1-s2.0-S0020748905000477-main

International Journal of Nursing Studies 43 (2006) 237–263

Nurse turnover: A literature review

Laureen J. Hayesa,!, Linda O’Brien-Pallasa, Christine Duffieldb, Judith Shamianc,James Buchand, Frances Hughese,f, Heather K. Spence Laschingerg,

Nicola Northf, Patricia W. Stoneh

aFaculty of Nursing, University of Toronto, Toronto, Ont., Canada M5S 3H4bCentre for Health Services Management, University of Technology, Sydney, P.O. Box 123 Broadway, NSW 2007, Australia

cVictorian Order of Nurses, Ottawa, Ont., Canada K2P 1B4dFaculty of Health and Social Sciences, Queen Margaret University College, Edinburgh, EH12 8TS, Scotland

eMinistry of Health, P.O. Box 5013, Wellington, New ZealandfSchool of Nursing, Faculty of Medical and Health Sciences, University of Auckland, Private Bag 92 019, Auckland, New Zealand

gSchool of Nursing, University of Western Ontario, London, Ont., Canada N6A 5C1hSchool of Nursing, Columbia University, New York, NY, 10032, USA

Received 25 October 2004; accepted 1 February 2005

Abstract

Ongoing instability in the nursing workforce is raising questions globally about the issue of nurse turnover. Acomprehensive literature review was undertaken to examine the current state of knowledge about the scope of the nurseturnover problem, definitions of turnover, factors considered to be determinants of nurse turnover, turnover costs andthe impact of turnover on patient, and nurse and system outcomes. Much of the research to date has focused onturnover determinants, and recent studies have provided cost estimations at the organizational level. Further research isneeded to examine the impact of turnover on health system cost, and how nurse turnover influences patient and nurseoutcomes.r 2005 Elsevier Ltd. All rights reserved.

Keywords: Nurse turnover; Job satisfaction; Nurse shortage; Job turnover literature; Literature review

1. Introduction

High nurse turnover can impact negatively on anorganization’s capacity to meet patient needs andprovide quality care (Gray and Phillips, 1996; Tai etal., 1998; Shields and Ward, 2001). At the nursing unitlevel, high turnover affects the morale of nurses and theproductivity of those who remain to provide care whilenew staff members are hired and orientated (Cavanaghand Coffin, 1992; Sofer, 1995). Ongoing workforce

instability in many countries is raising questionsabout the impact of nurse turnover on the well-beingof nurses, quality of patient care and system costs.Stated reasons for leaving that were common tonurses in the USA, Canada, England, Scotland andGermany include emotional exhaustion and problems inwork design (Aiken et al., 2001). While numerousstudies from several disciplines have been under-taken to better understand turnover behavior, fur-ther research is needed to address how turnoverwithin the skilled nursing workforce impacts thecapacity of health organizations to respond to demandsfor health care.

ARTICLE IN PRESS

www.elsevier.com/locate/ijnurstu

0020-7489/$ - see front matter r 2005 Elsevier Ltd. All rights reserved.doi:10.1016/j.ijnurstu.2005.02.007

!Corresponding author.E-mail address: [email protected] (L.J. Hayes).

vasanth kumar

vasanth kumar

vasanth kumar

www.elsevier.com/locate/ijnurstu

The past literature reviews and meta-analytic studieshave been conducted to review factors associated withnurse turnover. Irvine and Evans’s (1995) meta-analysisrevealed a strong positive relationship between beha-vioral intentions and turnover, a strong negativerelationship between job satisfaction and behavioralintentions, and a small negative relationship between jobsatisfaction and turnover. Turnover predictors wereidentified in a literature review by Tai et al. (1998) asage, tenure, job satisfaction, organizational commit-ment, perceived job possibilities and supervisor’s beha-vior. Similarly, Yin and Yang’s (2002) analysis revealedthat internal environmental factors such as stressresulting from staffing shortages, leadership style, super-visory relations, advancement opportunities and inflex-ible administrative policies were significantly related toturnover. For this paper, a comprehensive literaturereview was undertaken to examine the current state ofknowledge about the scope of the nurse turnoverproblem, definitions of turnover, factors considered tobe determinants of nurse turnover, turnover costs and ofmost importance to the authors, the impact of turnoveron patient, nurse and system outcomes. This reviewfocuses on the recent research, also identifying metho-dological challenges and implications for further study.

2. Methodological approach

Following the method for systematic review outlinedby Cooper (1998), the following electronic databaseswere broadly scanned prior to more refined searches:Health Sciences and Health Administration (CINAHL,MEDLINE, PubMed, EMBASE, HealthSTAR), Eco-nomics (EconoLit), Psychology (PsycInfo, PsycARTI-CLES) and Sociology (ASSIA, FRANCIS). Keywordsto search the literature included nurse turnover, nurseretention, staffing shortage, job satisfaction, nursevacancy, job stress, nursing leadership, nurse staffingand patient outcomes. As some publications might notbe covered by electronic searches or studies may havebeen indexed inaccurately (Droogan and Song, 1996),reference lists of articles were scanned for additionalitems and new releases of key journals were individuallyreviewed for recently published studies. Internetsearches of professional organization and governmentwww-sites were conducted, producing unpublishedresearch studies, discussion papers, media releases andaction reports relating to nurse staffing issues.Almost 200 documents were reviewed, of which 130

were used in this report. The selection was restrictedprimarily to the following studies: (1) published in thelate-1990s or afterward with the exception of the earlytheoretical models; (2) those written in English; and (3)examining turnover or turnover intention in employeepopulations of registered or practical/enrolled nurses

working in the hospital, long-term or community careareas. The primary intent of the review was to examinethe research in terms of how the evidence has advancedthe knowledge in this area. Using predominant studiesthat had undergone peer-review assured a high level ofquality, thereby promoting validity of the overallfindings and conclusions.From the body of research used in this report, a

sample of 37 studies reporting measures of turnover (orturnover intent) as a variable are highlighted in tablesprovided in the Appendices. The research objective(s),samples selection, data collection and analysis methods,and key findings are outlined. Wherever possible, theresearch instrument and its author(s) were stated (referto the respective articles for reference listings of theseauthors). Appendix A provides information about 32studies that examine determinants of nurse turnover,and Appendix B provides information about 5 studiesthat examine turnover consequences. The publicationyears are between 1991 and 2004, the majority being inthe past 5 years. The tables were developed in order toshow the progression of research methodologies and theknowledge advancement, in particular, that the body ofresearch represented in Appendix A (turnover determi-nants) is more developed than that represented inAppendix B (turnover consequences). This paper isorganized by definitions of turnover found in theliterature, theoretical models of turnover found in theliterature, review of evidence on determinant of nurseturnover, review of evidence on consequences of nurseturnover, and a discussion of the methodologicalchallenges in this body of evidence.

3. Turnover definition

Methodological challenges have plagued researcherswhen attempting to measure and compare turnoveracross diverse health care systems. Even at the locallevel, the lack of consistency in how records of turnoverare maintained presents difficulties, as the reliability ofturnover determinations varies according to record-keeping methods (Tai et al., 1998). The definitions ofturnover and accuracy of the reason for turnover areoften inconsistent, making it difficult to compare orgeneralize across studies (Tai et al., 1998 provide asummary table of turnover definitions and measures).Jones (1990a, b) defined nursing turnover as the processwhereby nursing staff leave or transfer within thehospital environment. This definition encompassesvoluntary and involuntary, as well as internal andexternal turnovers. Voluntary and involuntary turnoversare not always differentiated in studies because costs areincurred regardless of whether staff resign or arerequested to leave. Some studies define turnover as anyjob move while others consider nurse turnover as leaving

ARTICLE IN PRESSL.J. Hayes et al. / International Journal of Nursing Studies 43 (2006) 237–263238

vasanth kumar

vasanth kumar

vasanth kumar

the organization or even from the nursing profession.Mano-Negrin and Kirschenbaum (1999) suggest thatturnover reflects the effect of the balance betweenorganizational benefits (pull factors) and a careeristicattitude to work (push factors). Turnover behavior maybe counteracted by career aspiration if expectations foradvancement kept them in organization, added to byfear of unemployment.When considering the definition of turnover, it is

important to acknowledge that some renewal of nursingpersonnel can be viewed as beneficial to an organization.Nurse turnover offers opportunity for cost reductionwith decreased salaries, benefit costs and vacation payfor new employees (Jones, 1990a). Productivity couldincrease by providing better matches as employees moveto jobs where their performance will be greatest,contributing to an efficient labor market (Gray et al.,1996). Potential gains difficult to quantify are due toinfusion of knowledge and ideas by replacements, thestimulation of policy changes and the decreased exhibi-tion of withdrawal behaviors (Jones, 1990a). How-ever, as turnover reaches 50%, its net effect onproductivity is probably negative (Price and Mueller,1981a, b) as employees that the organization prefers toretain are lost.

4. Theoretical models of nurse turnover

Past theoretical models have synthesized turnoverresearch and specified relationships among determinantsof turnover (Price and Mueller, 1981a, b; Mobley, 1982;Hinshaw and Atwood, 1983; Abelson, 1986; Parasura-man, 1989). Price and Mueller (1981a, b) view turnoveras a product of job satisfaction and commitment, whichin turn are influenced by organizational factors, demo-graphics, and environmental factors such as alternativejob opportunity outside the organization. In Hinshawand Atwood’s (1983) model, anticipated and actualturnovers amongst nurses are determined by two typesof job satisfaction: organizational, relating to groupcohesion, job stress, and control over decisions; andprofessional, the nurse’s perception of the quality ofcare, time to do one’s job, and enjoyment derived fromit. Parasuraman (1989) also assessed the role of time lagin determining relationships between variables foundconsistently to explain and predict turnover acrossstudies, confirming that intention to leave was the mostimmediate determinant of actual turnover. Personal andorganizational variables were related to the intentionto leave and actual turnover only indirectly throughtheir effects on felt stress, job satisfaction, andorganizational commitment. Furthermore, as hypothe-sized, the strength of the intention-turnover relationshipdecreased as the time interval between expressedintentions and turnover behavior increased. While the

early models provided a well-developed body of researchfrom which to further examine the issue of turnover(Hinshaw and Atwood, 1983), limitations includedinadequate sample sizes, lack of generalizability, notcontrolling simultaneously for the influence of multiplevariables, cross-sectional nature of the data and the needto employ time series analysis, and the focus onindividuals’ data (Alexander, 1988; Cavanagh, 1989).The past decade has seen further modeling work on

nurse turnover. Irvine and Evans (1995) presented amodel based on Mueller and Price’s (1990) theory thatdifferent disciplinary perspectives contribute to explain-ing nurse turnover: economists who emphasize indivi-dual choice and labor market variables; sociologists,who emphasize the structural characteristics of the workenvironment and work content; and psychologists, whoemphasize individual variables and intra-psychic pro-cesses. Krausz et al. (1995) examined progressive with-drawal, testing the conceptualization that a nurse firstdecides to leave the ward, then the hospital and, finally,the profession. The researchers felt that previous studieswere narrow in scope, ignoring within-organizationturnover from one unit to another and, when relevant,turnover from the profession. Their hypothesizedprogression model was supported. Although the inten-tion to leave the profession was better predicted than theintention to leave the ward or the hospital, thecumulative effects of the lower two levels of withdrawalintention were its major determinants. Alexander et al.(1998) theorized that work environment affects groupsof nurses differently, conceptualizing turnover as amultistage process linking social and experiential or-ientations, attitudes toward the job, the decision to quit,and the behavior of actually quitting. As in othermodels, intention to quit was the strongest directpredictor of turnover.

5. Investigation of nurse turnover determinants

Considerable attention has been committed to under-standing organizational, individual and economicalfactors that influence turnover behavior. Administrativestrategies and organizational climate, in particular, havebeen investigated as to their effect on job satisfaction,intent to leave and actual turnover behavior.

