UBC Faculty Research and Publications

Nurse deployment patterns : examples for health human resource management Kazanjian, A.; Pulcins, I.; Kerluke, K. Feb 28, 1990

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-Kazanjian_A_et_al_Nurse_deployment.pdf [ 16.4MB ]
JSON: 52383-1.0075923.json
JSON-LD: 52383-1.0075923-ld.json
RDF/XML (Pretty): 52383-1.0075923-rdf.xml
RDF/JSON: 52383-1.0075923-rdf.json
Turtle: 52383-1.0075923-turtle.txt
N-Triples: 52383-1.0075923-rdf-ntriples.txt
Original Record: 52383-1.0075923-source.json
Full Text

Full Text

NURSE DEPLOYMENT PATTERNS:Examples for Health Human Resources ManagementHMRU 90 :1Health Manpower Research UnitDivision of Health Services Research and DevelopmentOffice of the Coordinator of Health SciencesThe John F . McCreary Health Sciences CentreThe University of British ColumbiaVancouver, B.C., CanadaV6T lZ6A. KazanjianI. PulcinsK. KerlukeFebruary 1990This report is one of a series describing the distribution of health humanresources in British Columbia. These reports, prepared for the HealthManpower Working Group of the Ministry of Health, are working documents andcomments or suggestions are welcome.~.,THE UNIVERSITY OF BRITISH COLUMBIADivision of Health ServicesResearch and DevelopmentOffice of the Co-ordinator,Health Sciences*400-2194 Health Sciences MallVancouver, B,C. Canada V6T 1Z6February 6, 1990Ms. Vicki Farra11yChairpersonHealth Manpower WorkingMinistry of Health1515 B1anshard StreetVictoria, B.C.V8W 3C8Dear Ms. Farra11y:GroupTelephone (604) 228-4810Fax (604) 228-2495It gives me great pleasure to submit to the Health Manpower Working Groupthis second report based on the Nurse Manpower Study, commissioned by theDeputy Minister of Health in 1987.We believe this study provides a new analytic tool for the examination ofissues pertaining to the management of nursing human resources, and shouldprove useful in the planning of future requirements.As usual, we look forward to suggestions and comments from the members ofthe Health Manpower Working Group, as well as other readers of this report.Y~:%~:1I12,,_"_--Arminee Kazanjian, Dr.Soc.Associate DirectorDivision of Health ServicesResearch and DevelopmentAssistant ProfessorDepartment of Health Careand EpidemiologyAK:daEne1.The John F. McCreary Health Sciences CentreAcknowledgementsWe wish to extend our appreciation to BCHA for providing the payroll data onwhich the analysis in this report is based.TABLE OF CONTENTSIntroduction and Purpose 1Literature Review 3Literature on Health Organizations 4Literature on Nurse Shortages 5Organization Design and Nurse Shortages 8Background to the Study 10Management of Nursing Resources 11Determinants of Efficient Personnel Deployment 13Data Development and Source 16Methods 18Patterns of Nursing Resource Deployment in British Columbia 22Implications for Health Managers 33References1Introduction and PurposeWhile the debate on the magnitude of current nurse shortages inhospitals - whether the province faces an absolute or relative shortage ­is far from being resolved, it is clear that corrective measures areneeded to first alleviate and eventually eliminate the problem. For thelong-term, proactive measures need to be adopted to prevent the recurrenceof the problem. For several reasons, these remedial measures are morelikely to be successful if directed at changing conditions internal toeach hospital. Implementing change on a small (i.e. at the agency level)scale is a lot more feasible than undertaking system-wide change.Furthermore, the organizational context is a key determinant inunderstanding labour force behaviour, and hence, should be the locus ofaction in addressing such problems. Finally, it is unlikely that, in theshort term, the management of any single hospital (multi-hospital society)could sufficiently influence environmental conditions external to theirinstitution in order to improve their current situation.Hospitals have traditionally been predisposed, for various reasons,to deal with nursing shortages by emphasizing recruitment rather thanretention of personnel. Similarly, studies describing nurse supplyusually focus on points of entry to (and exit from) the "active pool" ofemployed nurses and often ignore the most important component of thatpool; the cohort that remains in the workforce from one year to the next(Figure 1). High attrition and turnover rates have been consideredunavoidable because nursing is a female-dominated profession and women'sparticipation in the labour force is highly contingent on their familyFigure 1: Nurse Supply Model'" .""InactivePersonnel--- - - _.t>CurrentI t>I Penonnel.-t> I--I ~ ,NewPersonnelI Adapted from: Kazanjian et ai, -Modelling the Supply of Nurse Labour- .Medical Care.December 1986, 24(12)1067-10833roles. However, the health care sector's past record of reliance upon in­migration of nurses , in addition to incorporating modest increases in thenumber of nurses trained in-province (HMRU 88:7, 1988) has beeninefficient at best, and certainly has proven to be ineffective duringperiods of fiscal restraint .The purpose of this paper is to demonstrate that to invest in the"front end" of the management process is the best approach to thisproblem. Preventing human resource problems by optimizing regular staffdeployment is more cost-effective than reacting repeatedly to shortages ofqualified personnel through casual staffing. The thrust of the analysisis to examine and describe the staffing requirements of various patternsof nurse deployment and to develop a management model for maximizingcurrent staff resources. In addition, this report provides information onways to enhance the management of nursing resources through the systematicmonitoring of personnel retention.Literature ReviewWhile the literature on human resources management is quitevoluminous, it is appreciably more limited when it is concerned with themanagement of health human resources and even more contained when the areaof interest is nursing human resources. It should be noted as well thatU.S. studies in health human resources management are not oftengeneralizable to the Canadian context, due to differences in health care4funding and organization between the respective systems. The literaturereview indicated two very distinct fields of research related to the studyof nurse personnel management issues.Literature on Health OrganizationsThe first was comprised of publications in the field oforganizational theory and organizational behaviour which examineimplications of those theories and behaviours for management in the healthsector (Scott, 1981; Hage, 1980; Weber, 1964; Mooney, 1947; among others).This literature is based largely on the systematic investigation ofdifferent conceptions of how organizations work, and its purpose is tobroaden theories of strategic planning and strategic management . Thepractical relevance of this literature to nurse human resources managementwas rather limited, in view of the numerous steps of inferential reasoningrequired to apply to nursing (one part of the complex healthcare sector)management theories pertaining to complex organizational structures.One model of organizational behaviour described in this literature isthe closed-system model which is based on the assumption that the mostimportant features of organizations have to do with their internalstructures and processes . Its opposite is the open-system model which isin turn based on the assumption that an organization's behaviour is bestunderstood by taking into account its environment (Shortell and Kaluzny,1988). Both models can be usefully applied to the current situation inB.C. (as elsewhere), to examine the reasons for fluctuating imbalances innursing human resources . Unfortunately, no such effort exists. Our5review pointed to a serious gap in research concerned specifically withtheories in the management of nursing resources; this is particularlyalarming given the relatively large role this profession plays