Increased errors and decreased performance at night: A systematic review of the evidence concerning shift work and quality

Abstract

BACKGROUND:

Shift workers have worse health outcomes than employees who work standard business hours. However, it is unclear how this poorer health shift may be related to employee work productivity.

OBJECTIVE:

The purpose of this systematic review is to assess the relationship between shift work and errors and performance.

METHODS:

Searches of MEDLINE/PubMed, EBSCOhost, and CINAHL were conducted to identify articles that examined the relationship between shift work, errors, quality, productivity, and performance. All articles were assessed for study quality.

RESULTS:

A total of 435 abstracts were screened with 13 meeting inclusion criteria. Eight studies were rated to be of strong, methodological quality. Nine studies demonstrated a positive relationship that night shift workers committed more errors and had decreased performance.

CONCLUSIONS:

Night shift workers have worse health that may contribute to errors and decreased performance in the workplace.

Keywords

After-hours efficiency productivity safety

1 Introduction

Shift work is indispensable among industries that function 24 hours day. According to the United States (US) Bureau of Labor Statistics, shift work is defined as employees who work evening shifts (1400–2400), night shifts (2100–0800), rotating shifts, including employer-arranged irregular schedules, and/or hours in which employers can vary the time of the shift to meet the needs of the business [1]. In the US, the percentage of shift workers is approximately 27.5% and more than 21 million employees (18% of all workers) work outside the daytime shift range (between the hours of 0600 and 1800) [1]. In some countries, including China, Chile, Senegal, the UK, and the EU member states, a large percentage of workforce (38%) works outside standard business hours [1]. Countries with the highest instances of shift work have economies that rely heavily on industries, where shift work is necessary, such as manufacturing, transportation, retail and hospitality sectors [2]. Therefore, understanding how shift work impacts both the employee and the industry is important.

1.1 Shift work and employee health

It is well established that shift work is deleterious to employees’ health. While shift work is critical in certain industries and work sectors and the adverse association between non-standard work hours and employee health is well documented [3 –7]. For example, night shift workers are more likely to develop metabolic syndrome, a predecessor to obesity [3 , 8]. In one study, researchers found a 4-fold greater prevalence of elevated waist circumference among police officers working the midnight shift compared with those on the day shift (PR = 3.98; 95% CI, 1.72–9.22) [6]. This relationship was also true among female shift workers who had significantly elevated odds ratio for obesity (OR = 2.7; 95% CI, 1.6–4.5), central obesity (OR = 2.9; 95% CI, 1.7–5.1) and high blood pressure (OR = 2.3; 95% CI, 1.2–4.4) compared to female office workers [8].

In addition to metabolic syndrome, numerous researchers have established the link between shift work and impaired sleep across a variety of work settings. Impaired sleep may be related to circadian disruption results in the shift worker feeling tired and sleepy in the middle of the night [9]. This disturbance in the natural sleep cycle results in chronic fatigue [4 , 10–13]. In a randomized controlled trial of 288 automobile factory workers, 61% of night shift workers suffered from severe sleepiness and were also more likely to fall asleep while working [14]. Night shift workers who experience sleep deficits have decreased vigilance and concentration, as well as decreases in attention, memory, and accuracy as compared to well-rested workers [15]. A sleep deficit and impacted may lead to severe sleepiness at work, especially on the night shift [16]. Sleep impairments can impact the ability to drive home after working at night increasing the odds of an injury-involved motor vehicle crash six-fold [17, 18].

There are psychological and social effects of shift work. For example, shift workers often find it challenging to find a balance between their jobs, family, and social life [19]. In fact, night shift workers have been found to experience considerably greater work-family conflict than day shift workers [20]. Shift workers may experience a lack of social interaction with people who work a normal daytime schedule contributing to depressive symptoms [21]. Numerous studies show that shift workers experience a sense of isolation from family and friends and a lesser ability to fulfill domestic roles due to anxiety, irritability, and low self-esteem [20]. For female shift workers, additional responsibilities may include caring for children, housework, and chores, which all limit time for rest and recreation [22]. In a recent systematic review, researchers found that night shift workers and specifically, those employees who rotated to work at night had worse well-being and were less satisfied with their jobs [23].