5.1. Job satisfaction and nurse turnover

There is a vast body of the literature linkingjob satisfaction in nursing and turnover. Job satisfac-tion instruments have been developed, notably thescale by Mueller and McCloskey (1990) that includeseight satisfaction factors; extrinsic rewards, scheduling,family/work balance, co-workers, interaction, profes-sional opportunities, praise/recognition, and control/

ARTICLE IN PRESSL.J. Hayes et al. / International Journal of Nursing Studies 43 (2006) 237–263 239

responsibility. Job dissatisfaction has been frequentlyidentified as the reason why nurses leave their jobs (Lumet al., 1998; Tzeng, 2002). Using data from a nationalsurvey of National Health Service (NHS) nursing staff,Shields and Ward (2001) found job satisfaction to bemore important than the attraction of outside opportu-nities. Low job satisfaction was concentrated in young,newly qualified and highly educated nurses, andassociated with administration, promotional prospects,employment security and amount of time for clinicalduties.

5.2. Moderators of nurse turnover

Some studies suggest that certain moderators apartfrom external, personal and work-related variables,influence job satisfaction, turnover intention and turn-over behavior. Moore (2001) found that a sense ofprofessionalism mediated intention to quit despite theimpact of restructuring changes on hospital and nurseconditions, poor management and communication style,and burnout. Similarly, Angerami et al. (2000) deter-mined that nurses’ motives to remain in their jobsrelated to attachment to nursing, even though their workwas not recognized and they were poorly paid. Careercommitment might not, however, have the samemediating effect when nurses first start working. Basedon findings of a weak association between careercommitment and turnover intention in new nurses froma longitudinal survey, Gardner (1992) suggested thatcareer commitment is not a stable phenomenon in thefirst year, with greater susceptibility to organizationalfactors. As initial professional commitment is basedpartially on individual socialization into the profession(Blau and Lunz, 1998), more research examining theinfluence of nursing programs on career attitudes mayprovide insight into possible educational interventionsthat instill professional loyalty.Professional commitment has been shown to be more

effective in predicting intention to leave nursing thanintention to leave the organization (Lu et al., 2002).Chang (1999) suggested in a non-nursing study thatcareer commitment is distinct from organizationalcommitment, as individuals who are committed to boththeir career and their organization are even less willingto leave the company than those who are committedmainly to the organization. Likewise, Blau and Lunz,(1998) found in a longitudinal sample of medicaltechnologists, that professional commitment accountsfor significant additional variance in intent to leave theprofession after controlling for external, personal andwork-related variables. The findings support the posi-tion that career commitment is a more stable type ofwork commitment that can transcend situational andpersonal influences, thus supporting the role of continu-ing education and professional development opportu-

nities in nurse retention. In a recent study, Holtom andO’Neill (2004) examine a new construct, job embedded-ness, which focuses on the accumulated reasons why aperson stays in a job; being embedded in an organizationis associated with reduced intent to leave and actualleaving. Based on data from a sample of 232 employeesof US community-based hospital, job embeddednessassessed new variance in turnover in excess of thatpredicted by the variables in the major models ofturnover.Personal disposition has also been found to moderate

the relationship between job satisfaction and voluntaryturnover. Based on data collected from a sample ofnurses and other workers in a medical clinic, Judge(1993) found that job satisfaction and voluntary turn-over were more highly related for employees withpositive dispositions than for employees with negativedispositions, suggesting that individuals in a positiveframe of mind may take proactive steps such as quittingtheir jobs. Similarly, George and Jones (1996) deter-mined that the relationship between job satisfaction andturnover intention was stronger when positive mood washigh, consistent with research suggesting that people inpositive moods tend to have higher levels of self-efficacyand are more optimistic and action-oriented than peoplewho are not.

5.3. Organizational factors and nurse turnover

Much of the nurse turnover research explores howturnover behavior is influenced by organizationalcharacteristics associated with workload, managementstyle, empowerment and autonomy, promotional op-portunities and work schedules. Although increasingrecruitment of nurses and improved compensation mayhelp offset nurse shortages in the short term, researcherssuggest that administrative interventions to improvequality of work life are more effective long-term inreducing turnover (Bloom et al., 1992; Shields andWard, 2001; Gifford et al., 2002). Measurement oforganizational attributes that characterize professionalnursing practice has been advanced with the use of theRevised Nursing Work Index (NWI-R) (Aiken andPatrician, 2000), although further work is needed todetermine predictive capability and relevance to organi-zations across different nations (Estabrooks et al., 2002).

5.3.1. Workload, stress and burnoutA consistently heavy workload increases job tension

and decreases job satisfactions, which in turn, increasethe likelihood of turnover (Davidson et al., 1997; Tai etal., 1998; Hemingway and Smith, 1999; Strachota et al.,2003). Empirical evidence suggests that each additionalpatient per nurse is associated with a 23% increase inthe odds of burnout and a 15% increase in the oddsof job dissatisfaction (Aiken et al., 2002b). Workload


measurement has progressed over the past decade toreflect the variety of factors impacting a nurse’s workday. O’Brien-Pallas (1988) had emphasized that work-load measurement should focus on the complexity of thepatient situation rather than on the actual tasks. Amongvariables relating to nurse, patient and work environ-ment, the least studied are those involving schedulingand coordination of procedures, multiple and longprocedures, and characteristics and composition of thecaregiver team (O’Brien-Pallas et al., 1997). The practiceenvironment impacts work content that in turn,determines intrinsic work motivation (Janssen et al.,1999). While findings show that work overload exists, abetter understanding of the unpredictable nature ofnurses’ direct working environments, such as in the caseof frequent unforeseen patient events, would promote agreater appreciation of why nurses are frustrated.Some studies relating to work stress focus on a specific

unit type. Barrett and Yates, (2002) using a conveniencesample of oncology nurses, found nearly 40% weredealing with excessive workloads, 48% were dissatisfiedregarding pay, 70% experienced emotional exhaustionand 48% of the sample could not commit to remainingin the specialty for a further 12 months. Based oninterview data, Cartledge (2001) found work-relatedstress to be a major contributor to nurse turnover incritical care units. Cameron et al. (1994) conductedcomparative analyses by unit type, concluding that RNsin psychiatric settings were least satisfied with their jobs,more likely to leave their positions and reported moreburnout than did RNs in other settings. Other studiesexamine the effect of patient type on staff burnout(Evers et al., 2002; Molassiotis, and Haberman, 1996)and job satisfaction (Dougherty et al., 1992). In a studyof burnout in staff caring for the elderly, Evers et al.(2002) used hierarchical regression analysis to determinethat physical and psychological aggression and numberof weekly working hours have a significant relationshipwith emotional exhaustion, and psychological aggres-sion has a significant relationship with depersonaliza-tion. Perceived risk of assault also impacts turnoverbehavior, in one study predicting psychiatric nurses’intention to leave their current job (Ito et al., 2001).

5.3.2. Management styleResearch has demonstrated the importance of nursing

leadership to job satisfaction. Bratt et al. (2000) exploredthe effect of nurse attributes, unit characteristics andwork environment on job satisfaction of nurses inpaediatric critical care units, finding job stress andnursing leadership to be most influential. Leadershipthat values staff contribution promotes retention,evidenced by consistent themes in the literature relatingto autonomy, good working relationships and amanagement style that facilitates rather than directs.Magnet hospital research in the USA has contributed to

a better understanding of nursing practice withinhealthcare organizations (Scott et al., 1999). Supporthas been shown for the positive effects of a decentralizedorganizational structure, a commitment to flexibleworking hours, an emphasis on professional autonomy,and communication between management and staff,resulting in lower nurse turnover and higher levels of jobsatisfaction than other hospitals (Aiken et al., 1994;Buchan, 1994).Some studies refer to a participative management

style as enhancing job satisfaction (Jones et al., 1993;Nakata and Saylor, 1994; Moss and Rowles, 1997;Yeatts and Seward, 2000). Song et al. (1997) foundgreater satisfaction with supervisory dimensions in anurse-managed special care unit (SCU) with sharedgovernance and minimal technology than a traditionalintensive-care unit (ICU) characterized as being bureau-cratic and high-tech; nurses working in both the SCUand the ICUs preferred the SCU practice model over theICU model. Boyle et al. (1999) examined the effects ofnurse managers’ characteristics of power, influence andleadership style, finding managers’ position power andinfluence over work coordination to have a direct link tointent to stay in employment. Instrumental communica-tion, autonomy and group cohesion decreased job stressand increased job satisfaction, which in turn, weredirectly linked with intent to stay. Likewise, Leveck andJones (1996) found that management style, groupcohesion, job stress, organizational and professionaljob satisfaction influenced staff nurse retention andquality of care. In support of previous studies arefindings of a recently published study showing thatreward-based climates, high levels of communicationopenness and accuracy explained lower turnover innursing homes (Anderson et al., 2004).

5.3.3. Empowerment and autonomyStructural empowerment is the perception of the

presence or absence of empowering conditions in theworkplace, psychological empowerment is the employ-ees’ psychological interpretation or reaction to theseconditions (Laschinger et al., 2004). Although studieshave demonstrated that empowerment is associated withjob satisfaction, a direct link with nurse turnover wasnot determined. Larrabee et al. (2003) concluded thatthe major predictor of intent to leave was jobdissatisfaction and the major predictor of job satisfac-tion was psychological empowerment. Laschinger et al.(2004) used a longitudinal predictive design to test amodel linking changes in structural empowerment(opportunity, information, support, resources, formalpower, and informal power) and psychological empow-erment (meaningful work, competence, autonomy, andimpact) to changes in job satisfaction. Changes inperceived structural empowerment had direct effectson changes in psychological empowerment and job


satisfaction. Changes in psychological empowerment didnot explain additional variance in job satisfactionbeyond that explained by structural empowerment.The results suggest that fostering environments thatenhance perceptions of empowerment can have enduringpositive effects on employees.Kramer and Schmalenberg (2003) quantified nurse

autonomy and determined a strong relationship betweendegree of autonomy and rankings of job satisfaction andquality of care. Rafferty et al. (2001) found autonomy,control over resources, relationships with doctors,emotional exhaustion and decision-making to be corre-lated with one another as well as having a relationshipwith nurse-assessed quality of care and nurse satisfac-tion. These findings are consistent with those ofLaschinger et al. (2001) whose survey of hospital staffnurses suggested that perceived autonomy, control andphysician relationships influence the trust, job satisfac-tion, and perceived quality of patient care.

5.3.4. Promotional opportunitiesCareer development and life-long learning activities in

nursing promote job satisfaction, increased retention ofnurses and enable continued provision of high-qualitycare (Yoder, 1995; Kennington, 1999; Donner andWheeler, 2001; Davidson et al., 1997; Collins et al.,2000). Dissatisfaction with promotion and trainingopportunities has been shown to have a stronger impacton nurse turnover than workload or pay (Shields andWard, 2001). Multivariate analysis of data collected byDavidson et al. (1997) in a longitudinal survey of nursesin one hospital indicated that predictors of intent toleave were the perception of little promotional oppor-tunity, high routinization, low decision latitude andpoor communication. Perceived interest in one’s careerdevelopment and feelings of being valued influencenurses’ intent to stay (Yoder, 1995). Similarly, Szigeti etal. (1991) used correlational analysis to investigate thepotential factors that relate to the desire of registerednurses (RNs) and licensed practical nurses (LPNs) tocontinue practicing in rural hospitals. Overall jobsatisfaction and performance constraints were the onlyvariables to make significant contributions to theprediction of turnover intention for both RNs andLPNs. Satisfaction with promotion was the only work-related variable to make a significant contribution to theprediction of turnover intention for RNs. Performanceconstraints, role ambiguity and shift work were the onlywork-related variables contributing to the prediction ofturnover for LPNs.

