hasbeen hypothesized to impact patient care. This study investigated the hypothesis that fatigue and impairedmentation, due to sleep disturbance and shortened overall sleeping hours, would lead to increased nursingerrors.Methods: This is a prospective observational study of 30 ED nurses using self-administered survey andsleep architecture measured by wrist actigraphy as predictors of self-reported error rates. An actigraphydevice was worn prior to working a 12-hour shift and nurses completed the Pittsburgh Sleep QualityIndex (PSQI). Error rates were reported on a visual analog scale at the end of a 12-hour shift.Results: The PSQI responses indicated that 73.3% of subjects had poor sleep quality. Lower sleep qualitymeasured by actigraphy (hours asleep/hours in bed) was associated with higher self-perceived minorerrors. Sleep quantity (total hours slept) was not associated with minor, moderate, nor severe errors.Discussion: Our study found that ED nurses’ sleep quality, immediately prior to a working 12-hour shift,is more predictive of error than sleep quantity. These results present evidence that a “good night’s sleep”prior to working a nursing shift in the ED is beneficial for reducing minor errors.

© 2015 Elsevier Ltd. All rights reserved.

1. Introduction

Emergency departments (EDs) are 24-hour facilities known for over-crowding, long waits, and elevated patient and family anxiety (Bernsteinet al., 2009). In an environment that requires detailed and specific patientcare measures, sleep deprivation and fatigue can affect quality of caredelivered by an ED nurse (RN), which can negatively impact patientsafety (Dorrian et al., 2006). Because of the ongoing nursing shortage,nurses are working longer shifts and extra days to meet patient demandand provide necessary shift coverage (Griffiths et al., 2014). The impactof this necessity has led to increased stress, fatigue, and sleep disrup-tion (Barker and Nussbaum, 2011; Rogers et al., 2004).

Sleep deprivation and disruption may be associated with fatiguethat leads to errors which could negatively affect patient recoveryor lead to adverse outcomes or eventual death (Johnson et al., 2014).This study was conducted to determine if sleep quality (uninter-rupted sleep) or sleep quantity (number of hours of sleep), prior

to working a 12-hour shift in an ED, affects the perceived error rateof RNs in a critical care setting. Our hypothesis was that fatigue andimpaired mentation, due to sleep disturbance and shortened overallsleeping hours, would lead to increased nursing errors.

2. Background

According to the American Hospital Association, annual ED visitsin the United States have increased to 127.2 million in 2010 (USCensus Bureau, 2011). The ED is unique in that it is a complex areaknown for treating acutely ill patients with a wide array of ill-nesses and injuries, many of which are a mystery upon arrival. TheED is fast-paced and demanding. RNs are frequently asked to or-ganize complex care for patients in life-threatening situations.Interruptions, disruptions, and overcrowding can contribute to a de-crease in cognitive functioning for nurses (Epstein et al., 2012;Practices, 2011; Rivera-Rodriguez and Karsh, 2010). Thus, the EDis an environment ripe for preventable error (Brennan et al., 1991;Epstein et al., 2012; Gawande et al., 1999; Lecoanet et al., 2013).

Fatigue combined with critical situations may increase the riskof error (Griffiths et al., 2014). A 2014 study of critical care nursesin the United States reported that sleep loss (inadequate sleep) anddisrupted sleep negatively impacted nursing decision-making and

* Corresponding author. The University of Texas Southwestern Medical Center, 6201Harry Hines Boulevard, Dallas, TX 75390, United States. Tel.: 817 903 4802; fax: 214633 8725.

E-mail address: amy.weaver@utsouthwestern.edu (A.L. Weaver).

http://dx.doi.org/10.1016/j.ienj.2015.08.0031755-599X/© 2015 Elsevier Ltd. All rights reserved.