Shift work is also related to increased occupational injuries. In health care, shift work is associated with a higher risk of work-related injuries, which included blood or body fluid exposure from needle sticks and musculoskeletal injuries [24]. Occupational injuries also occur among different industries where production may vary or when more intense work is performed [25]. Indeed, researchers found that the highest risk of accidents and injuries occurs at about midnight and that risk increases if the employee has worked four successive night shifts [26].

1.2 Shift work and healthcare quality

Shift work has the potential to be detrimental to the healthcare employees who work at night as well as people who are under the care of the shift worker. Decreased levels of alertness and vigilance may increase the likelihood of errors and near errors, which may affect safety [27]. This is dangerous in an industry such as healthcare where errors decrease patient safety. For example, there is evidence to suggest that patients admitted to hospitals at night have worse outcomes. In a recent systematic review and meta-analysis, researchers found that patients who presented to hospitals with acute myocardial infarction during off-hours (nights and weekends) were more likely to die and have longer door-to-balloon times for patients with cardiac abnormalities [28]. In the emergency department, there was no impairment of quality measures (e.g., frequency of return after ED discharge, time to thrombolysis in acute myocardial infarction (AMI), frequency of aspirin use in AMI, and performance of endotracheal intubation) for patients at night, yet patient mortality was higher at night (0.5%; 95% CI, 0.0–1.0%) than during the day [9]. This effect may be related to variability in patient volume, staffing levels, and health system resources that vary with time of day [9].

Long hours, which are often required in shift work, may be associated with increased errors and decreased quality. For example, in a systematic review of 12 studies that examined shift length on quality of patient care, researchers found two studies in which health care providers who worked longer shifts were associated with errors and near-errors [29]. Additionally, in a large sample of 22,275 nurses across four states, researchers found that nurses who work longer than 10 hours were more likely to report that the nursing care quality was fair or poor compared to shorter shifts [30]. In contrast, Stone et al. found no difference in the quality of care on units in which the nurses worked 12 hours compared to 8 hours [31]. In healthcare, more research is needed to understand the optimal length of the shift considering employee preferences and need for continuity of patient care.

Given the well-documented relationship between shift workers and health outcomes, it is important to understand the relationship between shift work and errors. Guided by the Preferred Reporting Items for Systematic Reviews and Meta-analyses [32], a systematic review was conducted to assess the relationship between shift work and errors and performance.

2 Methods

2.1 Search strategy

An extensive literature search was conducted for English-language studies published in the past decade (between January 2003 and December 2013). Electronic databases were searched using MEDLINE/PubMed, EBSCOhost, and CINAHL. The specific search terms included “night shift and professions,” “shift work and accidents,” “shift work and errors”, “shift work and quality”, “night^* and errors” and “shift^* and errors”. The wildcard asterisk (^*) for the search term shift and night was used. After reviewing titles and abstracts, each article was reviewed for relevance. The reference lists of each included article were scanned to identify additional studies. Studies included were full text articles that were written in English, published in peer-reviewed journals, and were not review studies. Additionally, included studies examined the association of shift work with impaired quality (i.e., error rates, accident rates, or productivity outcomes). Excluded studies included research that examined metabolic and psychosocial outcomes and occupational injuries to the shift worker. Additionally, studies that examined sleep outcomes of shift workers without examining associations with quality were also eliminated. The search process was conducted by two researchers (PdC and MB).

2.2 Data abstraction and quality assessment

Data abstraction fields were created to facilitate contrast and comparison of studies. These included the author(s) of the study, year of publication, country of study, and the industry. The sample population, sample size, multisite status, and study design were also extracted because these helped to distinguish quality amongst the articles. Data on the outcomes of interest and whether the studies showed an increased risk for errors or decreased performance were also extracted (Table 1). Outcomes included errors and performance measures. Error was defined as a deviation from an ideal procedure for performing a task, which ultimately leads to a compromised product [33]. Performance was defined as the ability to complete a task accurately as a result of speed and precision influenced by vigilance, concentration, attention, memory, and being well-rested [15].