5.3.5. Work schedulesTo promote balance between work and home,

potential benefits of self-scheduling strategies have beendocumented, especially for nurses who have homeresponsibilities such as young children (Kane and

Kartha, 1992; Teahan, 1998; Kane, 1999; Hung, 2002).Long shifts, overtime, weekends, nights, holidays andweekend overtime were found to be predictors ofanticipated turnover (Shader et al., 2001; Strachota etal., 2003). Vetter et al. (2001) described a process of self-scheduling that met the needs of staff, promoted highjob satisfaction and maintained staffing standards on theunit. Using cross-tabulations when comparing well-being in nurses with young children at home, nursesemployed full-time, casual part-time, or those with job-sharing positions, Kane and Kartha (1992) found thatjob-sharing nurses had the highest ratings for jobsatisfaction and physical health status. Kane (1999)used analysis of variance to determine that job sharinghas a positive impact on job satisfaction and retentionwhen comparing differences among full-time, part-timeor job-sharing nurses in a large Canadian teachinghospital. These findings were supported in a qualitativestudy by Durand and Randhawa (2002) in which flexibleworking practices, increased salaries and demonstratingvalue to staff impacted career-break nurses’ decisionswhether they would return to practice in the NHS.Generally, to summarize the literature relating to the

impact of organizational factors on nurse turnover, todate there appears to be substantial evidence to informdecision-makers of the type of management style andworkload factors that promote job satisfaction innurses, as findings have been relatively consistent acrossvarious specialty unit types and jurisdictions. However,less well known is the impact of environmental complex-ities on nurses’ well- being and overall unit functioning,as well as the impact of unit, patient and nursecharacteristics on their work environments. Moreevidence to determine the interdependence of theserelationships would help decision-makers better under-stand how intervention directed to one component ofthe patient care system has an impact on the system aswhole.

5.4. Individual factors and turnover

Certain socio-demographic characteristics of nursespredispose to turnover, but are not usually consideredexplanatory variables in turnover behavior (Tai et al.,1998). An inverse relationship between age and turnoverhas been demonstrated for many years (Lowery andJacobsen, 1984; Parasuraman, 1989; Gray and Phillips,1994; Kiyak et al., 1997). McNeese-Smith and vanServellen (2000) suggest that mature nurses have greaterjob satisfaction, productivity and organizational com-mitment. Shader et al. (2001) found anticipated turnoverfor younger nurses to be associated with job satisfactionand stress, but no significant predictors of turnover orstress in nurses over 50, possibly due to their proximityto retirement, their marketability related to experience,and the fact that only those who are satisfied remain


working in a particular nursing environment. Contraryto other studies, Chan and Morrison (2000) found nostatistically significant difference in the proportion ofstayers and leavers on several demographic factorsincluding age.In terms of work experience, less experienced nurses

tend to be younger, participate less in decision-makingand have fewer home responsibilities (Price and Mueller,1981a, b), while those with more experience are moresatisfied with pay and less likely to leave (Lum et al.,1998). What remains unclear is whether it is workexperience that it is related to turnover, or that age,work experience and tenure are inextricably linked(McCarthy et al., 2002). The relationship between lengthof service (tenure) and turnover compared to thatbetween experience and turnover is said to be morecomplex. Gray and Phillips (1994) indicate that turnoverrates tend to be high in the first year of service andremain high, or even rise during the second year ofservice before declining. Fewer years of employment wasrelated to termination (Davidson et al., 1997; Kiyaket al., 1997) and RNs with more years of workexperience reported higher job satisfaction, lowest levelsof burnout and were less likely to leave their positions(Cameron et al., 1994). Lane et al. (1990) suggested thattenure as a correlate of turnover is inappropriate forcertain groups of nurses. While part-time unmarriednurses were more likely to leave their hospitals than full-time unmarried nurses, tenure was not found to be asignificant predictor of turnover for married nurses.Kinship responsibilities involve home obligations.

Children, spouses and aging parents affect the workand turnover habits of nurses, possibly requiring achange in work environment (Price and Mueller,1981a, b; Cavanagh, 1989). Bloom et al. (1992) pointout that nurse turnover in the USA during the 1960s wasattributed primarily to women leaving during child-bearing years. Tzeng (2002) found that age of youngestchild was a significant predictor of nurses’ intention toquit. Strachota et al. (2003) surveyed 84 nurses who hadvoluntarily terminated or changed their employmentstatus, 19 of whom indicated the reason as family,having to stay at home with children or elderly parents.In a study of public health nurses serving rural and smallurban areas, Henderson Betkus and MacLeod (2004)found that factors such as age, retirement, family needsand the economy affected intent to stay or leave, ratherthan job satisfaction or community satisfaction.Educational level is believed to impact turnover in

that more highly educated individuals are more likely toquit in order to seek career advancement, especially ifthere are limited opportunities in their current organiza-tion (Tai et al., 1998; Yin and Yang, 2002). Severalstudies suggest a relationship between educationalattainment, specifically the possession of a degree, andturnover (Krausz et al., 1995; Cavanagh and Coffin,

1992). A higher level of nursing education may alsoimply more varied placements during their preparationand development of a sense of loyalty to their professionrather than a particular health care service (Lane et al.,1990). Given the recent requirement of a degree topractice nursing in many countries, further researchshould examine the effects of a more highly educatednursing workforce on turnover rates. Aiken et al. (2004,p.70) suggest that ‘‘sustained underinvestment in nursingeducation is a theme across countries that are nowturning to aggressive international recruitment’’ thusincreasing nursing migration. This research highlightsthat increased investment in nursing education is key tocountries reducing their reliance on overseas nursingexpertise, particularly from developing countries.

5.5. Economic factors and nurse turnover

Studies that include remuneration as one componentof job satisfaction are inconsistent in their findings. Forexample pay does not have as strong an impact as workenvironment (Irvine and Evans, 1995), pay is not a highpriority (Frisina et al., 1988) and pay is not associatedwith turnover (Michaels and Spector, 1982; Mobleyet al., 1979; Borda and Norman, 1997). Shields andWard (2001) found that dissatisfaction with promotionhad a stronger impact than pay on intention to quit theNHS, therefore improved pay would only have limitedsuccess unless accompanied by improved opportunities.Other research suggests that pay can have both a directand indirect effect on turnover intent (Lum et al., 1998;Chan and Morrison, 2000; Tzeng, 2002; Yin and Yang,2002; Strachota et al., 2003). Lum et al. (1998) foundthat pay satisfaction correlated strongly with reducedturnover intent, but also had a weaker correlationmediated through job satisfaction. Lu et al. (2002)presented a matrix of correlation coefficients betweenvariables that indicated a significant negative correlationbetween turnover intent and wage, as well as a positivecorrelation between wage and professional commitment.There is scant and inconsistent evidence to inform us

whether alternative employment opportunity is adeterminant of nurse turnover. Price and Mueller(1981a, b) asserted that when jobs are plentiful, turnoveris high and when jobs are scarce, turnover is low. Krauszet al. (1995) found that even where many alternative jobopportunities exist, many nurses prefer an internalmove. Bloom et al. (1992) claimed that turnover isrelated to alternative employment opportunities only inhigher population areas. Yet in a study by Strachota etal. (2003), of 84 nurses who had voluntarily terminatedor changed their employment status, 31 indicated betterjob opportunity for more money and better hours asreasons for leaving. A recent study by Duffield et al.(2004) of nurses who already left the profession foundthat many moved to management outside health


industry with most taking additional study after leaving.Further research would provide insight into the natureof circumstances that lead to alternative employment,whether in or outside nursing.

5.6. Summary of factors associated with nurse turnover

In summarizing the factors associated with nurseturnover, job dissatisfaction and expressed intent toleave are most consistently reported as impactingturnover. Job satisfaction, turnover intention and turn-over behavior appear also to be influenced by certainmoderators including professional commitment andpersonal disposition. Organizational characteristics as-sociated with workload, management style, empower-ment and autonomy, promotional opportunitiesand work schedules are believed to contribute toturnover, and therefore researchers suggest thatadministrative interventions to improve quality of worklife are imperative for long-term resolution. Thesefindings are reflected in the studies described inAppendix A.While socio-demographic characteristics are not

usually considered explanatory in turnover behavior,and related findings have been somewhat inconsistent,factors such as younger age, inexperience and feweryears of the job, higher educational level, and kinshipresponsibilities are claimed to predispose nurses toturnover. In terms of economic factors, findings relatingto the impact of remuneration on turnover are varied,and there is little evidence as to whether alternativeemployment opportunity is a determinant of nurseturnover.

6. Consequences of nurse turnover

6.1. Economic impact

Inconsistent turnover definitions and measures haveimpeded a clear understanding of associated financialcosts, as replacement cost estimations vary according tolocation and components included in the measure.Direct costs are those incurred during the hiring process,such as advertising, recruiting, agency nurses and hiring.Indirect costs are due to RN termination, orientationand training, and decreased RN productivity. Jones(1990a, b) is supported by others (Gray et al., 1996;Johnson and Buelow, 2003) in pointing out that indirectcosts of nursing turnover could be significant because ofthe decreased initial productivity of new employees andthe decrease in staff morale and group productivity thatnursing turnover imposes. The health care providerswho switch employers or leave health care entirely alsosuffer, as professional disillusionment is a major andaccelerating problem (Waldman et al., 2004).

Jones (1990b) acknowledges that high nurse turn-over adversely impacts the nursing department, hospi-tal environment and healthcare system; however, nostudies were found that considered both incrementalstaffing costs and health care system costs. Previousresearchers have attempted to determine financialcosts of replacing individual nurses. Estimations rangefrom $10,000 to $60,000 per RN, depending on thenurse specialty (Jones, 1990b; Johnson and Buelow,2003 citing Curran, 1991). Likewise, Strachota et al.(2003) cited estimates (from The Advisory BoardCompany, 2000) of $42,000 to replace a medical-surgicalnurse and $64,000 for a specialty nurse. These figuresincluded the cost of recruitment, orientation, preceptingand lost productivity, the latter claimed to be nearly80% of the total turnover cost. While it has beenrecognized that the cost associated with lower produc-tivity of new hires is a significant component ofturnover, Waldman et al. (2004) state that this costhas not been quantified before and therefore has notappeared as a recognized budget expense. They indicatethat turnover cost models have typically omitted thismeasure, which requires calculations using learningcurve algorithms and retention methodologies. In theirapplication of rigorous accounting methodology tosix groups of health care workers, including nurses,the largest cost driver was the loss and necessaryreplacement of nurses. The total cost for a newly hirednurse averaged $15,825 and the cost of reducedproductivity ranged from $5,245 to $16,102 (Waldmanet al., 2004).The economic impact at a country level of this

turnover and ultimately nurse migration is now a topicof global discussion but little research and publicationsto date have occurred. In 1998 the United NationsConference for Trade and development estimated thatevery professional, from ages 25 to 35, who migratedfrom South Africa represented an annual loss of$184,000 for that country (United Nations Conferenceon Trade and Development, 1998). Interestingly, factorsthat are now found in regards to health professionals’decisions to emigrate are very similar to those ofturnover. A study of five African countries found themain factors causing health professionals to emigrate tobe lack of further training opportunities, and poorremuneration, working environment and health man-agement systems (International Organisation for Migra-tion, 2002 (IOM), 2002). The 2003 World Health Reportdescribes how the African countries’ ability to respondagainst HIV is undermined primarily by lack of skilledworkforce. The secretariat paper to the 2004 WorldHealth Assembly in Geneva describes how the turnoverof health professionals and ultimate migration, increasesnot only the workload of those who remain butultimately the capacity of country health services torespond. ‘‘Migration flow of health professionals


(largely nurses) account for 65% of economically activemigrants’’ (WHO, 2004 p.1).