in thedelivery of health care.Literature on Nurse ShortagesA second approach for the study of nursing resources management wasgleaned from the literature concerned with nurse shortages and itsimplications for staffing and scheduling of nursing departments. Thisliterature is almost exclusively confined to nursing journals and, whileit draws from other disciplinary perspectives such as economics andsociology, it is generally limited in theoretical breadth, but compensatesby its depth of analysis. As might be expected, the volume of thesepublications is inversely proportional to market conditions. That is tosay, there is less interest in studying nurse staffing issues duringperiods of relative surplus. Hence the previously mentioned proclivityfor the consideration of recruitment problems rather than the examinationof long-term measures for alleviating the situation, such as improvedretention and innovative management of human resources.Several studies from the second group are relevant to the situationin British Columbia, and describe rather simple management techniques thatcould result in the adoption of preventive measures. One such studypurports to demonstrate how monitoring turnover indices using hospitaladministrative records might have predicted a critical shortage of staffand prevented bed closures in the study hospital (Mann and Jefferson,61988). An attitude survey of nurses and supervisors in this hospitalcomplements the monitoring data and identifies critical components of theretention problem.Using measurement indices adopted from another study (Duxburg andArmstrong, 1982) I the study by Mann and Jefferson examines four turnoverindices for the years 1982 to 1986 in a Medical Intensive Care Unit(MICU). The Instability Rate measured the percentage of staff at thebeginning of the year who quit during the year; the Turnover Rate was ameasure of the number of nurses who quit relative to the average number ofnurses employed in the year; the Wastage Rate was the percentage of newlyhired nurses who quit during their first year; and the Mean Service ofLeavers was the average length of service among those who had quit. Thedata indicate that the rates for Instability and Turnover were similarwhile the Wastage rate was lower than both for 1982 to 1984, followed bysharp increases for all in 1985, signalling a severe degree of staffturbulence. These data, if collected, as part of routine monitoring ofhuman resources, would provide accurate advance warning of potentialproblems. The survey results indicated significant agreement amongrespondents on reasons for quitting the MICU. Two-thirds (67%) of thosewho had left indicated that if they were given the opportunity to changework conditions, they would implement "adequate staffing" measures .Another group of studies examines an alternative method of staffingwhich uses internal resource teams such as designated casual pools (Galeand Roark, 1985; Stenske et al, 1988). This method of staffing allows7flexibility of schedules and areas of practice, and provides financialincentives in the form of benefits or salary differentials for nurses'availability. While resource teams (RTs) are not a new concept in thisprovince (Pacific Health Forum, 1987) they are by no means a widespreadphenomenon. Nor has the idea been seriously evaluated as to its impact onnurse shortages. A recent U.S. study of a representative national sampleidentified the main characteristics of state-of-the-art RTs (Stenske eta1, 1988). Among the 11 study recommendations directed to managers andpertaining to the use of RTs six are particularly relevant to BritishColumbia: provide multiple options for employment; designate a nursemanager for the RT, establish minimum work schedule requirements; allowflexible scheduling and a choice of shifts; and, design an individualizedorientation program, with regular evaluation of RT members.Another group of recent studies on nurse shortages advocates theemployment of innovative methods of staff retention (Longo and Uranker,1987; Marquis, 1988; Wall, 1988). The authors argue that proactivestrategies for retention are not only desirable to ensure quality of carebut also happen to be cost-effective. Furthermore, the need for areliable database to monitor retention and its correlates is demonstratedand discussed in these studies. Without systematic evaluation of specificretention efforts, informed judgements about successful retention programscannot be made nor can hospitals learn from each others' experience(Weisman, 1982).8Organization Design and Nurse ShortagesFinally, a few research publications make a direct associat~nbetween organizational behaviour and nurse shortages. These studies aregrounded in organizational theory and explore organizing principles andensuing problems (McClure, 1982 and 1984; Kramer, 1988; Loveridge, 1988).McClure (1982) reported on a study designed to identify those hospitalsthroughout the u.S. that had reputations for being good places to work andhad been successful in attracting and retaining professional nurses. Inaddition, certain criteria regarding retention and turnover rates,proportion of RNs on staff and ratio of nurses to patients also had to bemet. Other work by the same author (McClure, 1984) relates modernmanagement theory and practice to the field of nurse human resourcesmanagement. Purpose, tasks, people, level of care and type of agency,technology, and structure are factors described as internal to theorganization and to be fully understood by management, not only asindependent factors but in concert with one another. External factorsthat must also be considered by managers include the economy, politicalpressures, legal aspects, sociocultural characteristics, and technologydiffusion.Until very recently hospital managers were mainly concerned withinternal structures (closed-system model) and were not concerned aboutexternal factors (open-system model). While it is true that nursemanagers may have little direct control of external factors, they shouldnevertheless be conscious that human resource management clearly affectsand is affected by the environment. Contingency theory, which posits that9organizational structure is contingent on environmental factors, hashelped some researchers in their understanding of the relationship betweenorganizational structure and staff nurse retention. A study by Loveridge(1988) tests the tenets of contingency theory (Perrow, 1967) which suggestthat a more bureaucratic, "mechanistic" form of organization is moreeffective when the environment is simple and stable, whereas a more"organic" form of organization is likely to be effective when theenvironment is complex and dynamic. Loveridge suggested that thecombination of more flexible organizational structures with more complextechnologic responsibilities is associated with a lower rate of staffnurse turnover.A more recent analysis of organizational conditions which areconducive to improved retention (Kramer, 1988) undertook a comparison of16 magnet hospitals with the 'best run' companies in a corporate communitydescribed in a 1982 study similar to the one on Magnet Hospitals (McClure,1982). The findings concluded that the same characteristics are found inmagnet hospitals and best run corporate communities. The analysisidentified two perspectives of the nursing shortage - internal andexternal - of which internal shortages are created by conditions thatexist in many hospitals (in varying degrees) for which the hospitals aremore or less responsible, such as the use of large numbers of float oragency nurses and inadequate support to nurse manpower. Externalshortages are caused by insufficient numbers of trained personnel,increased need for nurse manpower due to higher technology, or populationaging, etc., among others. In addition, internal nurse shortages create10and magnify external nurse shortages. What the magnet hospitals did wasto create conditions that obviate internal nurse