To evaluate study quality of the quantitative studies, we adapted a measure that included items examining each individual study’s research design, intervention measurement, threats to internal and external validity, statistical analyses (e.g., whether the use of appropriate statistical tests were appropriate), and the presence of power calculations [34]. To evaluate study quality of qualitative work, we adapted criteria that include evaluating the research question, sampling, data collection and analysis, and whether data claims were supported by sufficient evidence [35]. Inter-rater reliability of data abstraction was established by conducting a two-step process where 1) each rater independently scored the articles; and 2) the raters discussed the results of the evaluation. A study with a difference of two points between rater’s scores was discussed. After discussion the raters independently reassessed scores and agreed upon the study quality. Each study was given a quality score calculated as a percentages of applicable items met, which standardized the scores across study design. There was agreement that a percentage score of 80% or better indicated a high quality study.

3 Results

A total of 435 article titles and abstracts were screened; 380 articles were identified from PubMed and CINAHL and the remaining 55 from other sources. Of these, 388 did not meet initial inclusion criteria. Of the remaining 47 articles, 13 were eligible for inclusion (Fig. 1).

3.1 Description of studies

The majority of the 13 studies were conducted in Asia/Middle East (39%) followed by 23% in North America. The remainder was conducted in Europe (n = 2), Australia and New Zealand (n = 2), and one study was conducted in South America. There was a range of designs with a majority (n = 5, 38%) of the studies being interventional. One study used qualitative methods, which included semi-structured interviews. More than half (n = 8) of the studies were single-site studies. Of the variety of fields that these studies embody, the majority, (n = 7, 54%) were of the healthcare industry, about a third (n = 4, 31%) were industrial, one (8%) examined the transportation industry, and one (8%) dealt with safety officers. The studies were divided nearly in half in terms of outcomes, where (n = 7, 54%) studied the effects of shift work on error rates and (n = 6, 46%) studied the effects of shift work on other aspects of performance.

3.2 Quality assessment

Eight studies (seven quantitative and one qualitative study) were rated to be of strong, methodological quality receiving a weighted average score of greater than 75% on the evaluation criteria (Fig. 2). Overall, the strengths of the evidence included clearly stated objectives, clear descriptions of the sample clear description of the main outcome that was measured (excluding the qualitative study), and a thorough description of the main findings. Specifically for the quantitative studies, strengths included the reporting of actual probability values (e.g., 0.035 rather than <0.05) (83%), addressing external validity (61%), and including participants that were representative of the entire population (75%). For internal validity, the majority of the study designs (70%) were appropriate in assessing the main outcome.

Among the five interventional studies, four groups of researchers clearly described the interventions of interest and important adverse events that may have been a consequence that was acknowledged, but not reported [27 , 37]. All five studies provided a description of the compliance to the intervention, and three groups of researchers randomized participants to an intervention group [27 , 38]. However, only one study described the characteristics of the participants that were lost to follow-up [27], and only one study made an attempt to blind study participants [33].

With the four cross-sectional studies, three of the studies provided justification that there was adequate adjustment for confounding in the analyses [39 –41]. These three studies had large sample sizes ranging from 4,279 to 6,445 participants providing sufficient power to test associations between shift work and the quality outcomes.

The most common weakness of the studies included insufficient power or the researchers not providing a justification for sample size and/or a lack of a power analysis. Less than half (46%) of the studies conducted a power analysis to ensure that there would be power to detect a statistically significant effect. Additionally, the study designs chosen limited the ability to infer causality.

3.3 Errors

There were seven studies where researchers examined shift work and errors [27 , 39–43]. All of the investigators examined sleep as an additional outcome and potential contributing factor to committing errors. Four groups of researchers examined hospital providers as the shift worker and errors [27 , 42].