6.2. Nurse outcomes

While there is increasing attention on the impact ofinadequate nurse staffing and poor work environmentson nurse health and safety outcomes, few studies focusspecifically on turnover to determine its effect on nurseoutcomes. Inadequate staffing levels and work overloadhave been studied (Baumann et al., 2001; O’Brien-Pallaset al., 2001; Shamian and O’Brien-Pallas, 2001) andassociated negative effects on nurses’ well-being demon-strated. O’Brien-Pallas et al. (2001) examined trends inearned hours and complexity of inpatient hospital casesfor the years 1994/95 through 1998/99, finding increasedpatient acuity and demand accompanied by decreasednumbers of nurses working in hospitals. Findingssuggested that as hours of care per patient day increased,so did the overtime nurses were asked to work and theincidence of missed shifts due to illness. These findingsare consistent with Zboril-Benson’s (2000) study inwhich higher rates of absenteeism were found to beassociated with lower job satisfaction, longer shifts,working in acute care and working full-time. In anOntario study, more than one-third of nurses experi-enced high emotional exhaustion, had higher overall andmusculo-skeletal claim rates compared to non-nurses,and musculo-skeletal claims comprised the majority ofnursing claims (Shamian and O’Brien-Pallas, 2001).Clarke et al. (2002) found poor organizational climateand high workloads to be associated with 50% to 2-foldincreases in the likelihood of needlestick injuries andnear-misses (with a needle or sharp) to nurses.

6.3. Patient outcomes

While studies are beginning to emerge that correlatenurse staffing with patient outcomes, more data areneeded to support the claim that patient safety dependson the ability to recruit and retain sufficient numbers ofqualified nurses and shape a supportive practiceenvironment (Clarke and Aiken, 2003). A few studieswere found that linked nurse turnover with quality ofpatient care (Shortell et al., 1994; Leiter et al., 1998) butno studies were found that substantiated the effect ofnurse turnover on quality of care delivery. Leiter et al.(1998) examined data from 605 patients and 711 nursesto find that patients on units where nurses found theirwork meaningful were more satisfied with all aspects oftheir hospital stay, and patients who stayed on unitswhere nursing staff felt more exhausted or morefrequently expressed the intention to quit were lesssatisfied with the various components of their care. In astudy looking at how nursing home care affects residentinfection and hospitalization for infection, Zimmerman

et al. (2002) found that RN turnover was significantlyrelated to both outcomes; with each proportionate lossof an RN the risk of infection increases almost 30% andthe risk of hospitalization increases more than 80%.Recent studies have linked nurse staffing levels with

the length of stay (Lichtig et al., 1999; Needleman et al.,2002); complication rates (Blegen et al., 1998; Robertsonand Hassan, 1999; Dimick et al., 2001; Whitman et al.,2002; Unruh, 2003); patient mortality and ‘failure torescue’ (Blegen et al., 1998; Tourangeau et al., 2002) andpatient incidents that comprise a threat to patient safetyand well- being (Blegen et al., 1998; Whitman et al.,2002). Large sample sizes and sophisticated statisticalanalysis substantiate the claim that better nurse staffinglevels in terms of skill mix (Blegen et al., 1998; Lichig etal., 1999; Needleman et al., 2002; Tourangeau et al.,2002; Unruh, 2003); patient-to-nurse ratio (Dimick etal., 2001; Aiken et al., 2002b); years of related experience(Tourangeau et al., 2002) and hours of nursing care(Blegen et al., 1998) result in more desirable outcomesfor patients. Aiken et al. (2001) claim that increasesin patient acuity and nurses’ responsibilities haveincreased workload in ways that, when coupled with adeteriorating practice environment, adversely affectpatient outcomes.Whitman et al. (2002) examined relationships between

nursing staffing and specific nurse-sensitive outcomesacross cardiac and non-cardiac intensive care andintermediate care units, and medical-surgical units.Significant inverse relationships were present betweenstaffing and falls in cardiac intensive care, medicationerrors in both cardiac and non-cardiac intensive careunits, and restraint rates in the medical-surgical units.Blegen et al. (1998) found that as the RN proportionincreased, rates of adverse outcomes decreased up to87.5%. The higher the RN skill mix, the lower theincidence of adverse occurrences on inpatient care units.Lichig et al. (1999) used nursing intensity weights toacuity-adjust the patient data and found that bothhigher nurse staffing and higher proportion of RNs weresignificantly related to shorter lengths of stay.The importance of good management in promoting

quality of patient care has been supported. Shortell et al.(1994) determined that management ability of intensivecare units was significantly associated with lower risk-adjusted length of stay, lower nurse turnover, higherevaluated technical quality of care and greater evaluatedability to meet family member needs. Similarly, accord-ing to multivariate results in an international studymanagerial support for nursing had a pronounced effecton nurse dissatisfaction and burnout (Aiken et al.,2002a). Also, organizational support for nursing andnurse staffing were related to nurse-assessed quality ofcare. Qualitative research supports these findings:themes relating to workload, human resources, nursingshortage, restructuring and physical environment were


seen as an increasing risk in patient care (Nicklin andMcVeety, 2002).

6.4. Summary of turnover consequences

There is overall consensus that undesirable nurseturnover is costly as well as detrimental to nurse andpatient outcomes. Researchers agree that adverseimpacts relate to decreased initial productivity of newemployees and decreased staff morale and productivity.Financial costs of replacing individual nurses are beingestimated, however, studies have yet to produce evidencein terms of incremental staffing costs and health caresystem costs. Studies that examine the effect ofinadequate nurse staffing and poor work environmentson nurse health and safety outcomes often do notinclude variables explicit to nurse turnover. Similarly,few studies have produced findings that link nurseturnover to measurable patient outcomes. The scantevidence that does exist suggests that patients do suffermore physically and emotionally in health care environ-ments experiencing high nurse turnover. As suggested inthe study findings in Appendix B, there is clearly a needfor research committed to better understanding nurseturnover and its consequences.

7. Methodological challenges

Methodological challenges have been identified byresearchers attempting to study turnover. Cavanagh(1989) concluded that simple bivariate investigations toexamine why nursing staff leave their jobs producedconflicting results and underscored the complexity of theturnover problem. A variety of study populations,methodologies and inconsistent definitions and measure-ments made comparison of turnover studies difficult,due to confusion over the precise meaning of turnoverand who is included in the analysis (Cavanagh, 1989; Taiet al., 1998). Irvine and Evans (1992) pointed out thatdifferent operational definitions of concepts couldaccount for inconsistent findings and that conceptclarification is needed to define the relationships amongthe various concepts. Due to the complexity of definingand measuring multifaceted predictors and outcomeconstructs as well as differences among work contexts,study results are often inconsistent with each other (MorBarak et al., 2001).Irvine and Evans (1992) indicate that studies have not

always specified whether the focus is voluntary turnoveror involuntary turnover. Where studies distinguishbetween voluntary and involuntary turnover, there isstill a lack of clarity due to inaccuracy of organizationaldata. Institutional records cannot be trusted blindly,since ‘‘face-saving’’ reasons may have been used todocument employee withdrawal, altering the distinction

between voluntary and involuntary termination (Cam-pion, 1991). Also, as identified by Waldman et al. (2004),a practical reason for refining turnover cost measuresrelates to the required adaptation of accountingconcepts in health care as corporate practices are appliedto clinical settings.Much of the turnover research is characterized by

small sample sizes at the unit and organization levels.Mor Barak et al. (2001) state that gaps in existingknowledge include the examination of macrolevelvariables such as organization size, setting, structure,funding status and other economic factors, as well asspecific job conditions and employee characteristics,thereby limiting usefulness of findings for policyformation. As samples often do not include differentsubgroups within an organization’s workforce, such asthe study of differential turnover rates between man-agers and subordinates within the same unit (Cavanagh,1989), the evidence is less beneficial than it could be inaddressing high nurse turnover.In terms of data collection, many studies have used

questionnaires as the sole means (Cavanagh, 1989) aswas the case in the most of the studies used in thisreview. While this approach is most advantageous inlarge-scale investigation, more opportunity to commenton relevant aspects of turnover behavior could enrichthe overall findings. Fottler et al. (1995), however,cautions against the use of exit interviews of terminatingnurses, since they are likely to identify non-job-relatedfactors as reasons for leaving, due to reluctance tocriticize the organization for fear of alienating theinterviewer and jeopardizing future job references.Until very recently, many studies utilized cross-

sectional designs, almost exclusively examining turnoverfrom a fixed point in time and using a dichotomous(turnover or no turnover) dependent variable (MorBarak et al., 2001). Authors from the past few decadesargued that more longitudinal research designs wouldincrease ability to predict who will leave and determinefactors that moderate turnover behavior (Hinshaw andAtwood, 1983; Cavanagh, 1989). Earlier researchers hadsuggested that the study of nursing turnover is toocomplicated for simple correlation; that the lack ofmultivariate analysis resulted in a lack of clarity as towhat extent the host of variables hypothesized toimpact turnover were simple correlates or indepen-dent predictors. (Cavanagh, 1989; Irvine and Evans,1992). Also, the relationship between intention toleave and actual turnover merits further examina-tion, since intention to leave alone accounts for only aportion of actual turnover (Mor Barak et al., 2001).Current analysis procedures are including moreadvanced statistical methods such as multivariate ana-lysis and structural equation modeling, suggesting thatresearchers are responsive to previous critiques. Futureresearch to examine the strongest turnover predictors


simultaneously would help determine their relationshipsto one another and discover their mediating andmoderating influences (Mor Barak et al., 2001).

8. Concluding comments

Concerns have been expressed nationwide and glob-ally about the shortage of nurses and the issue ofturnover, yet the data to support an evidence base todevelop effective policies to deal with these problems arelacking. A comprehensive review of recent literature wasconducted to examine the current state of knowledgerelating to causes and consequences of nurse turnover. Itwas found that research has been hampered bymethodological challenges, even in the inconsistentdefinition of turnover itself. Most of the investigationto date has concentrated on determinants of nurseturnover, highlighting factors including organizational,work, management, job, individual, and economic, aswell as moderating factors. Theoretical modeling ofturnover that seeks to demonstrate relationships amongdeterminants has consistently shown a positive relation-ship between intent to leave and turnover. Very littleresearch was found that focused on the impact of nurse

turnover, with economic impacts, nurse and patientoutcomes attracting interest.Decision-makers require more evidence relating to

costs associated with turnover, how staffing decisionsinfluence turnover, how nurse turnover affects the healthteam functioning, and how nurse turnover affects healthsystem outcomes including nurse and patient outcomes.For such evidence to be useful and generalizable,however, the methodological limitations of researchneed to be reduced. Studies that reflect standardizeddefinitions of turnover, agreed formulae for identifica-tion of costs, distinction between voluntary andinvoluntary turnover, involve samples that includemultiple units and organizations, even countries, andthe use of longitudinal research designs and multivariateanalysis promise to produce more robust understand-ing of the complex phenomenon of nursing turnover.A skilled nursing workforce is required to strengthenthe health systems of all nations. Policy solutions tonursing turnover and shortages risk being around‘‘quick fixes’’: the issue is complex and ultimately it isonly through more research and detailed analysis ofimpact of turnover within health systems and at countrylevel that viable and sustainable solutions becomeapparent.