shortages. Consequently,if and when hit by an external shortage these hospitals are likely to feelit to a much lesser degree than those which have not created conditionswhich overcame the internal shortage.Clearly, this literature review is neither extensive norcomprehensive. The major point to be made, however, is that there arelessons to be learned from the examination of organizational structuresand environmental factors which circumscribe nurses' labour marketbehaviour.Background to the StudyThis study builds on our previous work included in The Nurse ManpowerStudy (Pu1cins, Kazanjian, and Ker1uke 1988). The earlier work was partof a three-volume study commissioned by the Deputy Minister of Health in1987 and its purpose was to assess the current status of nurse manpower,as indicated by a synthesis of currently available data. Therecommendations included in that report addressed the dual focus of theDeputy's study: the extent of manpower imbalances and the quality of thesupporting evidence.In brief, the 1988 study refined an earlier model for estimatingregistered nurse requirements (Kazanjian and Chan, 1984), by developingthe methodology for expressing requirements in terms of total number ofpersons as opposed to the less specific measure of fu11-time-equiva1ents11(FTE's). More importantly, this 1988 study examined the effects ofnurses' movement in the labour force and elaborated on the sensitivity ofthe health care system to fluctuations in both employment mix (that is,the mix of regular versus casual RNs), and deployment patterns (consistingof average annual hours of work).Interesting results emerged from the development of hypotheticalscenarios depicting changes in the composition of Regular and Casual staffas well as changes in the average hours of paid services by each group.This method of constructing scenarios can be used to test particular humanresources management decisions or intervention strategies. The currentstudy develops and further elaborates these management models.Management of Nursing ResourcesMany documented methods of evaluating the nursing shortage, quiteappropriately tend to focus on "external" or system-based roots andexacerbating factors in an attempt to alleviate any current "crises". Yetour previous research (Nurse Advisory, 1988) has also pointed to theimportance of a number of "internal" factors that are central to therelationship between nurse requirements in a particular facility (ormunicipality or region, for example) and the employment mix and personneldeployment patterns in that facility. Accordingly, the analysis presentedhere will explore the existence of different internal factors through areview of current personnel management practices in groups of facilities,and utilize this knowledge to maximize efficient deployment of existingpersonnel.12Whereas the number of nursing hours that will be required in ahospital or other medical facility is largely dependent on approved bedcapacity and level and type of care offered, the size of nursing staff per~ during the course of a year may fluctuate widely by type of facility aswell as by employment profile of nursing staff. It is clearly recognizedthat the number of budgeted positions does not, in any event, translateclearly to the number of individuals needed to fill these positions.Depending on the mix of full-time, part-time and casual nursing staff(employment mix), the average number of paid hours for each of these threegroups, and the level of stability of nursing workforce, the number ofnurses required may be quite different according to facility type, size orother variable. Indeed, even the characteristics of the surroundingenvirons (e.g. a family-oriented suburb in contrast with the moretransient inner-city) will to some extent determine the ratio of nursinghours to number of nurses required "on payroll" during a given period.Having identified these factors one develops the ability to evaluate moreaccurately the effect of such factors on the number of nurses that must berecruited every year. This knowledge may also be utilized to monitortheir current effects, manipulate these parameters, and map out futurestrategies to effectively improve the utilization of existing nursingpersonnel and reduce the need for intensive, and expensive, recruitingcampaigns.The ability to understand and explain the mechanisms that are centralto maximizing existing human resources in nursing, and moreover, to matchthe "right nurse" (i.e. with the appropriate level of clinical and/or13academic training) to the "right position", is particularly vital totoday's seemingly tenuous nurse manpower situation. The combined effectof high turnover in many facilities, reported workplace dissatisfaction(Layton, 1988) and the resultant "dropping out" of the nursing professionfor many, either on a permanent or temporary basis, has meant increasingdifficulty in filling nursing vacancies (DTF, September 1989). Theseproblems are compounded by intensified recruitment drives drawing on arelatively diminishing pool of potential supply, that is, increasedpersonnel requirements relative to supply. For these reasons alone, it isnecessary for managers to focus their attention on the efficientdeployment of existing staff by implementing more flexible personnelpolicies. Whereas in most cases it would certainly not be desirable todecrease the number of budgeted Fu11-Time-Equiva1ents (FTE's) or totalnursing hours to meet arising shortages, the need to reduce the number ofindividuals required to provide a given quantity of nursing care is highlydesirable.Determinants of Efficient Personnel DeploymentThe primary determinants of efficient personnel deployment arepresented schematically in Figure 2. Three sets of factors have beenidentified, each expressed as a bipolar continuum. The three-dimensionalnature of this representation not only reinforces the need to separatelyassess each of the factors presented here, but more importantly, thattheir inter-relatedness be seriously considered in the planning of nursingresources .14The first component considered in this management model is workforcestability. Simply expressed, the size of the total nursing workforcerequired to fill a given number of nursing hours will be considerablysmaller in a stable workplace environment than where recruitment for thesame position or set of positions must occur repeatedly over a givenperiod. Workforce stability may be considered on a variety of levels, andis typically undermined by either movement between hospitals (frequentlateral career movement in lieu of vertical movement, or perhaps simply aresponse to unfavourable workplace conditions), inter-regional migration,or movement out of nursing (into premature retirement or another choice ofcareer). Where the level of attrition exceeds that which can normally bereplaced by the flow of new graduates into the workforce, this type ofmovement may be labelled as "wastage".The second dimension in the model may be defined as the existingemployment mix, or the full-time employee ratio, as measured against thecontingent of part-time and casual nursing staff. A relatively highproportion of full-time staff again reduces the number of nurses requiredon payroll to fill a given workload. Although when viewed as a purelyarithmetic exercise, a full-time ratio approaching 1 would be consideredoptimal, it is recognized that a variety of tangible and intangibleconsiderations require a different proportion of full-time staff from onefacility to the next. In any event, the full-time ratio can have a quitesignificant impact upon recruitment efforts and therefore, upon unfilleddemand.15The final component of the model, average paid hours, describes themobilization of the paid workforce, and indirectly, the effect of unpaidleave and unpaid vacation days on personnel deployment. Our analyses showthat this is most clearly manifested with casual staff, who collectivelywork an appreciably wider range of hours