35-MSJ February 2008.ps2/11/084:03 PMPage 35Gina Aya NelsonMajor L. KingStephanie BrodineNurse-Physician CollaborationOn Medical-Surgical UnitsInterdisciplinary collaboration is viewed as a criticalfactor in delivering qualitypatient care. The purpose ofthis study was to describenurse-physician perceptionsof collaboration relationshipon general medical-surgicalunits.Although many factors contribute to the current nursing shortage, one ofthe primary reasons nurses leave the profession is dissatisfaction withtheir practice environment (Joint Commission on Accreditation ofHealthcare Organizations, 2001; U.S. General Accounting Office, 2001).Integrated structure and processes (collaboration) that allow nurses andphysicians to resolve their differences (conflict) are likely to increase nursesatisfaction, recruitment, and retention (Rosenstein, 2002; Schmalenberg etal., 2005). Achieving this goal requires significant commitment and supportby nursing administrators.Conflict with physicians has been identified as one stressor in the nursework environment (Greenfield, 1999). Nurses may face both verbal andphysical abuse when conflict arises with physicians (Rosenstein, 2002).Conflict may arise from differences in opinion regarding patient carerequirements, or from the longstanding hierarchical relationship of physician dominance and nursing subservience (Greenfield, 1999). Nevertheless,some degree of conflict exists. The goal should be to produce positiveresults through conflict resolution.Literature ReviewGina Aya Nelson, MSN, RN, is InternManager, Nursing Support, ResourcesPool, Scripps Hospital, La Jolla, CA.Major L. King, PhD, RN, CNS, is aProfessor, School of Nursing, AzusaPacific University, Azusa, CA.Stephanie Brodine, MD, is DivisionHead, Epidemiology and Biostatistics,Graduate School of Public Health, SanDiego State University, San Diego, CA.Note: A related article on this topic“Bridging the Professional Chasm: Toolsfor Collaborative Communication” canbe found in the January/February 2008issue of Medsurg Matters, the officialnewsletter of the Academy of MedicalSurgical Nurses ( Nursing—February 2008—Vol. 17/No. 1Interdisciplinary collaboration is viewed as a critical factor in deliveringquality patient care (Institute of Medicine [IOM], 2004; Rosenstein &O’Daniel, 2005). The benefits of positive nurse-physician relationships arewell documented in the literature. Interdisciplinary collaboration benefitsthe organization in terms of decreased costs, better patient care, and economy of decision making (National Joint Practice Commission, 1981;Schmalenberg et al., 2005), and decrease patient morbidity and mortality(Aiken, Clark, Sloane, Sochalski, & Huber, 2002). Moreover, a strong linkexists between effective interactions (collaboration) among team membersand risk-adjusted patient length of stay (Aiken, 2001). The nursing profession faces a critical nursing shortage with the nursing workforce expectedto decrease by 20% in 2020 (Buerhaus et al., 2005), making collaborationbetween nurses and physicians a high priority for nurse administrators.Studies show that nurses who practice collaboratively with physiciansexperienced less “burnout,” and recruitment and retention rates are higherthan nurses who practice in non-collaborative environments (Aiken, Clarke,Sloane, Sochalski, & Weber, 1999; Buchan, 1999).One focus of Nursing’s Agenda for the Future was the development of anurse practice environment that supported shared decision making and collaboration (American Nurses Association, 2002). The IOM (2004) also recommended that organizations adapt a structure that supports collaborationby encouraging interdisciplinary patient care rounds and providing ongoingeducation in the collaborative process and training for all staff.Many studies have explored nurse-physician collaboration, but mosthave been limited to intensive care units and emergency departments(Rosenstein & O’Daniel, 2005; Schmalenberg et al., 2005). Several studies35