International Emergency Nursing ■■ (2015) ■■–■■

ARTICLE IN PRESS

Please cite this article in press as: Amy L. Weaver, Sonja E. Stutzman, Charlene Supnet, DaiWai M. Olson, Sleep quality, but not quantity, is associated with self-perceived minorerror rates among emergency department nurses, International Emergency Nursing (2015), doi: 10.1016/j.ienj.2015.08.003

Contents lists available at ScienceDirect

International Emergency Nursing

journal homepage: www.elsevier.com/ locate /aaen

was associated with an increase in adverse patient outcomes (Scottet al., 2014). This study concluded that nurses who were eitherfatigued or sleep-deprived increased risk for both the nurse and thepatient. (Scott et al., 2014). Although there have been studies thatdescribe medical errors within inpatient and clinical settings, thereis a lack of literature that studies the effect of sleep quality and fatigueon medical errors among ED nurses (Dean et al., 2006; Scott et al.,2006, 2014). This study examined the impact of sleep on self-perceived medical errors of ED nurses.

2.1. Design

This study was a prospective, observational, self-administered surveycoupled with a biophysical measure. The study measured the qualityand quantity of sleep using the Pittsburgh Sleep Quality Index (PSQI)and wrist actigraphy (FitBit). The FitBit was worn prior to working a12-hour shift. After their 12-hour shift, RNs completed a survey to scoreself-perceived minor, moderate, and severe error rates (Table 1).

2.2. Setting

The study was conducted at a 460 bed university hospital, witha 40 bed emergency department. This hospital is a Joint Commis-sion Advanced Comprehensive Stroke Center and certified chest paincenter, located in Dallas County, which has an estimated popula-tion of 2,518,638 (US Census Bureau, 2015). Daily attendance rangesbetween 98 and 140 patients per day, with an average of 3224 pa-tients per month and an annual attendance estimate of 40,000patients for 2015. Patients were leveled 1–5, with level 5 being thehighest level of critical care and level 1 being the lowest level ofcare. Patient levels for 2014 included: Level 5 (13,330), Level 4 (6581),Level 3 (4199), Level 2 (9040), and Level 1 (1358).

Approval was obtained from the Institutional Review Board. ThirtyRNs were recruited who met the inclusion criteria of being as-signed to work full-time in the ED and having completed orientation.

2.3. Data collection

The study duration was six months. Because nurse experienceand education are linked to patient outcomes, each consented nursecompleted a Nurse Experience Form to document their work ex-perience (Aiken et al., 2003). To measure the quality and quantityof sleep, each nurse wore a small wrist actigraphy device once, fora period of approximately 15 hours, prior to his or her next sched-uled nursing shift (Fig. 1). Actigraphy devices record movementduring sleep, and assess sleep patterns while in the subject’s normalsleep setting (Martin and Hakim, 2011). The actigraphy device (FitBit)worn by the participants recorded the total number of hours slept,the number of times during a sleep cycle that the subject awoke(total wake after sleep onset [WASO] minutes), and the number oftimes during a sleep cycle that the subject was restless. The FitBitdevice has been shown to be both reliable and valid for collectingmovement during sleep (Montgomery-Downs et al., 2012).

At the beginning of the 12-hour shift, immediately followingwearing of the device, the nurse completed a PSQI form that was

then submitted, along with the actigraph, to a locked box. After their12-hour shift, the nurse was asked to complete a medical error Self-Assessment Survey, which indicated the amount of self-perceivedminor, moderate, and severe medical errors during the shift. Thesurvey was also placed in a locked box.

3. Measures

3.1. Nursing experience

Nursing experience and education are linked to patient outcomesand was measured with the Nurse Experience Form in this study (Aikenet al., 2003; Olson et al., 2008). The number of full-time equivalent (FTE)experience years of employment in nursing was calculated by multi-plying the number of years worked as a registered nurse by the FTEemployment status of each year and expressed as RN-FTE. Experi-ence in emergency critical care was computed in the same manner, butincluded only those years when the nurse worked in an emergency crit-ical care unit. Nursing degree and certification status data were alsocollected. Nurses participating in the study were assigned a unique iden-tification number and no personal identifiers were associated with anypaper work other than the consent.