as compared to the more set hoursrequired of full-time and part-time staff. But as evidenced here, eventhe average number of paid hours of full-time staff, or the proportion ofan FTE worked by each, may fluctuate significantly between facility orregional groupings.The "management cube" pictured in Figure 2 illustrates the interplaybetween these three dimensions. At one extreme, the "optimal" nursedeployment scenario is characterized by stability, or very little movementin or out of the workforce, a relatively high proportion of full-timestaff members, and a high number of average hours by each staffclassification. In contrast, the "pessimal" situation falls in adiametrically opposed position within this model. It is characterized bya high degree of movement within the workforce, a low full-time employeeratio and relatively low average number of paid hours per employee,thereby further increasing the number of nurses required to fill existingbudgetary FTE requirements. Of course, the majority of the observed casesfall at some other point in this cube, and reflect the mix of "high" and"low" ratings on these three bipolar continua. It is precisely thesecases that present the most interesting and relevant scenarios. Thefacility typified, for example, by even an extraordinarily high workforcestability will still require more nurses per FTE if their workforce is16compressed of a predominantly part-time, and especially casual staff, atthe expense of a sizeable full-time pool. Or, consider the case of afacility boasting a large full-time contingent with a high average numberof hours for all its staff, but which must continually undertakerecruitment drives due to low workforce stability (high turnover) by itsnursing staff.When operationalized, this management model serves as an evaluativetool which may be applied at the facility, local or regional levels.Specifically, the dimensions featured within the model may be applied tothe analysis of sensitivity of the system (i .e . a region or facility) toemployment mix, patterns of nursing resource deployment and labourmotility and wastage. In the following section, we apply this tool to thesituation in British Columbia , and calculate indices measuring therelative status of several types of "peer group" hospitals on the threebipolar scales featured in the model.Data Development and SourceThe application of Figure 2 to the nurse manpower situation inBritish Columbia requires data which would adequately represent allhospital types in the province (in light of the requirements for thelarger study within which this paper is couched) and be able to capturethe degree of workforce stability (movement), the proportion of regular,full-time staff, and the average number of paid hours per nurse onpayroll. Centralized payroll systems, such as the one provided by BritishColumbia Health Association (BCHA) , contain such information and areFigure 2: Schematic Representation of FactorsDetermining Efficient Personnel DeploymentIAverage Paid HoursI.. '...............~.9."...............Efficiency_ OptimalmID Pessimal...........18essential to monitoring workforce behaviour and management practice. Wehave used the most recently available at the time of the study, the yearending December 1986.The first step in the analysis of these data involved the examinationof workforce characteristics. During the course of data preparation anddevelopment, however, it became clear that examples of individualschanging job status, terminating employment or temporarily dropping out ofthe nurse workforce during the period under study would not provideaccurate and useful measures of nurse deployment because individuals couldeasily be double and triple counted, thus inflating the actual number ofindividuals involved. At the same time, it was recognized that thesedeviations from the "ideal" manpower scenario very clearly reflected theactual state of the nurse labour market in British Columbia . It was,therefore, very important to be able to accurately measure situationswhere there was frequent movement and contrast these to situation wheremovement was minimal .MethodsAlthough there has been much discussion regarding the movement ofnurses in and out of the workforce, this was to our knowledge represents afirst attempt to quantify movement in a much broader sense.The data indicated that in addition to changing employers, nurseswere sometimes simultaneously employed by more than one hospital. Allcombinations of employment status were possible; for example, a nurse19could be employed full-time in one hospital and on call-in basis (casual)in another, or, have two part-time positions, etc. Conversely, a nursecould change employment status during the year while remaining with thesame employer. Of course, the normal flow in and out of the workforce wasalso captured by these data. Thus, a nurse could have commenced orterminated employment during the course of the year and, therefore, wasnot fully productive that year.In contrast, the data indicated that some nurses were employed at onehospital all year, and even if changes in status had occurred, there hadbeen no discontinuities in employment.We therefore developed a methodology which would best describe twofacets of nurse resources management. Two scenarios were developed. The"static" or "ideal" scenario necessitated the exclusion of all recordswhich were not continuous in the year. According to this formula, allcases which switched employment status, or either terminated or commencedemployment during the course of the year were excluded from the sample.The "dynamic" or "real" scenario was constructed in a differentmanner. In this case, if the number of hours for an individual nurse wassplit between employment statuses, they were proportionally distributed.Likewise, the appropriate fractions of individuals who were not continuousemployees during the whole year were considered. Finally , if individualnurses had not been eliminated from the database at this point, and werefound to be working in more than one hospital within the system studied,20the total number of hours split between hospitals was verified and thecorresponding "proportion" of the individual was partitioned likewise. Asa result, movement in the workforce is fully captured in this analysis.Overall, the manner in which records are counted is the same for bothscenarios; however. the record selection process between the two differsmarkedly. The final step in this phase of data development was anexamination of the difference between the "static" and "dynamic"scenarios. The difference showed the degree of motility in anyonesetting and could be compared across settings. This measure provided uswith an accurate index of workforce stability (Figure 2) .The indices measuring both average number of paid hours per statuscategory and proportion of full-time nursing staff are self-evident andcomputationally simple . Total number of nursing personnel was used as thedenominator for the latter, and total number of paid hours by status forthe former. The data indicated that there were large variations in theproportionate distribution of Regular versus Casual components of thenurse workforce. It was noted that some employers tended to be heavierusers of casual staff than others . Similarly, there were variations inthe breakdown of Regular staff to its Full- and Part-time components .Closely related to this dimension, yet conceptually separate is themeasurement of the average number of paid hours per nurse, within eachemployment status . While it is expected that appreciable variation willexist within the Casual component, the data indicated that some variationexisted, by hospital, among Full -time staff. Thus, a full-time nurse is a21full-time nurse, is a full-time nurse, did not hold, the resultsdemonstrated variability even within a single hospital.For the analysis described here, these indices were calculated fornine different groupings of "peer hospitals", as