Researchers varied on how they operationalized error rate based on the industry studied as well as the measurement method (e.g., observational, simulation or self-report). Niu et al. [27] used a researcher-scored instrument to measure accuracy and errors of selection attention (d2 test) by completing a test. Hendy et al. [43] conducted a retrospective chart review to identify medical errors, and Gillberg et al. [33] used a simulator of a thermal power plant, which recorded errors. The remaining four research teams assessed errors by self-report questionnaires. For example, in Suzuki et al. [41] errors were recorded by a self-administered questionnaire. Only one of these teams found that self-reported errors among hospital nurses did not increase at night [42]. Additionally, Gillberg et al. [33] found higher sleepiness during the night shift, but with no significant difference in errors or performance of employees between shifts. However, the sample size consisted of only 12 power plant control room operators.

Five groups of researchers found significant relationships between working night shift and increased error rates [27 , 43]. For example, one group found that the error rate on the night shift was 0.44 times more than on the day shift and 0.62 times more than on the evening shift [27]. Additionally, in a subsample of 6,587 police officers who screened positive for sleep disorders related to shift work, 17.9% reported making important administrative errors versus 12.7% of those officers without a sleep disorder (OR = 1.43; 95% CI, 1.23–1.67), and 23.5% versus 15.5% reported making errors or committing safety violations due to fatigue (OR = 1.63; 95% CI, 1.43–1.85) [40].

3.4 Performance

There were six studies that examined shift workers’ performance [37 , 44–47] Overall, the sample sizes of shift workers were small with fewer than 25 shift workers in these studies with the exception of one study that sampled 62 nurses [38]. One of the research teams used qualitative methods and sampled 22 forest shift workers; they found that shift work performance is slower at night due to impaired working conditions [44].

In the five quantitative studies, the operational definition of performance differed. For example, in a study of 62 nurses, researchers operationalized “information processing” as measuring the nurses scores on a digit symbol substitution test and a symbol searching test, which was deemed equivalent as nursing checking medical orders and prescriptions [38]. Purnell et al. [37] operationalized performance by having participants complete a series of information and processing tests, measure reaction time, and complete a vigilance task. All five groups of researchers also operationalized sleep as an outcome and then examined the effect of sleep on performance.

Four teams found that performance was decreased at night [38 , 47]. Chang et al. [38] found that perceptual and motor ability was impaired among shift workers who worked two consecutive night shifts. Using a driving simulator, performance was impaired after working at night and the investigators believed that this translated into diminished performance in the work place [47]. Conversely, in an evaluation of a 24-hour glass factory, Lieber et al. [46] found no difference in productivity rates between shift or groups even when compared to fatigue conditions. Mixed findings came from Purnell et al. [37] who discovered that a 20-minute nap during the night shift led to better performance; however, this was only true for the first night shift worked and not for successivenight shifts.

4 Discussion

In this systematic review there was evidence that night shift is associated with higher error rate and decreased shift worker performance. Although there were weaknesses in the strength of the evidence, (i.e., lack of randomization in 2 the interventional studies) and varying definitions methods of measurement in the operationalization of errors and performance, the overall evidence evaluated was moderately strong. Although the interventional studies were not successful in blinding the shift worker, randomization did occur in three studies. For cross-sectional studies, three of the four studies had sufficient power and appropriately adjusted for confounding. Therefore the strength of the evidence was moderately strong given the limitation of not being able to conduct a trueexperiment.

The majority (n = 9, 75%) of researchers in this systematic review studies found that shift workers committed more errors or had decreased performance compared to their day counterparts. However, these results need to be interpreted with caution. For example, error reporting differed by study design. Self-reporting of errors may be critiqued because shift workers may be less likely to report errors because of fear of repercussions; nevertheless, self-report is a widely accepted method of reporting errors because it may yield data on otherwise unreported and unrevealed errors [41].

Workload and skill of the shift worker have a main role in increased errors and decreased performance at night. The workload is often less at night which may cause some employees to take more chances and perform jobs for which they are unqualified in order to maintain nighttime production [44]. Management may use less qualified staff to fill the staffing gaps. In health care, Glance et al. [48] showed an association that in hospitals where there was a less skilled workforce compared to a more skilled workforce, there was an increased rate of death and injury to patients. Additionally, a similar study by Thompson et al. [49] shows that a higher skill mix greatly improves conditions for the patient. Scheduling lower skilled workers at night is detrimental because they do not have the proper skill set to perform the job that needs to be done correctly, safely, or efficiently. A higher skill mix allows for a more adept approach to performing a job and which can decrease errors.