3.2. Pittsburgh Sleep Quality Index

The Pittsburgh Sleep Quality Index is a 19 item self-report scalethat measures quality of sleep (Buysse et al., 1989). Seven mecha-nisms are utilized to evaluate sleep behavior and measure the qualityof sleep. Sensitivity and specificity are high (89.6% and 86.5%, re-spectively), and the survey can be completed within five minutes.Previous reliability tests have shown an adequate reliability coef-ficient (α = 0.83) (Buysse et al., 1989).

3.3. Actigraphy

The FitBit is a noninvasive sensory unit that is worn on the wristand quantifies sleep, steps, active minutes, and stationary minutes.The benefit of actigraphy is that it can be worn in the subject’snatural environment and is both reliable and valid in measuring

Table 1Examples of error types.

Error level Definition Example 1 Example 2

Minor Does not reach the patient Choosing the wrong patient in the Pyxis,realizing it, then choosing the right patient.

Documenting in the wrong chart, realizing it, then documenting inthe correct patient’s chart.

Moderate Reaches patient, but does notcause harm

Entering the patient’s room and realizing youhave forgotten the proper supplies, having toleave the room and then return.

Telling the patient you will bring them a glass of water, forgetting,and having to be reminded by the patient a second time.

Major Reaches patient and causes harm The patient was given the wrong medication. A medication was delivered to a patient via the wrong route.

1. The number of minor medical errors I made today was: ___________________________________________________________ 0 100 None Excessive

2. The number of moderate medical errors I made today was: ___________________________________________________________ 0 100 None Excessive

3. The number of severe medical errors I made today was: ___________________________________________________________ 0 100 None Excessive

Fig. 1. The Visual Analogue Scale (VAS) used for self-assessment. *Note that eachVAS was precisely 100 mm in length when distributed to participants.

ARTICLE IN PRESS

Please cite this article in press as: Amy L. Weaver, Sonja E. Stutzman, Charlene Supnet, DaiWai M. Olson, Sleep quality, but not quantity, is associated with self-perceived minorerror rates among emergency department nurses, International Emergency Nursing (2015), doi: 10.1016/j.ienj.2015.08.003

2 A.L. Weaver et al./International Emergency Nursing ■■ (2015) ■■–■■

sleeping behavior (Montgomery-Downs et al., 2012). After use,actigraphy unit information can be uploaded to a computer and saved(Sadeh, 2011).

3.4. Medical error assessment survey

We measured medical errors at three levels and entitled themas “Minor Medical Errors,” “Moderate Medical Errors,” and “SevereMedical Errors” (Table 1). Minor errors were defined as mistakesthat were noticed by the nurse and were corrected before theyreached the patient. Moderate errors were defined as those that werenot noticed by the nurse until they reached the patient, but did notcause any harm. Severe medical errors were defined as errors thatreached the patient, and caused harm. The scales were in the formof a “visual analogue scale” (VAS) (Fig. 1). The VAS is a 100 mm lineand ranged from “no medical errors” to “excessive medical errors.”Nurses drew an intersecting line to indicate their position along thescale. The use of VAS is well validated and was chosen specificallyfor this study, as it allowed for self-perception of medical errors andis therefore subjective (Grant et al., 1999; Guyatt et al., 1987). Usingthis scale also allowed for further protection of nurse subjects, suchthat it would be impossible to identify which nurse completed eachself-report form.

4. Results

The thirty participants in this study were all registered nurses(RNs) working in the ED full-time. Participant gender included 24females and 6 males. Participants’ highest degree in nursing wasdiploma (2), associate (6), and baccalaureate (22). They had a meanof 10.2 years of experience in nursing and 6.4 years of experiencein ED nursing. Two were certified in emergency nursing (CEN), twocertified in critical care (CCRN), and one certified in progressive care(PCCN).