ARTICLE IN PRESS

Appendix A. Studies of nurse turnover determinants

Author(s) Purpose Sample/Setting Method/Analysis Key Findings

Szigeti et al.(1991)

Investigate factorsrelating to RNs’and LPNs’ desireto keep practicingin rural hospitals

291 RNs andLPNs fromhospitals with lessthan 100 beds

Questionnaire Overall job satisfaction andperformance constraintsare a predictor of turnoverintent for RNs and LPNs.Satisfaction withpromotion the only work-related predictor ofturnover intent for RNs.Performance constraints,role ambiguity, and shiftwork were the only work-related turnover predictorsfor LPNs

59% responserate

Correlationalanalysis

Bloom et al.(1992)

Examine thesisthat socialorganization ofwork in hospitalsis determinant ofvoluntaryturnover rate

Administrativedata from 435hospitals from theNursingPersonnel Survey

Measures:voluntaryturnover, skill andeducational ofnurses, workingconditions, workpatterns, careerladdering, work

Ratio of RNs to total staffis related to turnover.Total, primary, andmodular nursingorganization hadsignificant but marginaleffect on turnover. Careerladder unrelated to

L.J. Hayes et al. / International Journal of Nursing Studies 43 (2006) 237–263 247

among registerednurses

incentives,economicopportunity

turnover. Professionalbenefits related to higherrather than lower turnover.Hospitals with a largerproportion of longertenured nurses had lowerturnover. Economicopportunity hadinconsistent effects onturnover

Secondary dataanalysis Multipleregression

Gardner (1992) Measure careercommitment andexplore itsrelationship toturnover andworkperformance

320 newlyemployedregistered nursesat one hospital

A longitudinal,repeated-measuresdescriptive surveyto measure careercommitment andexplorerelationship toturnover andworkperformance

Career commitment scoresdropped over the first year.Although careercommitment correlateswith turnover, and there isa relationship with jobperformance, the directassociation is weak. Careercommitment in the firstyear in a new job issusceptible toorganizational factors

Judge (1993) Test hypothesisthat there isinteractionbetween affectivedisposition andjob satisfaction

Medical clinic.Subjects!N " 234#: RNs(56%), LPNs(15%), officeassistants (15%),lab techs (14%)

QuestionnaireMeasures:affectivedisposition, jobsatisfaction,voluntaryturnover.Regressionanalysis

Nurse data supportedhypothesis. The morepositive the disposition ofthe individual, the strongerthe relationship betweenjob dissatisfaction andturnover. Individualsdissatisfied with their jobsbut positively disposed tolife were the most likely toquit

Cameron et al.(1994)

Examinesimilarities anddifferencesbetween jobsatisfaction,propensity toleave, andburnout of RNsand RNAs

623 RNs and 231RNAs from threecommunityhospitals

Questionnaire-Quality ofworklifeconditions, indexof organizationalreactions,burnout scale,propensity toleave

Both groups moderatelysatisfied with jobs, burnoutor propensity to leave. RNswith more years ofexperience had highest jobsatisfaction, lowestburnout, and were lesslikely to leave. None ofthese variables were relatedto the length of time RNAswere employed. RNs inpsychiatric settings leastsatisfied with jobs, morelikely to leave, andreported more burnoutthan RNs in other settings

Multivariateanalysis


Gray and Phillips(1994)

Examine turnoverrates andrelationship toage and length ofservice

342,000employees fromnine staff groupsin NHS

Secondary dataanalysis

Annual turnover rateamong all NHS staff13.6%; higher turnoveramong full-time than part-time; turnover declineswith age, are high in firstyear of service, remainshigh in second year beforedeclining

Krausz et al.(1995)

Test withdrawalprogression-relationshipbetween predictorvariables andwithdrawal levelsof ward, hospitaland profession

Wave 1–390nurses in largehospital. One yearlater same nursescompleted similarquestionnaireresulting in 146matched pairs(37.4% of 390)

Questionnaire-Withdrawalintention, JobScope (Hackmanand Oldman),Burnout (Pines etal.)

Milder forms ofwithdrawal preceded moresevere forms, culminatingin intent to leave theprofession. Job scopeeffected intent to leave theward; burnout effectedintent to leave the hospital.Although intention to leavethe profession was betterpredicted than intention toleave the ward or thehospital, the cumulativeeffects of the two lowerlevels of withdrawalintention were the majordeterminants

LISRELGoodness-of-fitmeasures

Leveck and Jones(1996)

Examine effectsof management,group cohesion,job stress, jobsatisfaction(organizationaland professional)on retention andquality of care

Theoretical modelwas tested usingdata from 50nursing units atfour acute carehospitals in thesoutheast

Hinshaw andAtwood’S (1983)anticipatedturnover modelwas modified andexpanded toinclude relevantantecedent andoutcome variables

The model explained 49%of the variance in staffnurse retention and 39% ofthe variance in processaspects of quality ofnursing care. Experienceon unit and professionaljob satisfaction werepredictors of staff nurseretention; job stress andclinical service werepredictors of quality ofcare. The variablecontributing the most wasmanagement style

Davidson et al.(1997)

Examine effectsof hospital changeon jobsatisfaction andturnover

736 nurses from ahospital

Questionnaire-Satisfaction scalesby Hinshaw andAtwood and Priceand Mueller

Intent to leave predicted byperception of littlepromotion, highroutinization, low decisionlatitude and poorcommunication. Predictorsof turnover were feweryears on the job, expressedintent to leave, and lack oftime to do job well


73.5% responserate

Multivariateanalysis

Kiyak et al.(1997)

Examineemployees’ levelof commitment,operationalized asintent to leaveand subsequenttermination ofemployment

Staff of sixnursing homesand 12communityfacilities servingolder adults!N " 308#

Questionnaire-Job DescriptionIndex (Smith etal., 1974), jobsatisfaction,intention to leave

Turnover predicted best byintent to leave, followed bylength of employment(shorter), and age(younger). Intent to leavepredicted by age (younger),length of employment(shorter), jobdissatisfaction, and type ofagency worked(community).Dissatisfaction is a factorin desire to leave and maylead to turnover orcontinued dissatisfaction

MultipleregressionDiscriminantfunction analysis

Song et al. (1997) Compare jobsatisfaction,absenteeism, andturnover betweennurse-managedspecial care unit(SCU) and intraditional ICU

143 nurses- 34from SCU (90%response rate) and109 from ICUs(69% responserate)

Annualquestionnaires1991–1994- JobDiagnostic Survey(Hackman andOldman), Staffperception andpreference, nurseabsenteeism andturnoverANCOVA

Was greater supervisoryand payment satisfaction inSCU (shared governanceand minimal technology)than ICU (bureaucraticand high tech). Percentabsent hours for SCUnurses was lower thanICU. Turnover rates didnot differ between the twounits Nurses working inboth the SCU and theICUs preferred the SCUpractice model over theICU model

Lum et al. (1998) Assess direct andindirect impact ofpay policies onturnoverintentions ofpediatric nurses

361 RNs inneonatal andpediatric ICU

Questionnaire-Job SatisfactionScale, PaySatisfaction,Organizationalcommitmentscale, TurnoverIntention Chi-square Goodness-of-fit

Job satisfaction hasindirect and organizationalcommitment has directeffect on intent to quit. Paysatisfaction had direct andindirect effect on turnoverintent. Having children, adegree, and working 12-hshifts have direct andindirect influence on paysatisfaction and turnoverintent

77.5% responserate

Boyle et al. (1999) Examine effectsof managers’power, influence,

255 staff nurses inintensive care

Questionnaire-Manager,organization and

Managers’ position powerand influence over workcoordination had direct


and leadershipstyle on nurses’intent to stay

units at foururban hospitals

workcharacteristics;nursecharacteristics ofopportunityelsewhere; jobstress,satisfaction, andcommitment; andintent to stay

link to intent to stay.Instrumentalcommunication,autonomy, and groupcohesion decreased jobstress and thus increasedjob satisfaction. Jobsatisfaction was directlylinked with intent to stay

Causal modelingand multipleregression pathanalysis

Hemingway andSmith (1999)

Examine model:organizationalclimatehypothesized toaffect withdrawaland injuriesthrough themediating effectsof occupationalstressors

252 RNs fromdifferent wardsand shifts withinfour hospitals

Questionnaire-WorkEnvironmentscale, NursingStress Scale,Injuries (reported,unreported, near),Injury composite,absenteeism,Turnoverintention

Turnover intentions andinjury measures related toorganizational climate andoccupational stressors.However, climate andstressors not significantlyrelated to frequency ofshort-term absences. Highwork pressure waspredictive of greater roleconflict and role ambiguity,and of more stress fromheavy workload Poorsupervisor support wassignificantly related togreater stress from heavyworkload, and lack ofautonomy was related tohigher role ambiguity

MultipleregressionanalysisHierarchicalmultipleregression

Janssen et al.(1999)

Gain insight intothe relationshipsbetween workcharacteristicsand specific stressreactions

156 nurses fromnine departmentsat a generalhospital

Questionnaire-Quality of job,mental workoverload, Socialsupport, unmetcareerexpectations,Intrinsic workmotivation,Maslach BurnoutInventory

Emotional exhaustion waspredicted by lack of socialsupport from colleaguesand work demands.Turnover intention wasdetermined by unmetcareer expectations, such asa higher salary, moreresponsibility and to alesser extent by quality ofjob content

89% responserate

PearsoncorrelationsLISREL analysis


Kane (1999) Determine if aredifferences in jobsatisfaction,burnout, anddesire to leaveposition, in nursesemployed in full-time, part-time, orjob sharing

Sample !N " 269#drawn from largeCanadianteaching hospital.Three samplegroups weredeveloped,consisting of jobsharing, full-time,and part-time

Quasi-experimental.Questionnaire tomeasure jobsatisfaction,propensity toleave, burnoutand familystressors

Job-sharing nursesreported greatersatisfaction than full-timenurses. No statisticallyrelevant findings related toemployment status andpropensity to leave. Totalburnout scale notsignificantly different byemployment status butDepersonalizationSubscale showedsignificant difference withjob-sharing nurses faringbetter than the full-time orpart-time groups. Nosignificant relationshipnoted between familystressor and jobsatisfaction, burnout andpropensity to leave

DescriptivestatisticsMultivariateanalysisMANCOVA

Mano-Negrin andKirschenbaum(1999)

To understandimpact of‘careerist’attitudes (pusheffects) andorganizationalbenefits (pull) onturnover

707 health careworkers of which46% were nurses

Questionnaire-Voluntaryturnover adichotomousvariable (stayedvs. left)

Nurses and physicianturnover behavior isnegatively affected bycareer aspiration-expectations foradvancement kept them inorganization, added to byfear of unemployment(nurses). Turnover reflectsthe effect of the balancebetween organizationalbenefits (pull factors) and acareerist attitude to work(push factors)

29% responserate

Cross-sectionalanalysisANOVALogisticsregression

Chan andMorrison (2000)

Explore factorsinfluencingturnover intentionof RNs

114 nurses fromvarious units inlarge hospital

Questionnaire-adapted fromBattersby et al.(1990)

Leavers: Cert or diplomagrads, work non-ICU, RNs2–4 years. Stayers:Specialized qualification,work in ICU, had eithero2 years or between 6-10years experience. Workfactors of job satisfaction,collegiality, use of skills,


recognition, staffing andsalary influenced if stay orleave

95% responserate

Descriptivestatistics

Collins et al.(2000)

Examine views ininnovative roles,on jobsatisfaction,careerdevelopment,intention to leaveand factors thathinder andenhance working

452 nurses and162 professionalsallied to medicine(PAMs)

Questionnaire High level of jobsatisfaction in both groups(nurses and PAMs). Jobsatisfaction significantlyrelated to feeling integratedwithin professional groupand colleagues, feeling thatnew roles improved careerprospects, adequatelyprepared and trained forthe role. 68% !n " 415# feltthe role enhanced careerprospects but 27% !n "163# said they would leavetheir profession if theycould. Low job satisfactionwas significantly related tointention to leave theprofession

78.5% responserate

Descriptivestatistics Contentanalysis

Cartledge (2001) Explore factorsinfluencingturnover

Intensive carenurses who leftposition

Content analysisof eleveninterviews

Four themes emerged fromcontent analysis: stress,professional development,recognition and respect ofothers, and implications ofshift work

Ito et al. (2001) Relate intent toleave to jobsatisfaction,perceived risk ofassault, andsupervisorysupport

1,494 nurses from27 psychiatrichospitals

Questionnaire-National Institutefor OccupationalSafety and Healthjob stress

44% reported intention toleave their job, 89% ofthose perceived a risk ofassault. Younger age, fewerprevious job changes, lesssupervisory support, lowerjob satisfaction and moreperceived risk of assaultsignificant predictors ofintention to leave

76.5% responserate

Chi square, t testKendall’s tau bLogisticregression

Moore (2001) Examine impactof restructuringon conditions andnurses’ well-being, nursesreaction to