36-MSJ February 2008.ps2/11/084:03 PM(King & Lee, 1994; Rosenstein &O’Daniel, 2005; Schmalenberg et al.,2005) showed significant differencesin nurses’ and physicians’ perceptions of collaborative relationship,with physicians perceiving greatercollaborative behaviors than nurses. Alternately, nurses in criticalcare units are likely to have betterworking relationships with physicians when compared to nurses onmedical-surgical units because oftheir clinical expertise and theopportunity to work closely withphysicians (Greenfield, 1999; King &Lee, 1994).Definition of CollaborativePracticeCollaborative practice hasbeen defined in many ways. Weissand Davis (1985) offered a usefuldefinition that supports the theoretical framework for this study.The researchers defined collaborative practice as “the interactionsbetween nurse and physician thatenable the knowledge and skills ofboth professions to synergisticallyinfluence the patient care provided” (p. 299).Study purpose. Efforts to increase collaboration between nurses and physicians are notablebecause of their impact on patientoutcomes such as decreased morbidity and mortality (Kramer &Schmalenberg, 2003; Rosenstein &O’Daniel, 2005). It is important toevaluate nurses’ and physicians’perception of their interaction patterns in medical-surgical units,where physicians spend little timeinteracting with nurses (Greenfield,1999). The purpose of this studywas to describe medical-surgicalnurses’ and physicians’ perceptions of their collaborative behaviors on medical-surgical units.Theoretical FrameworkStyles of conflict managementhave evolved from several dimensions. Deutsch (1949) and Rubleand Thomas (1976) conceptualizedconflict management in one andtwo dimensions respectively.However, the concept of conflictresolution has evolved to fivedimensions in other sources (Blake& Mouton, 1970; Kilmann &Thomas, 1977; Rahim, 1983).Kilmann and Thomas (1977)36Page 36provided the theoretical frameworkfor this study. They suggested thatindividuals use one of five modes inconflict resolutions: competing, collaborating, compromising, avoiding, and accommodating. These fivemodes reflect independent dimensions of interpersonal conflictbehavior and build on earlier workby Ruble and Thomas (1976), whoproposed a classification schemebased on a two-dimensional modelof cooperativeness and assertiveness. The dimensions of cooperation and assertiveness have independent connotations for individuals. Cooperation is attempting tosatisfy the other person’s concerns,while assertiveness is attempting tosatisfy one’s own concerns. Thecombination of these two dimensions yields five types of conflictbehaviors that identify conflict ascompeting (assertive and uncooperative), collaborating (assertiveand cooperative), avoiding (unassertive and uncooperative),accommodating (unassertive andcooperative), and compromising(intermediate in both cooperativeness and assertiveness).Most nurse-physician conflictsoccur in the areas of general plan ofcare (Rosenstein, 2002), specificorders, and patient disposition.Conflict is natural and part of interactions with others (Zerwekh &Claborn, 2006); therefore, the goalshould focus on conflict resolution.Noted industrialist Mary ParkerFollett (1977) was one of the first tostudy organization conflict, suggesting that conflict be viewed asdifferences of opinions and differences of interest. Follet noted thatconflict is neither “good” nor “bad,”and that it should be used to identify the source of differences. Dealingwith conflict involves domination,compromise, or integration. Indomination, one party wins overthe other. In compromise, eachparty gives up a little to keep thepeace. Compromise is ineffectivebecause it means giving up a desire.According to Follett (1977),integration is an effective mode ofconflict resolution; “when twodesires are integrated, it means thata solution has been found in whichboth desires have a place that neither side has had to sacrifice anything” (p. 245). Integration involvesinventing new ways of solving differences. The key to this process isopen and effective communication.Differences cannot be resolved unless nurses and physicians acknowledge that differences exist intheir practice. These differencesmay be due to culture, gender, andto some extent, perceived differences in power base (for example,clinical expertise) (Kramer &Schmalenberg, 2003). Integration isan effective mode of conflict resolution because it creates win-win situations. In integration, solutionshave been found in which