Descriptive statistics and central tendencies were explored forindependent and dependent variables (Table 2). Emergency de-partment RNs, at this hospital, work an assortment of shiftassignments. Day-shift nurses were divided into the following as-signments: 7:00 am to 7:00 pm, 9:00 am to 9:00 pm, and 11:00 amto 11:00 pm. Night-shift assignments included: 1:00 pm to 1:00 am,

3:00 pm to 3:00 am, and 7:00 pm to 7:00 am. Day-shift nurses inthe study worked a mean of 2.9 shifts in a row, which was signifi-cantly higher compared with night-shift nurses with a mean of 2.4shifts (P < 0.05). The primary independent variables are reported inTable 2. In brief, the mean number of hours of sleep (6.63, SD = 1.3.),the mean number of WASO events (18.6, SD = 13.5) and the meannumber of minor (16.3, SD = 10.0), moderate (13.0, SD = 10.2), andsevere (0.3, SD = 0.6) errors was approximately normally distributed.

Error rates were computed using a 100 mm VAS (Fig. 1). All 30participants scored VAS by placing an intersection line along the100 mm VAS to indicate self-perceived minor, moderate, and severeerrors. Central tendency data for minor, moderate, and severe errorsis reported in Table 2. Day-shift nurses spent more hours in bed (8.3vs 7.1; P < 0.05), but did not spend more hours asleep (6.9 vs 6.3;P = 0.22), compared with night-shift nurses. Further, day-shift nurseswere more likely to wake up after falling asleep (23.4 vs 13.1;P < 0.05), compared with their night-shift counterparts. Althoughthe percentage of day-shift nurses (75%) who obtained at least 6hours of sleep was higher than the percentage of night-shift nurses(57%), this difference was not statistically significant in this smallsample (P = 0.32). Scoring responses from PSQI (mean = 7.3; SD = 3.6)indicated that 73.3% of subjects had poor sleep quality. There wasno association between PSQI and minor (P = 0.65), moderate (P = 0.87)or severe (P = 0.83) errors.

Independent associations between sleep and error rates were ex-amined for the total amount of time spent asleep and for the sleepquality prior to the shift worked. The total number of hours asleepwas not associated with minor (P = 0.77), moderate (P = 0.66) norsevere (0.72) errors. Lower sleep quality measured (hours asleep/hours in bed) was associated with higher self-perceived minor errors(P = 0.02). However there was no association between sleep qualityand moderate (P = 0.72) or severe (P = 0.54) errors. Notably, after con-trolling for nursing experience, sleep quality was negativelycorrelated with errors (r2 = 0.21, P = 0.04); however there was stillno association between time spent asleep and errors (P = 0.56).

5. Discussion

The results of this study are both surprising and insightful. Sur-prising because the data do not support our initial hypothesis thatthe number of hours of sleep would be inversely associated withself-perceived medical error rates among ED nurses. The results areinsightful because they reinforce the idea of sleep quality being su-perior to sleep quantity. Additionally, this was the first study totrichotomize self-perceived medical error as minor, moderate, andsevere. The data provide rationale for delineating between the dif-ferent types of self-perceived medical errors in future studies.

Sleep quality and sleep quantity (duration) are fundamentally dif-ferent (Pilcher et al., 1997). For the purpose of the current study, sleepquantity is a simple measure of the total number of hours or minutesof sleep over a specified period of time. Sleep quantity is a variablethat can be measured both directly (observation and actigraphy) andindirectly (e.g., questionnaire) (Montgomery-Downs et al., 2012; Scottet al., 2014; Slater et al., 2015). In this analysis, we excluded WASOfrom total sleep time because the subject is assumed to be awakeduring WASO; this may underestimate sleep quantity for individu-als with restless sleep scored as WASO (Luik et al., 2015).

Sleep quality is more difficult to define and is understandablydefined in a variety of ways by different authors (Krystal and Edinger,2008; Pilcher et al., 1997). Often, subjective and self-administeredtools such as the PSQI, the Epworth Sleepiness scale, and the 5-itemSleep Quality Index (SQI) are used (Bernert et al., 2014; Buysse et al.,1989; Violani et al., 2003). In this study, we used two measures ofsleep quality. The first was the PSQI, which is well validated but onlymeasures self-perceived sleep quality during the past month (Fujiiet al., 2015). Second, we derived sleep quality as a ratio of the number

Table 2Demographics.