well as the aggregatemeasures for the province of British Columbia as a whole. The hospital"peer groups" were identified on the basis of selected factors which were,on an a priori basis, deemed to be characteristic of major groups offacilities in the province. These include location (metropolitan, non­metropolitan, suburban, urban), function (e.g . teaching vs . non-teachinghospitals). size and bed type (extended vs. acute care). Clearly, thesecategories have some overlap, but that was not unintentional and did notin any way mislead the interpretation of the results.The final product of this exercise is a set of three indices whichmeasure the factors identified in the management model. These serve twofunctions. First, it is possible to evaluate the relative deploymentperformance of different groups of hospitals in the province. Secondly.one may then change the parameters according to hypothetical criteria, andestimate the impact of new personnel and/or managerial policies at thefacility (or regional) level on the need for recruiting additional nursingresources.In summary, the development of the management model (Figure 2)facilitates the understanding of how shortage situations may arise and maybe addressed, at least in part, through the proactive management of22resources. The model also provides a succinct conceptualization of thethree major dimensions contributing to the optimization of nurse humanresources.Patterns of Nursing Resource Deployment in British ColumbiaThe analysis of patterns of nursing resource deployment in any givenfacility or set of facilities calls for both the uni-dimensiona1 that is,each dimension taken alone, as well as multi-dimensional (i.e. takentogether) consideration of the three components of the management modelpresented here. In this section, we will evaluate the management ofnursing resources through the examination of several like hospitalgroupings in British Columbia. Following this, scenarios will beconstructed in order to demonstrate alternate management strategiesdesigned to both plan and enhance current personnel deployment strategies.In the first instance, employment status rates and average number ofpaid hours by status are displayed by peer-grouped hospitals in Table 1.The data show that in terms of the employment status ratio (Fu11­time:Part-time:Casua1) , the "urban" group displayed a markedly higherutilization of full-time staff at 0.791 (or 79%) vs. 0.599 for the lowgroups, and more restricted deployment of casual and especially part-timestaff, therefore appearing to have a more efficient pattern of manpowerdeployment than other groupings. On the other hand, those hospitals withmore than 40 percent extended care beds displayed a tendency to a highproportion of part time nursing staff at 0.232 (or 23%), and acorrespondingly low proportion of full-time nursing staff.23TABLE 1PERSONNEL DEPLOYMENT INDICES IN SELECTED HOSPITALS IN BRITISH COLUMBIAEMPLOYMENT STATUS RATIO AVERAGE HOURSHospital Group Full-Time Part-Time Casual Full-Time Part-Time CasualMetropolitan 0.681 0.168 0.151 1632.8 1137.6 602.2Non-Metropolitan 0.640 0.183 0.177 1666.2 1173.0 619.2GVRHD, >500 Beds 0.667 0.178 0.155 1582.6 1145.4 625.2GVRHD, <500 Beds 0.603 0.213 0.184 1709.0 1165. 1 527.6GVRHD Non-Teaching 0.636 0.195 0.169 1601 .6 1122.4 605.0GVRHD Teaching 0.698 0.161 0.141 1564.7 1176.1 652.8>40% Extended Care Beds 0.599 0.232 0.169 1629.4 1174.9 552.7Suburban 0 .624 0.209 0.167 1640.7 1140.2 607.0Urban 0.791 0.082 0.127 1540.9 1132.8 608.124While the patient population in extended care beds is more stablethat in acute care, it should be noted that these facilities with morethan 40 percent extended care beds were most likely to be located insuburban areas and, therefore, replicate closely the employment mix ratiosof the latter. In contrast, however, the measure on average hours wasquite different. The highest proportion of casual staff could beattributed to the smaller hospitals in the Greater Vancouver RegionalHospital District 0.184 (or 18%). Clearly, and perhaps somewhatsurprisingly, the data show that there do, in fact, exist substantialfluctuations in these ratios between British Columbia hospitals.Furthermore, the results suggest that this may be an "internal" factorthat maybe at least partially manipulated by personnel deploymentstrategy.Since an all full-time complement may not be the most efficient wayto staff a facility, nurse deployment patterns were examined in detail.The actual average number of paid hours per employment status revealedfluctuations similar to the previous factor. When measured in thismanner, a somewhat different picture of "efficiency" regarding personneldeployment emerges from these data. In this case, the urban hospitalsrated relatively low despite their large full-time contingent . Full-timestaff worked the lowest number of hours (at 1,540) than any of the otherpeer-grouped hospitals, despite their commendable high ratio of full-timenurses compared to part-time or casual staff. At the same time, the samehospitals, which ranked high in proportion of casual staff (small25hospitals, GVRHD) , obtained the highest number of average hours (at 1,709)from their full-time nursing staff.To render these results more meaningful one must go one step furtherto combine these indices and view them on at least a two-dimensionalplane. On the basis of these indices, it is possible to compare theefficiency of manpower deployment of peer-group hospitals using astandardized criterion. Here, results are standardized by calculating thetotal number of nurses required to provide 200,000 hours ofundifferentiated nursing care. Although this model does not differentiatebetween nursing specialties or the different levels of care which mayaccount for at least some of the observed differences between hospitals,it does nevertheless illustrate the possible effects of such variationsbetween facilities. The figure of 200,000 hours was chosen specificallybecause it was a "reasonable" approximate figure for many of thefacilities under study. It was representative of most regional hospitalsand some of the larger community hospitals. In addition, nurse managersare often in this position when they are given a budgetary allocation butno other information with which to plan their human resource requirements.The results of this evaluative exercise are presented in Table 2,which illustrates very clearly the variable number of nurses required tostaff 200,000 hours, taking into consideration the "normal" movement inthe workforce previously discussed for each of the groupings. The tableshows that it would take a minimum of 158 and a maximum of 176 individualTABLE 2NURSES REQUIRED PER 200,000 NURSING HOURSIN SELECTED HOSPITAL GROUPS IN BRITISH COLUMBIANURSES REQUIREDHospital Group Full-time Part-time Casual TotalMetropolitan 84 30 50 164Non-Metropolitan 76 32 58 166GVRHD, >500 Beds 84 32 50 164GVRHD, <500 Beds 70 36 70 176GVRHD Non-Teaching 80 34 56 170GVRHD Teaching 90 28 44 160>40% Extended Care Beds 74 40 62 174Suburban 76 36 56 168Urban 102 14 42 15827RNs to provide 200,000 hours of nursing services, given the observedemployment status ratios and average hours of the respective groupings .The difference of 18 nurses constitutes 11 percent of the staffcomplement . The combined effect of both employment status ratio andaverage paid hours per nurse over a period of one year has been that theurban peer-grouped hospitals require the least number of nurses over theperiod of one year to fill the required number of nursing hours, mainlydue to the large proportion of full-time staff. In contrast, smallhospitals in GVRHD « 500 beds) require the highest number of RN staffdespite their top ranking in average hours worked by full-time staff . Itshould be noted that casual staff in this type of facility provide verylow average hours and thus contribute to the inflation of the totalfigure. In non-metropolitan settings, while both the number of