bothdesires have a place, and neitherside needs to sacrifice anything(Follett, 1977).Obstacles to integration include undue influence of leadersand lack of training for using it(Follett, 1977). For example, Greenfield (1999) noted that nursing leaders (managers) may not be receptive to establishing collaborativerelationships with physicians because this may be seen as reinforcing the subservient role. He recommended that nurses and physiciansreceive training that recognizes theunique contributions of each in providing quality patient care. Leadership support and education are keyto improving relationships betweenindividuals with different worldviews (Follett, 1977). A recent studyby Rosenstein and O’Daniel (2005)supported this strategy. Kilmannand Thomas (1977), Rubel andThomas (1976), and Follett (1977)suggested that it is the manner inwhich nurses and physicians resolve their differences that determines whether collaborative practice exists in their relationships.Cooperation, assertiveness, andopen communication are healthyconflict resolution strategies inresolving conflict and are necessaryfor collaboration to occur in theorganization (King & Lee, 1994).MethodologyDesign and sample. A convenience sample of registered nursesand physicians was recruited frommedical-surgical units in a hospitalin San Diego, CA. A total of 120 surveys were distributed to both nurses and physicians respectively.Power analysis was used to determine the sample size with an alphaMEDSURG Nursing—February 2008—Vol. 17/No. 1

37-MSJ February 2008.ps2/11/084:03 PMof 0.05, power of 0.80, and a moderate effect size of 0.40. These parameters required a sample size of 196(98 nurses and 98 physicians).However, the available populationfor this study was less than thatrequired to meet this parameter;however, the effect size was largeenough to detect differences inmeans scores between nurses andphysicians in this study (Polit &Beck, 2004). The response rate was84% (101 surveys) for nurses, and43% for physicians (51 surveys).Eight questionnaires were discarded because they failed to meet thecut-off guidelines of answering atleast 50% of the questions. Dataanalysis was performed usingresponses from 95 nurses and 49physicians. It was large enough todetect a difference in mean scoresbetween groups with a partial η2(effect size) of 0.11, which according to Cohen’s taxonomy is a medium to large effect size.Page 37Figure 1.Collaborative Practice Scale – NursesData Collection ProceduresInstrumentation. The Collaborative Practice Scale (CPS) was usedto measure perceptions of collaboration between nurses and physicians (Weiss & Davis, 1985). TheCPS has two scales, one for nurses(see Figure 1) and one for physicians (see Figure 2). The CPS fornurses has 9 items with a possiblescore of 54. Each item is scored on6-point scale, ranging from never (1)to always (6).The nurse CPS has two factors,with one factor having a maximumscore of 30 and the other 24. Thefirst factor (five items) measuresthe degree to which a nurse directly asserts professional expertiseand opinions when interactingwith physicians about patient care.The second factor (four items)measures the degree to which anurse clarifies with the physicianmutual expectations regarding thenature of shared responsibilities inpatient care.The CPS for physicians has 10items that are divided into two factors of five items each. Each itemis scored on the same 6-pointscale. Each of the two factors hasa maximum score of 30 (total maximum score 60). The first factor(five items) measures the degreeto which a physician acknowl-MEDSURG Nursing—February 2008—Vol. 17/No. 1Figure 2.Collaborative Practice Scale – Physicians37