All nursesMean (SD)

Day shifta

Mean (SD)Night shiftb

Mean (SD)P-value

Demographics n = 30 n = 16 n = 14Years of nursing

experience10.2 (9.2) 10.9 (9.0) 9.4 (9.7) 0.66

Years of ED experience 6.4 (5.6) 6.6 (5.7) 6.2 (5.6) 0.85Number of shifts

worked per week2.6 (0.6) 2.9 (0.6) 2.4 (0.5) 0.02

Gender 30 13 female3 male

11 female3 male

0.89

Sleep measuresHours in bed 7.8 (1.6) 8.3 (1.7) 7.1 (1.4) 0.04Hours asleep 6.6 (1.3) 6.9 (1.1) 6.3 (1.5) 0.22Wake after sleep onset 18.6 (13.5) 23.4 (15.1) 13.1 (9.1) 0.04At least 6 hours of sleep 20 (66%) 12 (75%) 8 (57%) 0.32PSQI score 7.3 (3.6) 7.6 (4.3) 6.9 (2.9) 0.57Poor sleep by PSQI 22 (73.3%) 11 (69%) 11 (79%) 0.56

Perceived medical errorsMinor 16.3 (10.0) 17.0 (11.3) 15.6 (8.7) 0.72Moderate 13.0 (10.2) 14.0 (11.5) 11.8 (8.9) 0.56Severe 0.3 (0.6) 0.3 (0.6) 0.3 (0.7) 0.87

a Day shift included the following assignments: 7:00 am to 7:00 pm, 9:00 am to9:00 pm, 11:00 am to 11:00 pm.

b Night shift included the following assignments: 1:00 pm to 1:00 am, 3:00 pmto 3:00 am, and 7:00 pm to 7:00 am.

ARTICLE IN PRESS

Please cite this article in press as: Amy L. Weaver, Sonja E. Stutzman, Charlene Supnet, DaiWai M. Olson, Sleep quality, but not quantity, is associated with self-perceived minorerror rates among emergency department nurses, International Emergency Nursing (2015), doi: 10.1016/j.ienj.2015.08.003

3A.L. Weaver et al./International Emergency Nursing ■■ (2015) ■■–■■

of hours of sleep divided by the number of hours in bed. The findingthat low quality of sleep is associated with a higher self-perceivedmedical error rate extends the finding from a 2014 study. Scott et al.’s(2014) study investigated decision regret among critical care RNs,and it also examined self-perceived error among ED RNs. While bothstudies used self-report and PSQI, the current study also used com-mercial wrist actigraphy. Both studies also measured self-perceivedvariables using a VAS. Scott et al. (2014) links decision regret (a self-perceived measure) to sleep quality, wherein the current study linkshigher self-perceived medical error rates to lower sleep quality.

The lack of association between PSQI and self-perceived medicalerror rates is not surprising. The PSQI specifically asks subjects torespond according to the majority of days and nights in the pastmonth. An important finding of the current study is that sleep qualityimmediately prior to a working shift is more predictive of error thanthe long-term sleep quality and pattern experienced by the nurse.This finding is significant in that while it may be important to en-courage nurses to improve their general sleep quality, the resultspresent evidence that a “good night’s sleep” prior to working a shiftin the ED is beneficial for reducing minor errors.

5.1. Limitations

The use of the VAS is both a limitation and strength of this study.A limitation in that the VAS requires self-report and trichotomizederrors as minor, moderate, or severe. Self-report is associated withlower error rates compared with direct observation and thereforemay underestimate the true error rate (Patanwala et al., 2010; Scottet al., 2006). It is possible that error and harm occurred, but waseither not noted or not reported; it has also been found that, amongnurses, self-report of medical errors may underestimate the trueerror rate (Scott et al., 2006). The FitBit may not be considered equiv-alent to the hi-fidelity actigraphy devices available in 2015. However,they are of sufficient and similar quality to the devices available onlya decade ago and provide a low cost alternative to examine sleepand nursing behavior (Lyons et al., 2014; Perez-Macias et al., 2014;Vooijs et al., 2014).