full-timestaff and their average hours are similar to those of small hospitals, thedifference in casual average hours results in a smaller staff complement.But this still does not tell the whole story. Measuring a facility'sdeployment of nursing personnel against the third dimension of the model,workforce stability, may paint an entirely different picture than a two­dimensional analysis alone. This index is operationa1ized in a slightlydifferent manner from the above two measures. In this case, a motility,or wastage factor is calculated for each employee type (full-time, part ­time, casual) based on both observed paid hours for all staff and hourswhich are attributable only to those nurses who have been continuousemployees throughout the year under analysis. In other words, they havenot changed employers or dropped out of the workforce during the course of28the year under analysis. Specifically, this wastage factor measure is theproportional difference between the actual, total observed number of paidhours (which, of course, incorporates all individuals dropping into andout of the work force throughout the year) and the number of paid hoursfrom which have been filtered out all hours paid to noncontinuousemployees. This difference, then, represents the degree of wastage withinthe system because the labour market behaviour of the stable subgroupwithin each grouping represents the "gold standard" against which everyonecan be measured. As the index nears a (a score of a is only a theoreticalpossibility), wastage and movement in the system are minimized. It isimportant to remember that this is not an absolute measure . Nor does itmeasure retention, or conversely, turnover, per se especially since thisis a measure of movement in a limited time period (i.e. 12 months). Long­term retention may be inferred from this measure only if the period understudy can be ascertained to be a typical 12-month period. Rather, this isa relative measure of the approximate proportion of paid hours by "stable"versus "instable" or discontinuous staff. Expressed as a percentage, thisstatistic is rendered meaningful when compared to that for otherfacilities or to a regional aggregate, for example .The results from the peer-groups hospitals in British Columbia revealinteresting results (Table 3). Overall , the lowest wastage level wasexhibited by those facilities with a high proportion of extended care beds(23%), non-metropolitan hospitals (22%), and hospitals with less than 500beds and non-teaching hospitals in the Greater Vancouver Regional HospitalDistrict (31%) . The highest level of motility, or wastage, was seen to29TABLE 3NURSING PERSONNEL INSTABILITY IN SELECTED HOSPITAL GROUPS (%)Hospital Group Full-time Part-time Casual TotalMetropolitan 23.53 24.83 44.89 26.96Non-Metropolitan 17.26 19.84 42.99 22.28GVRHD, >500 Beds 24.94 22.75 43.22 27.38GVRHD, <500 Beds 18.14 20.97 45.17 23.71GVRHD Non-Teaching 21.51 20.06 39.92 24.33GVRHD Teaching 28.24 26.18 47.38 30.60>40% Extended Care Beds 18.80 19.05 41.10 22.63Suburban 25.89 22.43 43.78 25.89Urban 31. 79 28.71 42.99 31. 7930occur in urban hospitals (32%), closely followed by teaching hospitals(31%) (in the GVRHD). This is perhaps reflective of either personnelpolicies that results in workforce instability, an urban setting housing amore transient population or a combination of these and other factors.Expectedly, wastage was much higher in the casual sector in all cases,while those for full-time and part-time nurses did not differ markedlywithin each facility grouping. This analysis demonstrates the greatersensitivity of the market in British Columbia to fluctuations in thesupply of casual RNs than regular ones, since every position staffed bycasuals will bring an appreciably greater number of individual nursesthrough the system than will regular positions .In these examples we cannot say anything about management orpersonnel deployment practices of the hospital grouped the basis ofarbitrarily selected characteristics . Rather, we look at similarities anddifferences in deployment patterns of hospitals with certain likecharacteristics, and since we do not know from these data what theirmanagement practices have in common, if anything, it is not possible tocomment on whether one group advocated more astute human resource policiesthan any other. It is possible, however, for facilities to compare one'sown performance with that of other similar hospitals in one'sjurisdiction.Nevertheless, through the construction of scenarios to test alternatemanagerial policies , one can utilize the measures presented here as anevaluative tool in a facility. The human resource implications of31altering the current distribution of full-time, part-time and casual staffmay be investigated in depth and tested. In this sense, the effects ofcertain management decisions or simply expected changes in the compositionof the workforce may be examined. It is obvious that increasing thenumber of full-time staff in lieu of casual staff, for example, wouldresult in a decline in the number of persons needed to fill RN FIErequirements. The indices calculated by this method afford the additionaladvantage of estimating the exact magnitude of such changes.The sensitivity of the system to patterns of personnel deployment maybe examined in a similar fashion. By altering the average number of hoursworked (which may alternatively be expressed as the proportion of an FTEfilled) within the model, it should be possible to examine the effect ofmanagerial decisions concerning scheduling and staffing on nurserequirements.For example, the effects of such simple parameter changes areillustrated in Table 4. These data demonstrate the impact two differentparameter changes can have on the total number of nurses required to fill200,000 nursing hours over the period of one year. In the first scenario,the proportion of full-time staff is increased by 5 percent in relation topart-time and casual staff, whose proportions are accordingly decreased by2.5 percent each. The second scenario tests the possible effects of aimingto increase the average number of paid hours per casual nurse by 10percent. When these two strategies were combined (Scenario 3), reductionsin the number of individual nurses required throughout the year underTABLE 4EFFECT OF PARAMETER CHANGES ON NURSES REQUIRED PER 200,000 NURSING HOURSIN SELECTED HOSPITAL GROUPS IN BRITISH COLUMBIASCENARIO 1* SCENARIO 2** SCENARIO 3***ObservedHospital Group Total FIT pIT Casual Total FIT pIT Casual Total FIT pIT Casual TotalMetropolitan 164 90 26 42 156 84 30 46 158 90 26 38 152Non-Metropolitan 166 82 26 48 166 76 32 52 160 82 26 44 154GVRHD, >500 Beds 164 90 26 42 158 84 32 46 160 90 26 38 156GVRHD, <500 Beds 176 76 32 60 168 70 36 64 170 76 32 54 164GVRHD, Non-Teaching 170 86 30 48 164 80 34 50 164 86 30 44 160GVRHD, Teaching 160 96 24 36 154 90 28 40 156 96 24 32 150>40% Extended Care Beds 174 80 36 52 168 74 40 56 168 80 36 48 162Suburban 168 82 32 46 162 76 36 50 162 82 32 42 158Urban 158 110 10 34 152 102 14 38 156 110 10 30 150* Proportion of full-time nursing staff is increased 5%; part-time and casual proportions are decreased by 2.5% each.** Average paid hours per casual nurse increased 10%.