Poor communication between shift workers during shift change may explain increased errors at night. For example, in one of the studies reviewed, the researchers found Nicholls et al. [45] found that shift workers display 78% of the productivity of daytime workers, attributing this to poor transmission of information at shift change. Transfer of information at shift change can affect the next shift’s work if information is not communicated, skewed, or confusing. Additionally, poor communication with management can lead to errors if important information is left out about the job or if management shows a lack of leadership, guidance, and motivating properties [45]. Thus, workshops on how to improve communication on the job should be set in place [7].

There are several suggestions to improve errors and quality during these shifts. Mitchell et al. [44] recommend longer breaks for night shift workers, including two or more substantial rest breaks evenly spread throughout the shift. It has been suggested that nurses can improve shift-work tolerance by improving daytime sleeping and adjusting rhythm to the same time that they work [50]. Allowing shift workers to have more than two days off when changing over from the night shift to other shifts can provide the shift worker to adjust his or her circadian rhythms [27, 36].

Another recommendation from this review is to encourage shift workers to nap before they start work. The evidence suggesting the efficacy of napping during the shift has been mixed. For example, Purnell et al. [37] suggest that a short nap taken in the workplace may counteract performance deficits, but only during the first night shift in a series of many consecutive night shifts. In health care, Smith-Coggins et al. [51] found that a 40-minute nap at 3:00AM during a 12-hour shift can improve vigilance, memory, and mood. More research is needed regarding napping during night shifts and this intervention may vary in effectiveness based on type or work and employee demographics.

In two of the articles that were reviewed, the authors suggested that managerial recommendations should include providing support for the shift worker including stress-coping programs for personnel [41] and a worker-centered approach in proposing improvements to morale [45]. These suggestions may produce a more holistic approach to improving errors and quality in shift work. Cultural competence of the employee should also be considered because cultural differences impact communication not only in language differences, but also in contextual norms. For example, Brazilians may respond “yes” even when they mean “no,” which negatively impacts relationships between workers and management [46].

There are limitations to our review. We attempted to be exhaustive in our search, however, some articles pertaining to shift work and quality may have been missed. Although we established high inter-rater agreement, quality scores may have been subjective due to the diverse experience level of the researchers. The ability to synthesize a large amount of evidence in a review is an additional limitation because we were unable to present all details for each study that contributed to our conclusions.

There are increased errors and decreased performance at night. This is a concern for the shift worker as well as the people that may be impacted by the work that the shift worker provides. Providers in occupational health should be aware that shift work does not only impede health, but these impediments may contribute to increase errors and decreased performance in the workplace.

References

McMenamin

. A time to work: Recent trends in shift work and flexible schedules. Monthly Labor Review 2007 Available from: http://www.bls.gov/opub/mlr/2007/12/art1full.pdf.

Lee

, McCann

, Messenger

. Working time around the world: Trends in working hours, laws, and policies in a global comparative perspective. Routledge Studies in the Modern World Economy International Labour Office Geneva; 2007.

Biggi

, Consonni

, Galluzzo

, Sogliani

, Costa

. Metabolic syndrome in permanent night workers. Chronobiol Int 2008;25(2):443–454.

Chen

, Skillen

. Promoting personal safety of building service workers: Issues and challenges. AAOHN J 2006;54(6):262–269.

Demerouti

, Geurts

, Bakker

, Euwema

. The impact of shiftwork on work–home conflict, job attitudes and health. Ergon 2004;47(9):987–1002.

Violanti

, Burchfiel

, Hartley

, Mnatsakanova

, Fekedulegn

, Andrew

, et al. Atypical work hours and metabolic syndrome among police officers. Arch Environ Occup Health 2009;64(3):194–201.

Shen

, Dicker . The impacts of shiftwork on employees. Int J Hum Resour Manag 2008;19(2):392–405.

Chen

, Lin

, Hsiao

. Obesity and high blood pressure of 12-hour night shift female clean-room workers. Chronobiol Int 2010;27(2):334–344.