5.2. Implications for emergency nurses

Emergency departments are demanding work environments thatcan cause increased stress and decreased sleep quality among nurses,which may lead to increased medical errors. “Overworked, fa-tigued, and stressed nurses are at a higher risk of making mistakesthat threaten patient safety” (Hasson and Gustavsson, 2010, p. 1). Theresults of our study suggest that the quality of sleep prior to workingthe next shift is important in predicting perceived errors by nurses.Although daily sleep quality is significant to overall health, for nurses,the most important time to get quality sleep is before the start of thenext shift in order to optimize performance on the job. Obtainingquality sleep just prior to working a 12-hour clinical shift may de-crease fatigue and could increase safe delivery of care to patients. Ina 2008 publication by the Agency for Healthcare Research and Quality,Rogers states that “. . .several authorities have recommended that workshifts be limited to 12 hours in a 24-hour period and employees arelimited to working no more than 48 to 60 hours per week” (Rogers,2008, p. 1). Supplementary long-term studies are needed to deter-mine how the quality of sleep over a longer period of time might affectnurse performance and error rates.

6. Conclusions

The quality of sleep before a working shift may be more impor-tant than the quantity of sleep for optimal nurse performance.Restlessness and poor sleep quality appear to affect error rates morethan quantity of sleep. The use of an off-the-shelf actigraph, such

as the FitBit, provides the potential for low cost, high quality re-search with large numbers of participants. Further studies on a largerscale are recommended to gain a better understanding of sleepquality and its effects on medical errors and patient safety. The find-ings support a larger trial with both more nurses and repeatedobservations per nurse. It is important that sleep quality depriva-tion, and not simply sleep quantity deprivation, be recognized asan important factor in contributing to medical errors.

Acknowledgements

The authors wish to thank the emergency department nursesof UTSW for their willingness to take part in this study, and to theUTSW Neuroscience Nursing Research Center staff for their con-tinuous support and insight. Sincere appreciation is accorded toRegina McGary for her expertise in preparing and formatting thispaper.

Conflict of interest

None declared.

References

Aiken, L.H., Clarke, S.P., Cheung, R.B., Sloane, D.M., Silber, J.H., 2003. Educational levelsof hospital nurses and surgical patient mortality. JAMA: The Journal of theAmerican Medical Association. 290, 1617–1623.

Barker, L.M., Nussbaum, M.A., 2011. The effects of fatigue on performance in simulatednursing work. Ergonomics. 54, 815–829.

Bernert, R.A., Turvey, C.L., Conwell, Y., Joiner, T.E., Jr., 2014. Association of poorsubjective sleep quality with risk for death by suicide during a 10-year period:a longitudinal, population-based study of late life. JAMA Psychiatry. 71, 1129–1137.

Bernstein, S.L., Aronsky, D., Duseja, R., Epstein, S., Handel, D., Hwang, U., et al., 2009.The effect of emergency department crowding on clinically oriented outcomes.Academic Emergency Medicine. 16, 1–10.

Brennan, T.A., Leape, L.L., Laird, N.M., Hebert, L., Localio, A.R., Lawthers, A.G., et al.,1991. Incidence of adverse events and negligence in hospitalized patients. Resultsof the Harvard Medical Practice Study I. The New England Journal of Medicine.324, 370–376.

Buysse, D.J., Reynolds, C.F., 3rd, Monk, T.H., Berman, S.R., Kupfer, D.J., 1989. ThePittsburgh Sleep Quality Index: a new instrument for psychiatric practice andresearch. Psychiatry Research. 28, 193–213.

Dean, G.E., Scott, L.D., Rogers, A.E., 2006. Infants at risk: when nurse fatiguejeopardizes quality care. Advances in Neonatal Care. 6, 120–126.