*** Both of the above changes are incorporated into this scenario.33analysis ranged from 4.9 percent (eight fewer nurses to staff each 200,000hours in each of the large GVRHD hospitals) to over 7 percent (12 fewer inmetropolitan hospitals). All three scenarios imply that a heavy relianceon casual staffing as a stop-gap measure in response to local nursingshortages may not be the most appropriate solution, and may, in fact,increase exponentially the total requirement of nurses in any givenfacility.These tools, which have been designed to assess current, and testpotential, nurse resource deployment policies and approaches, and whichmay be used at the provincial, regional, facility, or even departmentallevels, also demonstrate that there is significant variation betweenfacility type and location in the efficiency with which existing nursingresources are utilized. Clearly, retention, or conversely, movement whichresults in wastage (of human and financial resources to recruit and orientnew nursing staff) plays a major role in estimating future total nurserequirements, when measured in terms of individual nurses rather thanhours or FTE's. Personnel management and staffing policies play animportant role in determining the direction of deployment trends. A roughindication of the direction of change desirable in management policies maybe gauged from conceptually mapping out an agency's current "position" oneach dimension of the management model discussed in this paper.Implications for Health ManagersPerhaps the most important aspect of the management model discussedhere is the assessment, by each facility, of the three factors which have34been demonstrated to be central to nurse requirements and the inter­relationships of the three factors unique to each facility . Referringagain to Figure 2, although it may be necessary to analyze the three axesseparately, that is only a starting point to the analysis of the three­dimensional cube, and more importantly, one's position within this cube.Clearly, depending on factors such as hospital type and role, bedcapacity, location, and as in the examples offered here, some facilitiesmay exhibit efficient deployment practices in one, two, or all three ofthe dimensions. Frequently, seemingly contradictory results may occur, inthat indicators of both "efficiency" and "inefficiency" may exist in thesame facility. For example, as shown in Table 1, the "urban" group employthe highest proportion of full-time staff of any of the peer groups (thusexhibiting an "efficient" deployment of staff), yet their full-time nursesprovide the lowest observed average number of hours. Furthermore, thedata in Table 3 show that this group exhibited the highest level ofpersonnel instability, and one that markedly surpassed the level of anyother of the peer grouped hospitals under analysis. Thus at the sametime, this group appeared to be the "most" effective in terms ofemployment status mix as well as the "least" effective.This observation leads to two important conclusion made earlier.First, it should be quite obvious that a show of seeming ineffectivenessalong any particular dimension, such as high turnover, need not mean thatthe facility, or group of facilities under question has dealt with nursedeployment in an inefficient manner. Furthermore, frequently any two35single dimensions may exhibit contrary trends, but do not necessarilycancel each other out. As a matter of fact, even low ratings on one ofthe dimensions, when teamed with more efficient practice on another maysimply be a condition of the characteristics unique to the facility orperhaps are determined by the behaviours exhibited on the otherdimensions. For example, the high turnover exhibited by the urban groupcould be attributed simple to the higher rates of transient populationstypically associated with urban neighbourhoods, or it may be linked insome manner to either relatively low average hours for full-time staff(that is, low average hours are due to high turnover), or even to therather high contingent of full-time staff (that is, the employer is lesslikely to use casual staff and is perhaps more demanding of full-timestaff).The temporal order of these interrelationships has not been exploredin any depth to date . However, it may be important to know whether, forexample, low average hours may simply be a manifestation of generaldiscontent and therefore also linked to higher turnover rates, orconversely. whether instability may be causally associated to low averagehours of nursing service provided by full-time staff.Secondly, it becomes very clear that it is essential to analyze theefficiency of staff deployment practices in a multi-dimensional manner.i.e. within the framework of all three axes of the management model .Turning again to the example of the urban hospital group, the analysis ofthe dimensions individually might lead one to quite different conclusions36about the overall efficiency of their deployment practices. Only byconsidering the intersection of these dimensions can a true pictureemerge. This is illustrated in Table 4, the so-called "bottom line". Inspite of higher turnover, and a lower average number of hours, or perhapsbecause of a relatively large full-time contingent, the lowest observednumber of individuals to fill 200,000 nursing hours was for this group.The analysis of turnover or average hours alone would not have necessarilypointed to this fact.Still, this does indicate that it is the size of the full-timecontingent, per se, that should be the dominant consideration for alltypes and sizes of facility. This was the case in urban hospitals, butthe specific proportion of full-time staff will most likely differ withthe unique needs of individual hospitals. Very definitely, what is"optimal" for one facility may be quite simply an unrealistic target foranother. Consider the case of a rehabilitation facility, which arrangesits staffing around a large full-time component and which rarely requireseither emergency (on call), or supplementary staff (due to short-termworkload fluctuations) compared to a general hospital removed from a majorcentre which relies on a large casual pool to fill need when it arises inthe absence of a large regular staff component, which is not needed underevery-day circumstances.A number of issues arise from this examination of nurse deploymentpatterns, one of which is that of recruitment versus retention. Althoughthe exact extent of the effect of retention for each facility type and37size is not clearly delineated, one can safely assume that retentionpolicies will generally have a positive impact on nurse requirements.Since recruitment is very taxing in terms of fiscal and human resources(constant recruiting may be a drain on the budget due to expensiverecruitment trips and other such strategies, and on both budget and staffdue to double staffing during orientation periods, person-hours requiredto complete recruitment drives, etc.) and indirectly, in terms of qualityof care (through frequent orientation periods), the more cost-effectivealternative to managing nursing resources must lie in the area of policiesdesigned to enhance retention.Traditional strategies which depend on in-migration and new graduatesto fill nursing vacancies may sometimes not be able to adequately providethe personnel to fill the nursing hours that are budgeted and required.This is demonstrated as provincial figures indicate that increasingly, thenumber of new registrants who had graduated from a nursing program inBritish Columbia is becoming almost equal to those graduating outside ofB.C. (1,246 to 1245 respectively, for a two-year period ending in June,1987). This is due to a decrease in the number of out-of-provinceregistrants and not to the appreciable increase of in-provinceregistrants. Thus, heavy demands are already being placed on the out-of­province pool of new registrants. Even if this were not the case, themigration of nurses has tended to be a "coat-tail" phenomenon, respondingto fluctuations in, for example, the forestry industry and the inter­provincial or international patterns in migration that are primarilycaused by such economic waves.38Regardless of which efficiency dimension becomes the principal targetof strategies designed to improve deployment of nurses, it cannot beevaluated or even identified without the proper monitoring of site­specific trends over a period of time. These results indicate that evenretroactive monitoring of these patterns are useful and may point todeficiencies or strengths within a given employment situation, and ifacted on quickly, can be used as a basis for the testing of futuremanagement strategies by substituting new parameters as