Silbergleit

, Kronick

, Philpott

, Lowell

, Wagner

. Quality of emergency care on the night shift. Acad Emerg Med 2006;13(3):325–330.

10.

Son

, Kong

, Koh

, Kim

, Harma

. Effects of long working hours and the night shift on severe sleepiness among workers with 12-hour shift systems for 5 to 7 consecutive days in the automobile factories of Korea. J Sleep Res 2008;17(4):385–394.

11.

Jay

, Dawson

, Ferguson

, Lamond

. Driver fatigue during extended rail operations. Appl Ergon. 2008;39(5):623–629.

12.

Bonnefond

, Muzet

, Winter-Dill

, Bailloeuil

, Bitouze

, Bonneau

. Innovative working schedule: Introducing one short nap during the night shift. Ergon 2001;44(10):937–945.

13.

Garbarino

, De Carli

, Nobili

, Mascialino

, Squarcia

, Penco

, et al. Sleepiness and sleep disorders in shift workers: A study on a group of italian police officers. Sleep 2002;25(6):648–653.

14.

Sallinen

, Harma

, Mutanen

, Ranta

, Virkkala

, Muller

. Sleep-wake rhythm in an irregular shift system. J Sleep Res 2003;12:103–112.

15.

Kahol

, Leyba

, Deka

, Mayes

, Smith

, et al. Effect of fatigue on psychomotor and cognitive skills. Am J Surg 2008;195(2):195–204.

16.

Harma

, Sallinen

, Ranta

, Mutanen

, Muller

. The effect of an irregular shift system on sleepiness at work in train drivers and railway traffic controllers. J Sleep Res 2002;11:141–151.

17.

Ingre

, Akerstedt

, Peters

, Anund

, Kecklund

. Subjective sleepiness, simulated driving performance and blink duration: Examining individual differences. J Sleep Res 2006;15(1):47–53.

18.

Herman

, Kafoa

, Wainiqolo

, Robinson

, McCaig

, Connor

, et al. Driver sleepiness and risk of motor vehicle crash injuries: A population-based case control study in Fiji. Injury 2014;45(3):586–591.

19.

Costa

, Akerstedt

, Nachreiner

, Baltieri

, Carvalhais

, Folkard

, et al. Flexible working hours, health, and well-being in Europe: Some considerations from a SALTSA project. Chronobiol Int 2004;21(6):831–844.

20.

Wilson

. The impact of shift patterns on healthcare professionals. J Nurs Manag 2002;10(4):211–219.

21.

Cheng

. The Impact of Shiftwork on the Everyday Worker. Unpublished master thesis 2008.

22.

Trinkoff

, Geiger-Brown

, Brady

, Lipscomb

, Muntaner

. How long and how much are nurses now working? Am J Nurs 2006;106(4):60–71.

23.

de Cordova

, Phibbs

, Bartel

, Stone

. Twenty-four/seven: A mixed-method systematic review of the off-shift literature. J Adv Nurs 2012;68(7):1454–1468.

24.

Zhao

, Bogossian

, Turner

. Shift work and work related injuries among health carer workers: A systematic review. Aust J Adv Nurs 2010;27(3):62–74.

25.

Folkard

, Lombardi

, Spencer

. Estimating the circadian rhythm in the risk of occupational injuries and accidents. Chronobiol Int 2006;23(6):1181–1192.

26.

Folkard

, Akerstedt

. Trends in the risk of accidents and injuries and their implications for models of fatigue and performance. Aviat Space Environ Med 2004;75(3):A161–A167.

27.

Niu

, Chu

, Chen

, Chung

, Chang

, Liao

, et al. A comparison of the effects of fixed- and rotating-shift schedules on nursing staff attention levels: A randomized trial. Biol Res Nurs 2013;15(4):443–450.

28.

Sorita

, Ahmed

, Starr

, Thompson

, Reed

, Prokop

, et al. Off-hour presentation and outcomes in patients with acute myocardial infarction: Systematic review and meta-analysis. BMJ 2014;348:f7393.

29.

Estabrooks

, Cummings

, Olivo

, Squires

, Giblin

, Simpson

. Effects of shift length on quality of patient care and health provider outcomes: Systematic review. Qual Saf Health Care 2009;18(3):181–188.