Dorrian, J., Lamond, N., Van Den Heuvel, C., Pincombe, J., Rogers, A.E., Dawson, D.,2006. A pilot study of the safety implications of Australian nurses’ sleep and workhours. Chronobiology International. 23, 1149–1163.

Epstein, S.K., Huckins, D.S., Liu, S.W., Pallin, D.J., Sullivan, A.F., Lipton, R.I., et al., 2012.Emergency department crowding and risk of preventable medical errors. Internaland Emergency Medicine. 7, 173–180.

Fujii, H., Fukuda, S., Narumi, D., Ihara, T., Watanabe, Y., 2015. Fatigue and sleep underlarge summer temperature differences. Environmental Research. 138c, 17–21.

Gawande, A.A., Thomas, E.J., Zinner, M.J., Brennan, T.A., 1999. The incidence and natureof surgical adverse events in Colorado and Utah in 1992. Surgery. 126, 66–75.

Grant, S., Aitchison, T., Henderson, E., Christie, J., Zare, S., McMurray, J., et al., 1999.A comparison of the reproducibility and the sensitivity to change of visualanalogue scales, Borg scales, and Likert scales in normal subjects duringsubmaximal exercise. Chest Journal. 116, 1208–1217.

Griffiths, P., Dall’ora, C., Simon, M., Ball, J., Lindqvist, R., Rafferty, A.M., et al., 2014.Nurses’ shift length and overtime working in 12 European countries: theassociation with perceived quality of care and patient safety. Medical Care. 52,975–981.

Guyatt, G.H., Townsend, M., Berman, L.B., Keller, J.L., 1987. A comparison of Likertand visual analogue scales for measuring change in function. Journal of ChronicDiseases. 40, 1129–1133.

Hasson, D., Gustavsson, P., 2010. Declining sleep quality among nurses: a population-based four-year longitudinal study on the transition from nursing education toworking life. PLoS ONE. 5, e14265.

Johnson, A.L., Jung, L., Song, Y., Brown, K.C., Weaver, M.T., Richards, K.C., 2014. Sleepdeprivation and error in nurses who work the night shift. The Journal of NursingAdministration. 44, 17–22.

Krystal, A.D., Edinger, J.D., 2008. Measuring sleep quality. Sleep Medicine. 9 (Suppl.1), S10–S17.

Lecoanet, A., Sellier, E., Carpentier, F., Maignan, M., Seigneurin, A., François, P., 2013.Experience feedback committee in emergency medicine: a tool for securitymanagement. Emergency Medicine Journal. 31 (11), 894–898.

ARTICLE IN PRESS

Please cite this article in press as: Amy L. Weaver, Sonja E. Stutzman, Charlene Supnet, DaiWai M. Olson, Sleep quality, but not quantity, is associated with self-perceived minorerror rates among emergency department nurses, International Emergency Nursing (2015), doi: 10.1016/j.ienj.2015.08.003

4 A.L. Weaver et al./International Emergency Nursing ■■ (2015) ■■–■■

Luik, A.I., Direk, N., Zuurbier, L.A., Hofman, A., Van Someren, E.J.W., Tiemeier, H., 2015.Sleep and 24-h activity rhythms in relation to cortisol change after a verylow-dose of dexamethasone. Psychoneuroendocrinology. 53, 207–216.

Lyons, E.J., Lewis, Z.H., Mayrsohn, B.G., Rowland, J.L., 2014. Behavior change techniquesimplemented in electronic lifestyle activity monitors: a systematic contentanalysis. Journal of Medical Internet Research. 16, e192.

Martin, J.L., Hakim, A.D., 2011. Wrist actigraphy. Chest Journal. 139, 1514–1527.Montgomery-Downs, H.E., Insana, S.P., Bond, J.A., 2012. Movement toward a novel

activity monitoring device. Sleep and Breathing. 16, 913–917.Olson, D.M., Kelly, A.P., Washam, N.C., Thoyre, S.M., 2008. Critical care nurses’

workload estimates for managing patients during induced hypothermia. Nursingin Critical Care. 13, 305–309.