201 nursesworking in threehospitals

Internetquestionnaire-RestructuringInitiative Index,Impact ofRestructuring

Despite high conditions,nurses kept strong sense ofprofessional efficacy.However, impact predictednurse burnout.Management


changes; anddegree to whichnurses felt theywere consultedand supported

Index andHospitalConsultationScale (Burke andGreenglass);Social Support(Tate, 1996);Deakin CopingScale (Moore andGreenglass, 1999),burnout and self-efficacy items,Intention to Quit

communications proposedas a mediator againstburnout but was notsupported. Nursesconsidered management atop-down approach andthis non-consultativecommunication alsopredicted nurse burnout.The restructuring changes,their impact, and poorcommunication style, andburnout also predictednurses’ intentions to quit;however, a sense ofprofessionalism mediatedthis intent

Asymptoticvariance-covariance matrixstructuralequationmodeling

Shader et al.(2001)

Examinerelationshipbetween jobsatisfaction, agestress, cohesion,work scheduleand turnover

241 nurses andnurse managers in12 units in a 908-bed hospital

Questionnaire-Index of WorkSatisfaction, JobStress Scale,Byrne GroupCohesion Scale,AnticipatedTurnover Scale

Job stress, worksatisfaction, groupcohesion and weekendovertime were predictors ofanticipated turnover. Therewere differences in thefactors predictinganticipated turnover fordifferent age groups

Unit demographicdata fromhospital databasesPearson’scorrelations

Shields and Ward(2001)

Investigatedeterminants ofjob satisfaction,establishimportance of jobsatisfaction inintent to quit

Data of 9625nurses from 1994national survey ofnursing staff

Secondary dataanalysis jobsatisfaction scalesfor work aspectsTurnover—Stayervs. Quitter

Job satisfaction is mostimportant determinant inintent to quit, more so thanoutside opportunities. Poorcareer advancementopportunities (largestquantitative affect),increased workload, payand workplace relationsimportant in quittingoutcome. Workdissatisfaction is greater inyoung, male, ethnicminority and highlyeducated nurses

Cross-tabulationst statistics


Lu et al. (2002) Investigaterelationshipsamong turnoverintent,professionalcommitment, andjob satisfaction ofRNs

Random sampleof 2197 hospitalnurses

Questionnaire—ProfessionalCommitmentscale (Lu, Chiouand Chang, 2000)TurnoverIntention (Blauand Lunz, 1998)Job Satisfaction

Professional commitmentwas more effective inpredicting intent to leavenursing profession thanintent to leaveorganization. Professionalcommitment related tointent to leave professionwas more stronglycorrelated than intent toleave the organization. Jobsatisfaction related toorganizational leave intentwas more stronglycorrelated thanprofessional leave intention

86.2% responserate

DescriptivestatisticsCorrelationmatricesDiscriminantanalysis

Tzeng (2002) Hypothesis: thehigher the generaljob satisfactionand happiness,and satisfactionwith role, thelower the intent toquit

648 hospitalnurses 82%response rate

Questionnaire-Job satisfactionsubscales, generalperceptual factorsordinal logisticsregressionanalyses

General job satisfaction,general job happiness,satisfaction with salary andpromotion, institution,educational background,and age of nurses’ youngestchild were significantpredictors of nurses’intention to quit

Larrabee et al.(2003)

Investigate therelative influenceof attitudes andcontext of andstructure of careon nurses’jobsatisfaction andintent to leave

90 RNs employedon two medical,two surgical, andthree intensivecare step-downunits at a medicalcenter

Questionnaire-Work QualityIndex,MultifactorLeadershipQuesionnaire,Profession workenvironmentsubscale, NurseCollaborativePractice Subscale,Group CohesionScale, Spreitzer’s12-itemquestionnaire,Personal ViewsSurvey

Major predictor of intentto leave was jobdissatisfaction, and themajor predictor of jobsatisfaction waspsychologicalempowerment. Predictorsof psychologicalempowerment werehardiness,transformationalleadership style, nurse/physician collaboration,and group cohesion

Strachota et al.(2003)

Investigatereasons nurseschangedemploymentstatus

84 hospital nurseswho changedemploymentstatus

Telephoneinterviews over 4months

Reasons for change were:hours worked; better jobopportunity; family; poorpay and benefits; poorstaffing; unsupportivemanagement; unacceptable


work environment; workstress; no opportunity foradvancement

46% responserate

Data analyzed forthemes andcategories

Anderson et al.(2004)

Tested effect ofadministrativeclimate,communicationpatterns, andinteractionbetween the twoon turnover

3449 employees in164 randomlysampled nursinghomes

Perceptions ofadministrativeclimate andcommunicationlinked tosecondary data

Climate andcommunication affectedturnover, but lowerturnover dependent oninteraction between thetwo. With reward-basedclimates, higher levels ofcommunication opennessand accuracy explainedlower turnover, relative tonursing homes with anambiguous climate.Adequate staffing andlonger tenure of nursingdirector also importantpredictors of turnover

Hierarchicalregression

Duffield et al.(2004)

Identify positionsof RNs moved towhen left nursingand exploreperceptions aboutthe skills gainedfrom nursing

154 nurses nolonger employedin nursing

Questionnaire-Current job, skillsgained fromnursing, reasonsfor becoming anurse, reasons forleaving nursing(Duffield andFrank)

Many moved tomanagement outsidehealthcare, most tookadded study. Few haddifficulty in non-nursingemployment, most agreednursing skills assisted inattaining positions.Reasons for leaving relatedto work aspects, structuralaspects, professional issues,team support, salary andprestige, employer care,and legal concerns

Kruskal-Wallistest

HendersonBetkus andMacLeod (2004)

Examine publichealth nurses’(PHNs) job andcommunitysatisfaction, andrelate it todecision to stay

124 PHNs from48 rural and smallurbancommunities inBritish Columbia

Self-administeredquestionnaireStramps andPiedmontes WorkSatisfaction Index

PHNs most satisfied withprofessional status,interaction and autonomy,and friendliness of thecommunity and theirfriends, and least satisfiedwith salary. Jobsatisfaction or communitysatisfaction did notinfluence retention. Age,retirement, family needsand the economy affectedintent to stay or leave


76% responserate

Independent t-testContent coding

Holtom andO’Neill (2004)

Examine jobembeddedness inpredictingretention andassess whetherfactors thatinfluence nurseretention aredifferent fromother workers

232 employees,including nurses,from communityhospital

Questionnaire 1year later,hospital dataregarding leavers

Job embeddednesspredicted turnover beyonda combination of perceiveddesirability of movementmeasures (job satisfaction,organizationalcommitment) andperceived ease ofmovement measures (jobalternatives, job search).Job embeddedness assessednew variance in turnover inexcess of that predicted bymajor variables in majorturnover models. Leversfor influencing retentionare substantially similar fornurses and other healthcareworkers

46.4% responserate

Alpha coefficient

ARTICLE IN PRESS

Appendix B. Studies of nurse turnover consequences

Author(s) Purpose Sample/Setting Method/Analysis Findings

Jones (1990b) Develop amethodology tomeasure nurseturnover costs

Turnover andturnover cost datafrom four acutecare hospitalsgrouped intodirect and indirectcosts

Total nursingturnover costdivided by thenumber of nursesreported asturning over ateach of thehospitals

Mean turnover rate 26.8%

Mean cost per RNturnover was $10,198,ranging from $6,886 to$15,152

Shortell et al.(1994)

Examine factorsassociated withrisk-adjustedmortality, risk-adjusted averagelength of stay,nurse turnover,technical qualityof care, andability to meetfamily memberneeds

17,440 patientsacross 42 ICUs

Apache IIImethodology forrisk-adjustment

Technological availabilitysignificantly associatedwith lower risk-adjustedmortality (beta " $0.42);diagnostic diversitysignificantly associatedwith greater risk-adjustedmortality (beta " 0.46);and caregiver interactioncomprising the culture,leadership, coordination,communication, andconflict management


ARTICLE IN PRESS

abilities of the unit issignificantly associatedwith lower risk-adjustedlength of stay(beta " 0.34), lower nurseturnover (beta " $0.36),higher evaluated technicalquality of care(beta " 0.81), and greaterevaluated ability to meetfamily member needs(beta " 0.74)

Leiter et al. (1998) Study relationshipof nurse burnout,intent to quit, andmeaningfulness ofwork with patientsatisfaction withcare, informationprovided and carecoordination, andoutcomes of thehospital stay

605 patients and711 nurses from16 inpatient unitsfrom two hospitalsites

Staff survey:Maslach BurnoutInventory,meaningfulness ofwork, intention toquit,

Patients where nursesfound work meaningfulwere more satisfied with allaspects of hospital stay.Patients who stayed onunits where nursing stafffelt exhausted or expressedthe intention to quit wereless satisfied with thevarious components oftheir care. Although nursecynicism was reflected inlower patient satisfaction,the correlations betweencynicism and other aspectsof care fell below statisticalsignificance. No significantcorrelations were foundbetween nurse professionalefficacy and patientsatisfaction

Patient survey:Satisfactionmeasures andoutcomes

Zimmerman et al.(2002)

Determinerelationshipbetween structureand processelements ofnursing home careand residentoutcomes

Senioradministrators

Interviews forfacility data

RN turnover significantlyrelated to outcomes; witheach proportionate loss ofan RN the risk of infectionincreased almost 30% andthe risk of hospitalizationincreased more than 80%.When other predictiveprocess and structurevariables are included,associations persist,implying that therelationships betweenturnover and infection &hospitalization cannot beexplained by confoundingwith other processes andstructure variables

L.J. Hayes et al. / International Journal of Nursing Studies 43 (2006) 237–263258

ARTICLE IN PRESS

2015 newadmissions aged65 and older froma stratifiedrandom sample of59 nursing homes

Baseline data ofresidents followedfor 2 years.Outcomemeasures:infection andhospitalizationfor infection.Regressionanalysis

Waldman et al.(2004)

Examine turnoverand its costs in thehealth careenvironment

Multipledatabases tomeasure costs ofhiring, training,and termination,and to measureemployeeproductivity

Prior work oncosting turnoverused for newaccountingmethodology.This model wasapplied to a largeacademic medicalcenter

The total costs for a newlyhired nurseaveraged$15,825. CoRPsfor nurses ranged from$5,245 to$16,102

The training of nursesgenerated 59% of totaltraining costsTurnover costs representan expenditure of about5% of the annual operatingbudget

References

Abelson, M.A., 1986. Strategic management of turnover: amodel for the health service administrator. Health CareManagement Review 11 (2), 61–71.

Aiken, L.H., Patrician, P.A., 2000. Measuring organizationaltraits of hospitals: the revised nursing work index. NursingResearch 49 (3), 146–153.

Aiken, L.H., Smith, H.L., Lake, E.T., 1994. Lower Medicaremortality among a set of hospitals known for good nursingcare. Medical Care 32 (8), 771–787.

Aiken, L.H., Clarke, S.P., Sloane, D.M., Sochalski, J.A., Busse,R., Clarke, H., Giovannetti, P., Hunt, J., Rafferty, A.M.,Shamian, J., 2001. Nurses’ reports on hospital care in fivecountries. Health Affairs 20 (3), 43–53.

Aiken, L.H., Clarke, S.P., Sloane, D.M., 2002a. Hospitalstaffing, organization, and quality of care: cross-nationalfindings. Nursing Outlook 50 (5), 187–194.

Aiken, L.H., Clarke, S.P., Sloane, D.M., Sochalski, J., Silber,J.H., 2002b. Hospital nurse staffing and patient mortality,nurse burnout, and job dissatisfaction. Journal of theAmerican Medical Association 288 (16), 1987–1993.

Aiken, L.H., Buchan, J., Sochalski, J., Nichols, B., Powell, M.,2004. Trends in international nurse migration. HealthAffairs 23 (3), 69–77.