38-MSJ February 2008.ps2/11/084:03 PMedges the importance of nurses’unique contribution to differentresponsibilities in patient care.The second factor (five items)measures the degree to which aphysician seeks consensus withnurses regarding mutual responsibilities and patient care goals.Higher scores on the instrumentimply greater use of collaborativebehaviors by the nurse or physician (Weiss & Davis, 1985).This instrument is reliable andvalid. Cronbach’s alpha coefficientsof 0.80 and 0.84 were reported fornurses and physicians respectively(Weiss & Davis, 1985). Anotherstudy reported Cronbach’s alphacoefficients of 0.83 for the nurseCPS total scales and 0.86 for thephysician CPS total scales (King &Lee, 1994). In the current study, theCPS demonstrated acceptable internal consistency reliability withCronbach’s alpha coefficients of0.87 for the nurse CPS and 0.88 forthe physician CPS. Factor analysiswas used to determine constructvalidity for the nurse and physicianCPS, which was similar to thatreported by Weiss and Davis (1985)and King and Lee (1994).This study was approved bythe institutional review boards(IRB) at San Diego State Universityand a southern California hospital,and IRB protocols were followed inthis study. A cover letter explainingthe purpose of the study, a consentform, the CPS, and a demographicrecord (see Table 1) were includedin the packet that was distributedto the sample via unit mail boxes.No identifying markers were placedon the surveys. The directions oneach survey noted whether it was anurse or physician survey.Page 38Table 1.Demographics for Nurses and PhysiciansMean SDN (%)Nurse36 1222-63Physician50 maleMale7 (13.7)*44 sicianCaucasian41 (80.4)*Others10 (19.6)Years ExperienceNurse0-10 years66 (69.5)11-20 years10 (10.5) 20 years19(20)7(7.8)Physician2-10 years11-20 years17 (33.3) 20 years27 (58.9)Shift Worked (Nurses)0700-193053 (55.8)1930-073034 (35.8)Others8(8.4)6(6.4)Degree (Nurses)DiplomaData AnalysisADN40 (42.6)Data were analyzed usingStatistical Package for the SocialSciences (SPSS, Version 14) applications software (2005). Univariateanalysis was used to test the differences in mean scores betweennurses’ CPS and the adjusted meanscores on the physicians’ CPS.Univariate analysis also was used totest group differences for selecteddemographics (age, gender, education, experience, and certifications).BSN46 )No2(4)CertificationNursesPhysicians* Missing dataMEDSURG Nursing—February 2008—Vol. 17/No. 1

39-MSJ February 2008.ps2/11/084:03 PMResultsSample demographics areshown in Table 1. Univariate analysis showed that physicians hadhigher total mean scores on theCPS than nurses (F [(1, 142] 18.16,p 0.05, partial η2 0.113). The partial η2 suggested that 11% of thevariability was due to group membership. Item mean scores for nurses and physicians are shown inTables 2 and 3. Nurses’ and physicians’ mean scores ( SD) on theCPS were 3.5 1.04 and 4.3 1.06respectively. Nurses and physicianswith more education [F (4, 84) 3.59, p 0.010) and experience ([F (5, 78] 5.25, p 0.0001) were likelyto perceive collaborative relationships, as were nurses with a jobtitle ([F (1, 42] 4.33, p 0.044), andadvanced certification (F [1, 82] 17.4, p 0.0001). King and Lee (1994)demonstrated that those withadvanced clinical expertise are likely to perceive collaborative relations exist.Page 39Table 2.Item Means for the Nurse CPS(N 95)Mean ( SD)Item 12.4 (1.5)Item 22.8 (1.4)Item 33.7 (1.7)Item 43.3 (1.6)Item 54.6 (1.1)Item 63.2 (1.4)Item 73.8 (1.4)Item 84.6 (1.2)Item 93.3 (1.5)Note: Scale range from 1 to 6.Table 3.Item Means for the Physician CPS(N 49)Mean ( SD)Item 14.9 (1.14)Item 24.7 (1.11)DiscussionItem 33.2 (1.36)This study showed significantdifferences in perceptions of collaborative behaviors between nursesand physicians on general-medicalsurgical units. Findings were consistent with other studies that evaluated the concept of collaboration(King & Lee, 1994; Schmalenberg etal., 2005). Decades of researchshow that increased collaborativepractice between nurses and physicians results in better patient outcomes. However, this mode of conflict resolution remains elusivebetween the two professions (King& Lee, 1994). In this study, communication between nurses and physicians was distant as suggested bythe differences in mean scores onthe CPS between the two groups.Nurses’ mean scores on theCPS suggested they lacked assertiveness skills in communicatingtheir unique contribution to patientcare requirements when interactingwith physicians. Findings are consistent with published reports byTimmins and McCabe (2005a,2005b). Gender, education preparation, and the nursing culture mayplay an important part in nurses’lack of assertive behaviors whencommunicating with physicians.Nurses had mean scores less than 3Item 44.9 (.26)Item 54.5 (1.26)Item 63.9 (1.48)Item 74.7 (1.03)Item 83.5 (1.39)Item 95.1 (1.20)Item 103.7 (1.45)MEDSURG Nursing—February 2008—Vol. 17/No. 1Note: Scale range from 1 to 6.and physician) recognizes and values the contributions of the other;thus power is equal in their relationships. Nurses share some of theresponsibility for the manner inwhich they present themselveswhen interacting with physiciansbecause their approach may contribute to the perceived powerimbalance. Greenfield (1999) notedthat nurses used more “support/agreement” messages when interacting with physicians, while physicians used more “give opinion”messages when interacting withnurses. These findings suggest thatassertiveness training may be beneficial for nurses; research indicatesthat this training improvesassertiveness skills and self-esteem(Lin et al., 2004).Physicians’ higher mean scoressuggest that they value and use theinput from nurses, and are comfortable with the role of the physiciannurse team with respect to patientcare delivery. However, this perception was not shared by nurses inthis study. Of the items on thephysicians’ scale, item #3 (“I discuss with nurses the similaritiesand differences in medical andnursing approaches to care”) hadthe lowest mean and SD (3.2 1.36.)These findings may reflect ineffective communication patternsbetween the two professionals, asdemonstrated by the study’s theoretical framework.Study Limitation(on a scale of 1-6) on item #1 (“I askphysicians about their expectationsregarding the degree of my involvement in health care decisions”) anditem #2 (“I negotiate with the physician to establish our responsibilities for discussion of different kindsof information with patients”).Until these issues are resolved,nurses may not enjoy collegial, collaborative relationships with physicians in decision making. Moreover, nurses may be reluctant toshare their expertise and opinionsregarding patient care requirements because of perceived powerdifferences. Kramer and Schmalenberg (2003) noted that in collaborative relations, power is mutualbut not equal. In true collaborativerelations, each professional (nurseBecause the small sample sizerepresented one hospital in SanDiego, results cannot be generalized. However, the partial η2 (effectsize) suggested it was large enoughto detect differences in meansscores between the two groups.Because this appears to be the firststudy to describe nurses’ and physicians’ perceptions of collaborative behaviors in the medical-surgical environment, further studiesusing a larger sample size are needed to determine if findings represent the true population means.ConclusionFindings indicated that collaborative practice between nursesand physicians occurred at low-tomoderate levels on medical-surgicalunits. Findings suggest that the rela-39

40-MSJ February 2008.ps2/11/084:04 PMtionship between nurses and physicians has not changed over time.Maximizing nurse-physician collaboration holds promise for improvingquality patient care and creating satisfying work environments for nurses and physicians (Rosenstein &O’Daniel, 2005). Collaborative practice is a process that involves thevalued contributions of all teammembers in reaching the best possible solutions. Additionally, collaboration between nurses and physicians requires intentional teambuilding at schools and work settings (Hojat et al., 2001). It is important for nurses and physicians todevelop a new culture of collaboration which merges the uniquestrengths of each discipline with themutual goal of quality patient care. ReferencesAiken, L.H. (2001). Evidence-based management: Key to hospital workforce stability.Journal of Health AdministrationEducation, Spec. No., 117-124.Aiken, L.H., Clarke, S.P., Sloane, D.M.,Sochalski, J., & Huber, J.H. (2002).Hospital nurse staffing and patient mortality, nurse burnout, and job dissatisfaction. Journal of the American MedicalAssociation, 288(16), 1987-1993.Aiken, L.H., Clarke, S.P., Sloane, D.M.,Sochalski, J., & Weber, A.L. (1999).Organization and outcomes of an inpatient AIDS ward. Medical Care, 37(8),760-767.American Nurses Association (ANA). (2002).Nursing’s agenda for the future.Retrieved July 1, 2005, from, R., & Mouton, J. (1970). The fifthachievement. Journal of BehavioralScience, 6, 413-426.Buerhaus, P., Donelan, K., Ulrich, B.T.,Norman, L., Williams, M., & Dittus, R.