Patanwala, A.E., Warholak, T.L., Sanders, A.B., Erstad, B.L., 2010. A prospectiveobservational study of medication errors in a tertiary care emergency department.Annals of Emergency Medicine. 55, 522–526.

Perez-Macias, J.M., Jimison, H., Korhonen, I., Pavel, M., 2014. Comparative assessmentof sleep quality estimates using home monitoring technology. ConferenceProceedings – IEEE Engineering in Medicine and Biology Society. 2014, 4979–4982.

Pilcher, J.J., Ginter, D.R., Sadowsky, B., 1997. Sleep quality versus sleep quantity:relationships between sleep and measures of health, well-being and sleepinessin college students. Journal of Psychosomatic Research. 42, 583–596.

Practices, E.I.A.I.F.S.M., 2011. Medication errors in the emergency department:need for pharmacy involvement? Pennsylvania Patient Safety Authority. 8, 1–7.

Rivera-Rodriguez, A.J., Karsh, B.T., 2010. Interruptions and distractions in healthcare:review and reappraisal. Quality and Safety in Health Care. 19, 304–312.

Rogers, A.E., 2008. Chapter 40, p. 1. The effects of fatigue and sleepiness on nurseperformance and patient safety, in: Hughes, R. (Ed.), Patient Safety and Quality:

An Evidence-based Handbook for Nurses. Agency for Healthcare Research andQuality, U.S. Dept. of Health and Human Services, Rockville, MD.

Rogers, A.E., Hwang, W.-T., Scott, L.D., Aiken, L.H., Dinges, D.F., 2004. The workinghours of hospital staff nurses and patient safety. Health Affairs. 23, 202–212.

Sadeh, A., 2011. The role and validity of actigraphy in sleep medicine: an update.Sleep Medicine Reviews. 15, 259–267.

Scott, L.D., Rogers, A.E., Hwang, W.T., Zhang, Y., 2006. Effects of critical care nurses’work hours on vigilance and patients’ safety. American Journal of Critical Care.15, 30–37.

Scott, L.D., Arslanian-Engoren, C., Engoren, M.C., 2014. Association of sleep and fatiguewith decision regret among critical care nurses. American Journal of Critical Care.23, 13–23.

Slater, J.A., Botsis, T., Walsh, J., King, S., Straker, L.M., Eastwood, P.R., 2015. Assessingsleep using hip and wrist actigraphy. Sleep and Biological Rhythms. 13, 172–180.

US Census Bureau 2011. Avalere Health Analysis of American Hospital AssociationAnnual Survey data, 2011. US Census Bureau: National and State PopulationEstimates, July 1, 2011. http://www.aha.org/content/12/preparedtocare.pdf.

US Census Bureau (2015) Dallas county QuickFacts. http://quickfacts.census.gov/qfd/states/48/48113.html.

Violani, C., Lucidi, F., Robusto, E., Devoto, A., Zucconi, M., Ferini Strambi, L., 2003. Theassessment of daytime sleep propensity: a comparison between the EpworthSleepiness Scale and a newly developed Resistance to Sleepiness Scale. ClinicalNeurophysiology. 114, 1027–1033.

Vooijs, M., Alpay, L.L., Snoeck-Stroband, J.B., Beerthuizen, T., Siemonsma, P.C., Abbink,J.J., et al., 2014. Validity and usability of low-cost accelerometers for internet-based self-monitoring of physical activity in patients with chronic obstructivepulmonary disease. Interactive Journal of Medical Research. 3, e14.

ARTICLE IN PRESS

Please cite this article in press as: Amy L. Weaver, Sonja E. Stutzman, Charlene Supnet, DaiWai M. Olson, Sleep quality, but not quantity, is associated with self-perceived minorerror rates among emergency department nurses, International Emergency Nursing (2015), doi: 10.1016/j.ienj.2015.08.003

5A.L. Weaver et al./International Emergency Nursing ■■ (2015) ■■–■■