Alexander, J.A., 1988. The effects of patient care unitorganization on nursing turnover. Health Care Manage-ment Review 13 (2), 61–70.

Alexander, J.A., Lichtenstein, R., Oh, H.J., Ullman, E., 1998. Acausal model of voluntary turnover among nursing person-nel in long-term psychiatric settings. Research in Nursingand Health 21 (5), 415–427.

Anderson, R.A., Corazzini, K.N., McDaniel Jr., R.R., 2004.Complexity science and the dynamics of climate andcommunication: reducing nursing home turnover. Geron-tologist 44 (3), 378–388.

Angerami, E.L., Gomes, D.L., Mendes, I.J., 2000. Permanenceof nurses in their profession. Revista Latino-Americana deEnfermagem 8 (5), 52–57 [Article in Portuguese].

Barrett, L., Yates, P., 2002. Oncology/haematology nurses: astudy of job satisfaction, burnout, and intention to leave thespecialty. Australian Health Review 25 (3), 109–121.

Baumann, A., O’Brien-Pallas, L., Armstrong-Stassen, M.,Blythe, J., Bourbonnais, R., Cameron, S., Irvine Doran,D., Kerr, M., McGillis Hall, L., Vezina, M., Butt, M.,Ryan, L., 2001. Commitment and care: the benefits of ahealthy workplace for nurses, their patients and the system.Canadian Health Services Research Foundation and TheChange Foundation, Ottawa.

Blau, G., Lunz, M., 1998. Testing the incremental effect ofprofessional commitment on intent to leave one’s professionbeyond the effects of external, personal, and work-relatedvariables. Journal of Vocational Behavior 52 (3), 260–269.

Blegen, M.A., Goode, C.J., Reed, L., 1998. Nurse staffing andpatient outcomes. Nursing Research 47 (1), 43–50.

Bloom, J.R., Alexander, J.A., Nuchols, B.A., 1992. The effectof the social organization of work on the voluntary turnover


rate of hospital nurses in the United States. Social Scienceand Medicine 34 (12), 1413–1424.

Borda, R.G., Norman, I.J., 1997. Testing a model of absenceand intent to stay in employment: a study of registerednurses in Malta. International Journal of Nursing Studies 34(5), 375–384.

Boyle, D.K., Bott, M.J., Hansen, H.E., Woods, C.Q., Taunton,R.L., 1999. Managers’ leadership and critical care nurses’intent to stay. American Journal of Critical Care 8 (6),361–371.

Bratt, M.M., Broome, M., Kelber, S., Lostocco, L., 2000.Influence of stress and nursing leadership on job satisfactionof pediatric intensive care unit nurses. American Journal ofCritical Care 9 (5), 307–317.

Buchan, J., 1994. Lessons from America? US magnet hospitalsand their implications for UK nursing. Journal of AdvancedNursing 19 (2), 373–384.

Cameron, S.J., Horsburgh, M.E., Armstrong-Stassen, M.,1994. Job satisfaction, propensity to leave and burnout inRNs and RNAs: a multivariate perspective. CanadianJournal of Nursing Administration 7 (3), 43–64.

Campion, M.A., 1991. Meaning and measurement of turnover:comparison of alternative measures and recommendationsfor research. Journal of Applied Psychology 76 (2), 199–212.

Cartledge, S., 2001. Factors influencing the turnover ofintensive care nurses. Intensive and Critical Care Nursing17 (6), 348–355.

Cavanagh, S.J., 1989. Nursing turnover: literature review andmethodological critique. Journal of Advanced Nursing 14(7), 587–596.

Cavanagh, S.J., Coffin, D.A., 1992. Staff turnover amonghospital nurses. Journal of Advanced Nursing 17 (11),1369–1376.

Chan, E.-Y., Morrison, P., 2000. Factors influencing theretention and turnover intentions of registered nurses in aSingapore hospital. Nursing and Health Sciences 2 (2),113–121.

Chang, E., 1999. Career commitment as a complex moderatorof organizational commitment and turnover intention.Human Relations 52 (10), 1257–1278.

Clarke, S.P., Aiken, L.H., 2003. Failure to rescue. AmericanJournal of Nursing 103 (1), 42–47.

Clarke, S.P., Rockett, J.L., Sloane, D.M., Aiken, L.H., 2002.Organizational climate, staffing, and safety equipment aspredictors of needlestick injuries and near-misses in hospitalnurses. American Journal of Infection Control 30 (4),207–216.

Collins, K., Jones, M.L., McDonnell, A., Read, S., Jones, R.,Cameron, A., 2000. Do new roles contribute to jobsatisfaction and retention of staff in nursing and professionsallied to medicine? Journal of Nursing Management 8 (1),3–12.

Cooper, H., 1998. Synthesizing Research: A Guide forLiterature Reviews, third ed. Sage Publications, ThousandOaks, CA, USA.

Curran, C.R., 1991. Calculating the costs of nurse turnover.Recruitment and Retention Report 4 (10), 1–2.

Davidson, H., Folcarelli, P.H., Crawford, S., Duprat, L.J.,Clifford, J.C., 1997. The effects of health care reforms onjob satisfaction and voluntary turnover among hospital-based nurses. Medical Care 35 (6), 634–645.

Dimick, J.B., Swoboda, S.M., Pronovost, P.J., Lipsett, P.A.,2001. Effect of nurse-to-patient ratio in the intensive careunit on pulmonary complications and resource use afterhepatectomy. American Journal of Critical Care 10 (6),376–382.

Donner, G.J., Wheeler, M.M., 2001. Career planning anddevelopment for nurses: the time has come. InternationalNursing Review 48 (2), 79–85.

Dougherty, L.M., Bolger, J.P., Preston, D.G., Jones, S.S.,Payne, H.C., 1992. Effects of exposure to aggressivebehavior on job satisfaction of health care staff. Journalof Applied Gerontology 11 (2), 160–172.

Droogan, J., Song, F., 1996. The process and importance ofsystematic reviews. Nurse Researcher 4 (1), 15–26.

Duffield, C., Aitken, L., O’Brien-Pallas, L., Wise, W.J., 2004.Nursing: a stepping stone to future careers. Journal ofNursing Administration 34 (5), 238–245.

Durand, M.A., Randhawa, G., 2002. Nurses’ views aboutreturning to practice after a career break. British Journal ofNursing 11 (7), 477–485.

Estabrooks, C.A., Tourangeau, A.E., Humphrey, C.K., Hes-keth, K.L., Giovannetti, P., Thomson, D., Wong, J., Acorn,S., Clarke, H., Shamian, J., 2002. Measuring the hospitalpractice environment: a Canadian context. Research inNursing and Health 25 (4), 256–268.

Evers, W., Tomic, W., Brouwers, A., 2002. Aggressivebehaviour and burnout among staff of homes for theelderly. International Journal of Mental Health Nursing 11(1), 2–9.

Fottler, M.D., Crawford, M.A., Quintana, J.B., White, J.B.,1995. Evaluating nurse turnover: comparing attitudesurveys and exit interviews. Hospital and Health ServicesAdministration 40 (2), 278–295.

Frisina, A., Murray, M., Aird, C., 1988. What Do NursesWant? A Review of Job Satisfaction and Job TurnoverLiterature. Health Care Research Unit, University ofToronto, Toronto.

Gardner, D.L., 1992. Career commitment in nursing. Journal ofProfessional Nursing 8 (3), 155–160.

George, J.M., Jones, G.R., 1996. The experience of work andturnover intentions: Interactive effects of value attainment,job satisfaction, and positive mood. Journal of AppliedPsychology 81 (3), 318–325.

Gifford, B.D., Zammuto, R.F., Goodman, E.A., 2002. Therelationship between hospital unit culture and nurses’quality of work life. Journal of Healthcare Management47 (1), 13–25 discussion 25-6.

Gray, A.M., Phillips, V.L., 1994. Turnover, age, and length ofservice: a comparison of nurse and other staff in theNational Health Service. Journal of Advanced Nursing 19(5), 819–827.

Gray, A.M., Phillips, V.L., 1996. Labour turnover in the BritishNational Health Service: a local labour market analysis.Health Policy 36 (3), 273–289.

Gray, A.M., Phillips, V.L., Normand, C., 1996. The costs ofnursing turnover: evidence from the British National HealthService. Health Policy 38 (2), 117–128.

Hemingway, M.A., Smith, C.S., 1999. Organizational climateand occupational stressors as predictors of withdrawalbehaviours and injuries in nurses. Journal of Occupationaland Organizational Psychology 72 (3), 285–299.


Henderson Betkus, M., MacLeod, M.L.P., 2004. Retainingpublic health nurses in rural British Columbia. CanadianJournal of Public Health 95 (1), 54–58.

Hinshaw, A.S., Atwood, J.R., 1983. Nursing staff turnover,stress, and satisfaction: Models, measures, and manage-ment. Annual Review of Nursing Research 1, 133–153.

Holtom, B.C., O’Neill, B.S., 2004. Job embeddedness: atheoretical foundation for developing a comprehensivenurse retention plan. Journal of Nursing Administration34 (5), 216–227.

Hung, R., 2002. A note on nurse self-scheduling. NursingEconomics 20 (1), 37–39.

International Organisation for Migration, 2002. Migration fordevelopment in Africa. Geneva

Irvine, D., Evans, M., 1992. Job satisfaction and turnoveramong nurses: a review and meta-analysis. Toronto,Faculty of Nursing, University of Toronto, Quality ofNursing Worklife Research Unit.

Irvine, D., Evans, M., 1995. Job satisfaction and turnoveramong nurses: integrating research findings across studies.Nursing Research 44 (4), 246–253.

Ito, H., Eisen, S.V., Sederer, L.I., Yamada, O., Tachimori, H.,2001. Factors affecting psychiatric nurses’ intention to leavetheir current job. Psychiatric Services 52 (2), 232–234.

Janssen, P.P.M., de Jonge, J., Bakker, A.B., 1999. Specificdeterminants of intrinsic work motivation, burnout andturnover intentions: a study among nurses. Journal ofAdvanced Nursing 29 (6), 1360–1369.

Johnson, J.E., Buelow, J.R., 2003. Providing staff feedback tonurse managers using internal resources. Journal of NursingAdministration 33 (7/8), 391–396.

Jones, C.B., 1990a. Staff nurse turnover costs: Part I, aconceptual model. Journal of Nursing Administration 20(4), 18–23.

Jones, C.B., 1990b. Staff nurse turnover costs: Part II,measurements and results. Journal of Nursing Administra-tion 20 (5), 27–32.

Jones, C.B., Stasiowski, S., Simons, B.J., Boyd, N.J., Lucas,M.D., 1993. Shared governance and the nursing practiceenvironment. Nursing Economics 11 (4), 208–214.

Judge, T.A., 1993. Does affective disposition moderate therelationship between job satisfaction and voluntary turn-over? Journal of Applied Psychology 78 (3), 395–401.

Kane, D., 1999. Job sharing: a retention strategy for nurses.Canadian Journal of Nursing Leadership 12 (4), 16–22.

Kane, D., Kartha, A., 1992. Job sharing: impact on the generalwell-being of female nurses. Canadian Journal of NursingAdministration 5 (1), 6–10.

Kennington, G., 1999. A case for a formalized careerdevelopment policy in orthopaedic nursing. Journal ofOrthopaedic Nursing 3 (1), 33–38.

Kiyak, H.A., Namazi, K.H., Kahana, E.F., 1997. Job commit-ment and turnover among women working in facilitiesserving older persons. Research on Aging 19 (2), 223–246.

Kramer, M., Schmalenberg, C.E., 2003. Magnet hospital staffnurses describe clinical autonomy. Nursing Outlook 51 (1),13–19.

Krausz, M., Koslowsky, M., Shalom, N., Elyakim, N., 1995.Predictors of intentions to leave the ward, the hospital, andthe nursing profession. Journal of Organizational Behavior16, 277–288.