30.

Stimpfel

, Aiken

. Hospital staff nurses’ shift length associated with safety and quality of care. J Nurs Care Qual 2013;28(2):122–129.

31.

Stone

, Du

, Cowell

, Amsterdam

, Helfrich

, Linn

, et al. Comparison of nurse, system and quality patient care outcomes in 8-hour and 12-hour shifts. Med Care 2006;44(12):1099–1106.

32.

Moher

, Liberati

, Tetzlaff

, Altman

. Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. Phys Ther 2009;89(9):873–880.

33.

Gillberg

, Kecklund

, Goransson

, Akerstedt

. Operator performance and signs of sleepiness during day and night work in a simulated thermal power plant. Int J Ind Ergonom 2003;31:101–109.

34.

Downs

, Black

. The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. J Epidemiol Commun H 1998;52:377–384.

35.

Dixon-Woods

, Shaw

, Agarwal

, Smith

. The problem of appraising qualitative research. Qual Saf Health Care 2004;13(3):223–225.

36.

Akerstedt

. Shift work and disturbed sleep/wakefulness. Sleep Med Rev 1998;2(2):117–128.

37.

Purnell

, Feyer

, Herbison

. The impact of a nap opportunity during the night shift on the performance and alertness of 12-h shift workers. J Sleep Res 2002;11(3):219–227.

38.

Chang

, Wu

, Hsu

, Tang

, Yang

, Su

. Impairment of perceptual and motor abilities at the end of a night shift is greater in nurses working fast rotating shifts. Sleep Med 2011;12(9):866–869.

39.

Arimura

, Imai

, Okawa

, Fujimura

, Yamada

. Sleep, mental health status, and medical errors among hospital nurses in Japan. Ind Health 2010;48(6):811–817.

40.

Rajaratnam

, Barger

, Lockley

, Shea

, Wang

, Landrigan

, et al. Sleep disorders, health, and safety in police officers. JAMA 2011;306(23):2567–2578.

41.

Suzuki

, Ohida

, Kaneita

, Yokoyama

, Miyake

, Harano

, et al. Mental health status, shift work, and occupational accidents among hospital nurses in Japan. J Occup Health 2004;46(6):448–454.

42.

Arakawa

, Kanoya

, Sato

. Factors contributing to medical errors and incidents among hospital nurses –nurses’ health, quality of life, and workplace predict medical errors and incidents. Ind Health 2011;49(3):381–388.

43.

Hendey

, Barth

, Soliz

. Overnight and postcall errors in medication orders. Academic emergency medicine: Official journal of the Society for Acad Emerg Med 2005;12(7):629–634.

44.

Mitchell

, Gallagher

, Thomas

. The human factors of implementing shift work in logging operations. J Agric Saf Health 2008;14(4):391–404.

45.

Nicholls

, Bren

, Humphreys

. Harvester productivity and operator fatigue: Working extended hours. Int J Forest Eng 2004.

46.

Lieber

, Kvieska

, Delamaro

. Administrative rationality and coping strategies in shift work. Work 2012;41(Suppl 1):5847–5849.

47.

Akerstedt

, Peters

, Anund

, Kecklund

. Impaired alertness and performance driving home from the night shift: A driving simulator study. J Sleep Res 2005;14(1):17–20.

48.

Glance

, Dick

, Osler

, Mukamel

, Li

, Stone

. The association between nurse staffing and hospital outcomes in injured patients. BMC Health Serv Res 2012;12:247.

49.

Thompson

, Hsu

, Chang

, Martsteller

. Impact of nursing staffing on patient outcomes in intensive care unit. Journal of Nursing & Care 2013;2(2):1–6.

50.

Berger

, Hobbs

. Impact of shift work on the health and safety of nurses and patients. Clin J Oncol Nurs 2006;10(4):465–471.

51.

Smith-Coggins

, Howard

, Mac

, Wang

, Kwan

, Rosekind

, et al. Improving alertness and performance in emergency department physicians and nurses: The use of planned naps. Ann Emerg Med 2006;48(5):596–604, e1–3.