required. Whilethe focus here has been on short-term monitoring only, the need for 10ng­term monitoring clearly exists. One must be cautioned not to rely onassumptions regarding a ten-year period based on one-year's data (i.e. ifretention was high in 1986 it cannot be assumed to continue at a high forthe following five year period). At least one study discussed earlier(Mann and Jefferson, 1988) demonstrated that this is an unwise inference.Of course, ongoing monitoring is the most valuable indicator ofeffectiveness of specific personnel policy shifts, and will alertmanagement regarding the need for change of strategy where necessary.In addition, the impact of staffing and scheduling on total nurserequirements cannot be underestimated. Again, while much attention, sofar, has been focused on reducing overall nursing hours (e.g. by bedclosures) to deal with shortages, an absence of any thoughtful discussionon maximizing existing resources through optimal manpower deploymentshould be apparent. A decrease in the number of individuals required, andindirectly, a more stable and dedicated nursing pool, can not only be39theorized about but can be concretely evaluated through new staffing andscheduling policies for nurses .Monitoring and evaluating current and past practices of nursedeployment in the hospital setting should be a vital and inseparable facetof staffing. The results discussed here show the value of taking anexplicitly planned approach when addressing not only immediate personnelconcerns. but also the more serious nurse shortage problem . Using actualdeployment statistics that apply to specific hospitals, groups ofhospitals or entire jurisdictions provides a more rational, and certainlymore tailor-made approach to the management of nursing resources. Thisapproach facilitates the design of strategies specific to each case and isconducive to early intervention. With baseline information which depictsthe local (or general) situation, the construction of various hypotheticalscenarios allows managers to test new staffing policies prior toimplementation, and to assess their case-specific appropriateness andeffectiveness. As no two hospitals, or facilities are exactly the same.it is not reasonable to expect that a single "management strategy" willsuccessfully address staff deployment problems in all facilities at alltimes .Research on nurse turnover (Weisman, 1982) indicates that theorganizational context (that is, each hospital culture) provides themeaning of otherwise apparently similar working conditions. For example,if shifts are less negotiable in one hospital than another. rotatingshifts will .mor e likely cause some dissatisfaction in the first facility40than in the second. The key to the solution of each problem situation isthe manager's ability to understand the reason(s) for the problem, and toachieve the appropriate mix of nursing resources: full-time as well ascasual, experienced as well as new graduate, specialist as well asgeneralist. Innovative management strategies would ensure that theexisting and prospective supply of nurses adequately meet nursing servicerequirements (Ginzberg et aI, 1982).REFERENCESAbelson, M.A. (1986), "Strategic Management of Turnover: A Model for theHealth Service Administrator", Health Care Management Review, 11(2):61-71.Aiken, L.H. and Mullinix, C.F. (1987), "The Nurse Shortage: Myth orReality", New England Journal of Medicine, 317(10):641-645.Becker, E.R., and Foster, R.W. (1988), "Organizational Determinants ofNursing Staffing Patterns", Nursing Economics, 6(2):71-75.B1anchf10wer, S. (1986), "Alternative Rota Systems", Nursing Times,82(10):55-58.Bosanquet, N. (1985), "Manpower Audit. Distribution Problems", NursingTimes, 81(38):21.Braddy, P. (1987), "Scheduling Alternatives for Administrators", NursingForum, 23(2):70-77 .Davis-Flood, S. and Diers, D. (1988), "Nurse Staffing, Patient Outcome andCost", Nursing Management, 19(5):34-43.Englefield, J. (1988), "Part-time Staff: A Blessing in Disguise?",Professional Nurse, 3(12):524-526.Friss, L. (1988), "The Nursing Shortage: Do We Dislike It Enough to CureIt?", Inquiry, 25:232-242.Gibson, L.W. and Dewhirst, H.D. (1986), "Using Career Paths to MaximizeNursing Resources", Health Care Management Review, 11(2):73-82.Ginzberg, E., Patray, J., Ostow, M. and Brann, E.A. (1982), "NurseDiscontent: The Search for Realistic Solutions", Journal of NursingAdministration, November.Halloran, E.J. and Kiley, Marylou (1987), "Nursing Dependency, Diagnosis­related Groups, and Length of Hospital Stay", Health Care FinancingReview, 8(3):27-36.Halloran, E.J . and Hadley-Vermeersch, P.E. (1987), "Variability in NurseStaffing Research", Journal of Nursing Administration, 17(2):26-34.Helmer and McKnight (1989), "Management Strategies to Minimize NursingTurnover", Health Care Management Review, 14(1):73-80.Howard, J.R. (1988), "TICK - A Staffing Tool for Floor Nurses", NursingManagement, 19(4):102-103.Kazanjian, A., Brothers, K. and Wong, G. (1986), "Modeling the Supply ofNurse Labor. Life-Cyclel Activity Patterns of Registered Nurses in OneCanadian Delivery System", Medical Care, 24(12):1067-1083.Kazanjian, A. and Stark, A.J . (1985), "RegLs t ered Nurses in B.C.: 1979­83", Health Management Forum, Spring, p.p. 35-44.Kramer, M. and Schmalenberg, C. (1988), "Magnet Hospitals: Part II.Institutions of Excellence", Journal of Nursing Administration, 18(2):11­19.Kramer, M. and Schmalenberg, C. (1988), "Magnet Hospitals: Part I.Institutions of Excellence", Journal of Nursing Administration, 18(1):13­24.Lindquist, K. and Hart, K. (1988), "How Hospitals are Responding to theShortage", American Journal of Nursing, 88(9):1206-1210.Longo, R.A., Uranker, M.M. (1987), "Why Nurses Stay: A positive approachto the nursing shortage", Nursing Management, 18(7)":78.Loveridge, C.E . (1988), "Contingency Theory: Explaining Staff NurseRetention", Journal of Nursing Administration, 18(6):22-25.Mann, E.E. and Jefferson, K.J. (1988), "Retaining Staff: Using TurnoverIndices and Surveys", Journal of Nursing Administration , 18(7,8):17-23 .Marquis, B. (1988), "Attrition: The Effectiveness of RetentionActivities", Journal of Nursing' Administration, 18(3) :25-29.Martin, J. (1988), "Operation SST. Supplemental Staffing Team", NursingManagement, 19(4) :72.McClure, M.L. (1984), "Managing the Professional Nurse. Part II.Applying Management Theory to the Challenges", Journal of NursingAdministration, March:11-17 .McClure, M.L. (1984), "Managing the Professional Nurse. Part I. TheOrganizational Theories", Journal of Nursing Administration, February:15­21.Nutt, P .C . (1984), "Decision-modeling Methods Used to Design DecisionSupport Systems for Staffing", Medical Care, 22(11):1002-1013.Ontario Ministry of Health, Advisory Committee on Nursing Manpower (1988),Report on Nursing Manpower, June.Pulcins, I., Kazanj ian, A. and Kerluke, K. (1988), "The Nurse ManpowerStudy. Volume II: A Synthesis of the Nurse Manpower Data in BritishColumbia", HMRU 88:1(2), University of British Columbia, February.Robertson, M. and Meehan, M.E. (1987), "Temporary Staffing: A PositiveApproach", Nursing Management, 18(7):80, 82.Rouvet, V. (1987), "Personnel Staffing and Scheduling", Soins, (502): 59­60.Scherer, P. (1987), "When Every Day is Saturday: The Shortage", AmericanJournal of Nursing, 87(10):1284-1290.Shaheen, P.P. (1985), "Staffing and Scheduling: Reconcile Practical MeansWith the Real Goa1", Nursing Management, 16(10):64-72.Stenske, J.E., et a1 (1988), "Resource Teams: Their Structure and Use",Journal of Nursing Administration, 18(4):34-38.Wall, L.L. (1988), "P1an Development for a Nurse Recruitment RetentionProgram", Journal of Nursing Administration, 18(2):20-27.Weisman, C.S. (1982), "Recruit from Within: Hospital Nurse Retention inthe 19805", Journal of Nursing Administration, May.Wolf, S.M. (1987), "We Give Every Nurse the Schedule She Wants II , RN,50(10):23-26.


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items