Lane, I.M., Matthews, R.C., Presholdt, P.H., 1990. Educa-tional background, marital status, moral obligation andnurses’ decision to resign from her hospital. Journal ofApplied Social Psychology 20, 1432–1443.

Larrabee, J.H., Janney, M.A., Ostrow, C.L., Withrow, M.L.,Hobbs Jr., G.R., Burant, C., 2003. Predicting registerednurse job satisfaction and intent to leave. Journal ofNursing Administration 33 (5), 271–283.

Laschinger, H.K., Shamian, J., Thomson, D., 2001. Impact ofmagnet hospital characteristics on nurses’ perceptions oftrust, burnout, quality of care, and work satisfaction.Nursing Economics 19 (5), 202–219.

Laschinger, H.K., Finegan, J., Shamian, J., Wilk, P., 2004. Alongitudinal analysis of the impact of workplace empower-ment on work satisfaction. Journal of OrganizationalBehavior 25, 527–545.

Leiter, M., Harvie, P., Frizzell, C., 1998. The correspondence ofpatient satisfaction and nurse burnout. Social Science andMedicine 47 (10), 1611–1617.

Leveck, M.L., Jones, C.B., 1996. The nursing practice environ-ment, staff retention, and quality of care. Research inNursing and Health 19 (4), 331–343.

Lichtig, L., Knauf, R.A., Milholland, D.K., 1999. Someimpacts of nursing on acute care hospital outcomes. Journalof Nursing Administration 29 (2), 25–33.

Lowery, B.J., Jacobsen, B.S., 1984. On the consequences ofoverturning turnover: a study of performance and turnover.Nursing Research 33 (6), 363–367.

Lu, K.Y., Lin, P.L., Wu, C.M., Hsieh, Y.L., Chang, Y.Y.,2002. The relationship among turnover intentions, profes-sional commitment, and job satisfaction of hospital nurses.Journal of Professional Nursing 18 (4), 214–219.

Lum, L., Kervin, J., Clark, K., Reid, F., Sirola, W., 1998.Explaining nursing turnover intent: job satisfaction, paysatisfaction, or organizational commitment? Journal ofOrganizational Behaviour 19 (3), 305–320.

Mano-Negrin, R., Kirschenbaum, A., 1999. Push and pullfactors in medical employees’ turnover decisions: the effectof a careerist approach and organizational benefits on thedecision to leave the job. The International Journal ofHuman Resource Management 10 (4), 689–702.

McCarthy, G., Turrell, M., Cronin, C., 2002. National Study ofTurnover in Nursing and Midwifery. Department ofNursing Studies, National University of Ireland Cork.Commissioned by the Department of Health and Children.Accessed March 10, 2004, at http://www.doh.ie/pdfdocs/tover.pdf.

McNeese-Smith, D.K., van Servellen, G., 2000. Age, develop-mental, and job stage influences on nurse outcomes.Outcomes Management for Nursing Practice 4 (2), 97–104.

Michaels, C.E., Spector, P.E., 1982. Causes of employeeturnover: a test of the Mobley, Griffeth and Meglinomodel. Journal of Applied Psychology 67 (1), 53–59.

Mobley, W.H., 1982. Employee Turnover: Causes, Conse-quences and Control. Addison-Wesley, Reading, MA.

Mobley, W.H., Griffeth, R.W., Hand, H.H., Meglino, B.M.,1979. Review and conceptual analysis of the employeeturnover process. Psychological Bulletin 36 (3), 493–521.

Molassiotis, A., Haberman, M., 1996. Evaluation of burnoutand job satisfaction in marrow transplant nurses. CancerNursing 19 (5), 360–367.


http://www.doh.ie/pdfdocs/tover.pdf

http://www.doh.ie/pdfdocs/tover.pdf

Moore, K.A., 2001. Hospital restructuring: impact on nursesmediated by social support and a perception of challenge.Journal of Health and Human Services Administration 23(4), 490–517.

Moore, K.A., Greenglass, E.R., 1999. Development andapplication of the Deakin Coping Scale. Unpublishedpaper.

Mor Barak, M.E., Nissly, J.A., Levin, A., 2001. Antecedents toretention and turnover among child welfare, social work,and other human service employees: what can we learn frompast research? A review and metanalysis. Social ServiceReview 75 (4), 625–661.

Moss, R., Rowles, C.J., 1997. Staff nurse job satisfaction andmanagement style. Nursing Management 28 (1), 32–34.

Mueller, C.W., McCloskey, J.C., 1990. Nurses’ job satisfaction:a proposed measure. Nursing Research 39 (2), 113–117.

Mueller, C.W., Price, J.L., 1990. Economic, psychological, andsociological determinants of voluntary turnover. Journal ofBehavioral Economics 19 (3), 321–335.

Nakata, J.A., Saylor, C., 1994. Management style and staffnurse satisfaction in a changing environment. NursingAdministration Quarterly 18 (3), 51–57.

Needleman, J., Buerhaus, P., Mattke, S., Stewart, M.,Zelevinsky, K., 2002. Nurse-staffing levels and the qualityof care in hospitals. New England Journal of Medicine 346(22), 1715–1722.

Nicklin, W., McVeety, J.E., 2002. Canadian nurses’ perceptionsof patient safety in hospitals. Canadian Journal of NursingLeadership 15 (3), 11–21.

O’Brien-Pallas, L., 1988. An analysis of the multiple ap-proaches to measuring nursing workload. Canadian Journalof Nursing Administration 1 (2), 8–11.

O’Brien-Pallas, L., Irvine, D., Peereboom, E., Murray, M.,1997. Measuring nursing workload: understanding thevariability. Nursing Economics 15 (4), 171–182.

O’Brien-Pallas, L., Thomson, D., Alksnis, C., Bruce, S., 2001.The economic impact of nurse staffing decisions: time toturn down another road? Hospital Quarterly 4 (3), 42–50.

Parasuraman, S., 1989. Nursing turnover: an integrated model.Research in Nursing and Health 12 (4), 267–277.

Price, J.L., Mueller, C.W., 1981a. A causal model of turnoverfor nurses. Academy of Management Journal 24 (3),543–565.

Price, J.L., Mueller, C.W., 1981b. Professional Turnover. SPMedical and Scientific Books, New York.

Rafferty, A.M., Ball, J., Aiken, L.H., 2001. Are teamwork andprofessional autonomy compatible, and do they result inimproved hospital care? Quality in Health Care 10 (Suppl.II), ii32–ii37.

Robertson, R.H., Hassan, M., 1999. Staffing intensity, skill mixand mortality outcomes: the case of chronic obstructive lungdisease. Health Services Management Research 12 (4),256–258.

Scott, J.G., Sochalski, J., Aiken, L., 1999. Review of magnethospital research: findings and implications for professionalnursing practice. Journal of Nursing Administration 29 (1),9–19.

Shader, K., Broome, M., Broome, C.D., West, M., Nash, M.,2001. Factors influencing satisfaction and anticipated turn-over for nurses in an academic medical center. Journal ofNursing Administration 31 (4), 210–216.

Shamian, J., O’Brien-Pallas, L., 2001. Effects of job strain,hospital organizational factors and individual characteris-tics on work-related disability among nurses. Final Reportsubmitted to the Ontario Workplace Safety and InsuranceBoard.

Shields, M.A., Ward, M., 2001. Improving nurse retention inthe National Health Service in England: the impact of jobsatisfaction on intentions to quit. Journal of HealthEconomics 20 (5), 677–701.

Shortell, S.M., Zimmerman, J.E., Rousseau, D.M., Gilles, R.R.,Wagner, D.P., Draper, E.A., Knaus, W.A., Duffy, J., 1994.The performance of intensive care units: does good manage-ment make a difference. Medical Care 32 (5), 508–525.

Sofer, S., 1995. Determinants of nursing turnover. DissertationAbstract International: Section B: The Sciences and En-gineering 55 (9-B), US: University Microfilm International.

Song, R., Daly, B.J., Rudy, E.B., Douglas, S., Dyer, M.A.,1997. Nurses’ job satisfaction, absenteeism, and turnoverafter implementing a special care unit practice model.Research in Nursing and Health 20 (5), 443–452.

Strachota, E., Normandin, P., O’Brien, N., Clary, M., Krukow,B., 2003. Reasons registered nurses leave or changeemployment status. Journal of Nursing Administration 33(2), 111–117.

Szigeti, E., Laxdal, S., Eberhardt, B.J., 1991. Barriers to theretention of registered and licensed practical nurses in smallrural hospitals. Journal of Rural Health 7 (3), 266–277.

Tai, T.W.C., Bame, S.I., Robinson, C.D., 1998. Review ofnursing turnover research, 1977–1996. Social Science andMedicine 47 (12), 1905–1924.

Tate, U., 1996. Cross-cultural validation of measures of socialsupport. Psychological Reports 79, 271–274.

Teahan, B., 1998. Implementation of a self-scheduling system: asolution to more than just schedules!. Journal of NursingManagement 6 (6), 361–368.

The Advisory Board Company, 2000. Reversing the flight oftalent. Executive Briefing, vol. 1. Nursing Executive Center,Washington, DC.

Tourangeau, A.E., Giovannetti, P., Tu, J.V., Wood, M., 2002.Nursing-related determinants of 30-day mortality forhospitalized patients. Canadian Journal of Nursing Re-search 33 (4), 71–88.

Tzeng, H.M., 2002. The influence of nurses’ working motiva-tion and job satisfaction on intention to quit: an empiricalinvestigation in Taiwan. International Journal of NursingStudies 39 (8), 867–878.

United Nations Conference on Trade and Development, 1998.World Health Organization, International Trade in healthService a development perspective. Geneva.

Unruh, L., 2003. Licensed nurse staffing and adverse events inhospitals. Medical Care 47 (1), 142–152.

Vetter, E., Felice, L.D., Ingersoll, G.L., 2001. Management.Self-scheduling and staff incentives: meeting patient careneeds in a neonatal intensive care unit. Critical Care Nurse21 (4), 52–59.

Waldman, J.D., Kelly, F., Arora, S., Smith, H.L., 2004. Theshocking cost of turnover in health care. Health CareManagement Review 29 (1), 2–7.

Whitman, G.R., Kim, Y., Davidson, L.J., Wolf, G.A., Wang,S.L., 2002. The impact of staffing on patient outcomes


across specialty units. Journal of Nursing Administration 32(12), 633–639.

World Health Organisation, 2004. Recruitment of healthworkers from the developing world. Executive board 114session. Agenda item 4.3 Geneva WHA, www.who.int/gb/ebwha/pdf_files/EB114/B114_5-en.pdf.

Yeatts, D.E., Seward, R.R., 2000. Reducing turnover andimproving health care in nursing homes: the potential effectsof self-managed work teams. Gerontologist 40 (3), 358–363.

Yin, J.T., Yang, K.A., 2002. Nursing turnover in Taiwan: ameta-analysis of related factors. International Journal ofNursing Studies 39 (6), 573–581.

Yoder, L.H., 1995. Staff nurses’ career development relation-ships and self-reports of professionalism, job satisfaction,and intent to stay. Nurse Researcher 44 (5), 290–297.

Zboril-Benson, L.R., 2002. Why nurses are calling in sick: theimpact of health-care restructuring. Canadian Journal ofNursing Research 33 (4), 89–107.

Zimmerman, S., Gruber-Baldini, A.L., Hebel, J.R., Sloane,P.D., Magaziner, J., 2002. Nursing home facility riskfactors for infection and hospitalization: importance ofregistered nurse turnover, administration, and social fac-tors. Journal of the American Geriatrics Society 50 (12),1987–1995.


http://www.who.int/gb/ebwha/pdf_files/EB114/B114_5-en.pdf

http://www.who.int/gb/ebwha/pdf_files/EB114/B114_5-en.pdf

1-s2.0-S0020748905000477-main

Documents

deakin coping

advisory board

united nations

intrinsic

decreased

meet family

international

unmet career