(2005). Hospital RNs’ and CNOs’ perception of the impact of the nursing shortageon the quality of care. NursingEconomics , 23(5), 214-221.Buchan, J. (1999). Still attractive after all theseyears? Magnet hospitals in changinghealth care environment. Journal ofAdvanced Nursing, 30(1), 100-108.Deutsch, M. (1949). A theory of co-operationand competition. Human Relations, 2,129-152.Follett, M.P. (1977). Constrictive conflict. InC.H. Summer, J.J. O’Connell, & N.S.Peery, Jr. (Eds.), The managerial mind(4th ed., pp. 245-248). Homewood, IL:Richard D. Irwin.Greenfield, L.J. (1999). Doctor and nurses: Atroubled partnership. Annals of Surgery,230(3), 279-288.Hojat, M., Nasca, T., Cohen, M., Fields, S.,Rattner, S., Griffiths, M., et al. (2001).Attitudes toward physician-nurse collaboration: A cross-cultural study of maleand female physicians and nurses in theUnited States and Mexico. NursingResearch, 50(2), 123-128.40Page 40Institute of Medicine (IOM). (2004). Keepingpatients safe: Transforming the workenvironment of nurses. Retrieved April15, 2005, from Commission on Accreditation of Healthcare Organizations. (2001, May 17).Statement by the Joint Commission onAccreditation of Healthcare Organizations hearing on addressing direct carestaffing shortages before the SenateCommittee on Health, Education, Laborand Pension. Washington, DC: Author.Kilmann, R., & Thomas, K.W. (1977). Developing a forced-choice measure of conflict-handling behavior: The “MODE”instrument. Educational and Psychological Measurement, 37, 309-325.King, M.L., & Lee, J.L. (1994). Perceptions ofcollaborative practice between navynurses and physicians in the ICU setting.American Journal of Critical Care, 3(5),331-336.Kramer, M., & Schmalenberg, C. (2003).Securing “good” nurse physician relationship. Nursing Management, 34(7), 34-38.Lin, Y.R., Shiah, I.S., Chang, Y.C., Lai, T.J.,Wang, K.Y., & Chou, K.R. (2004).Evaluation of an assertiveness trainingprogram on nursing and medical students’ assertiveness, self-esteem, andinterpersonal communication satisfaction. Nurse Education Today, 24(8), 656665.National Joint Practice Commission. (1981).Guidelines for establishing joint or collaborative practice in hospitals. Chicago:Author.Polit, D., & Beck, C.T. (2004). Nursing research: Principles and methods (7th ed.).Philadelphia: Lippincott, Williams &Wilkins.Rahim, M. (1983). A measure of styles of handling interpersonal conflict. Academy ofManagement Journal, 26, 368-376.Rosenstein, A.H. (2002). Nurse-physician relationships: Impact on nurse satisfactionand retention. American Journal ofNursing, 102(6), 26-34.Rosenstein, A.H., & O’Daniel, M. (2005).Disruptive and clinical perceptions ofbehaviors outcomes: Nurses and physicians. American Journal of Nursing,105(1), 54-64.Ruble, T.L., & Thomas, K.W. (1976). Supportfor a two-dimensional model of conflictbehavior. Organizational Behavior andHuman Performance, 16(1), 143-155.Schmalenberg, C., Kramer, M., King, C,Krugman, M., Lund, C., Poduska, D., etal. (2005). Excellence through evidence:Securing collegial/collaboration nursephysician relationships, Part 1. Journal ofNursing Administration, 35(10), 450-458.Timmins, F., & McCabe, C. (2005a). Howassertive are nurses in the workplace? Apreliminary pilot study. Journal of NursingManagement, 13, 61-67.Timmins, F., & McCabe, C. (2005b). Nurses’and midwives’ assertive behavior in theworkplace. Journal of Advanced Nursing,51(1), 38-45.U.S. General Accounting Office. (2001).Nursing workforce: Emerging nurseshortage due to multiple factors (GAO10-944). Washington, DC: Author.Weiss, S.J., & Davis, H.P. (1985). Validity andreliability of the collaborative practicescale. Journal of Nursing Research,34(5), 299-305.Zerwekh, J., & Claborn, J.C. (2006). Nursingtoday: Transition and trends (5th ed., pp.271-288). St. Louis: Elsevier.MEDSURG Nursing—February 2008—Vol. 17/No. 1

MEDSURG Nursing—February 2008—Vol. 17/No. 1 35 Gina Aya Nelson, MSN, RN,is Intern Manager, Nursing Support, Resources . Professor, School of Nursing, Azusa Pacific University, Azusa, CA. Stephanie Brodine, MD,is Division Head, Epidemiology and Biostatistics, Graduate School of Public Health, San Diego State University, San Diego, CA. Note .