Abstract
Enhancing nurse involvement in decision-making is a starting point in addressing the nursing shortage, recruitment, and retention. The purpose of this descriptive comparative secondary data analysis was to describe the level of registered nurses’ actual and preferred decisional involvement (DI) in two studies carried out during 2004 and 2010 and to describe the difference in the levels of actual and preferred DI between the 2004 study (N = 290) and the 2010 (N = 111) study in a Midwestern medical center after a new shared governance structure was implemented. In the 2004 and 2010 studies, there were statistically significant differences between actual and preferred levels of DI. A statistically significant decrease in means occurred in actual and preferred DI in 2010 as compared with that in 2004. A concerted effort must be made by nursing leaders to enhance DI, offer adequate resources, promote learning and growth, and recognize nursing contributions within a shared governance.
By the year 2025, it is projected that there will be at least 260,000 fewer registered nurses (RNs) available to provide care than will be needed (Buerhaus, Auerbach, & Staiger, 2009). The shortage is expected to intensify as Baby Boomers age and the need for health care services grow (American Association of Colleges of Nursing [AACN], 2012). Factors contributing to nursing vacancies and turnover include unsupportive practice environments, long work hours, an aging workforce, and excessive physical and psychological demands (AACN, 2012). A management style such as shared governance that emphasizes the decisional involvement (DI) of nurses has been associated with enhancement of the nursing practice environment (Kowalik & Yoder, 2010; Moore & Hutchison, 2007; Weston, 2008) and improved health care organizational outcomes (Kowalik & Yoder, 2010; Kramer et al., 2009; Moore & Hutchison, 2007; Weston, 2008). Enhancing nurse involvement in the decision-making process contributes to nurse retention and recruitment (Havens & Vasey, 2003), empowers nurses to take control of their practice, and may act as a starting point in addressing the nursing shortage caused from vacancies and turnover.
DI, shared governance, control over nursing practice, and empowerment are noted to be interchangeable within the literature. All of these concepts play an integral part in nursing organizations by involving nurses in patient care and organizational decisions, while addressing accountability and responsibility for nursing practice (Jones, Stasiowski, Simons, Boyd, & Lucas, 1993). The purpose of this literature review is to discuss the empirical literature related to the concept of DI as a component of shared governance, control over nursing practice, empowerment, and their relationship to adequate resources, education, and personal growth.
Shared governance is a management style that facilitates the DI of nurses in decisions that affect their practice, empowers nurses to contribute, and participate in their work environments (Kowalik & Yoder, 2010; Laschinger, 2008). Defined by Kowalik and Yoder (2010), DI is “a complex collaboration between the nursing personnel of an organization and the organization’s leadership. Decisional involvement requires nurses to choose to participate in a process that involves organizational structures” (p. 259). Shared governance affords nurses the opportunity of having a voice, staying informed, being heard, and included in the decision-making process if one so chooses; this sets the stage for DI (Kowalik & Yoder, 2010).
DI of nurses in a shared governance structure allows nurses to influence the decisions that affect their practice, work environment, professional development, and personal fulfillment (Kowalik & Yoder, 2010). Having a voice, greatly improves nurse job satisfaction (Angermeier, Dunford, Boss, Smith, & Boss, 2009; Bogue, Joseph, & Sieloff, 2009; Kowalik & Yoder, 2010; Laschinger, 2008; Macphee, Wardrop, & Campbell, 2010), and nursing morale (Kowalik & Yoder, 2010) while contributing to lower levels of burnout (Angermeier et al., 2009). The shared decision-making approach of shared governance also promotes greater employee engagement (Macphee et al., 2010); an engaged nurse is more involved and has increased knowledge about what is happening in their work environment (Williamson, 2005).
Shared governance models are often found to enhance nurse recruitment and retention in an organization (Kowalik & Yoder, 2010), and foster greater nurse commitment (Kowalik & Yoder, 2010; Moore & Hutchison, 2007). An environment that creates a proactive culture with committed employees and a higher level of work satisfaction is predisposed to generating a higher quality of patient care (Laschinger, 2008) and better patient outcomes (Kowalik & Yoder, 2010; Moore & Hutchison, 2007).
In contrast to the positive outcomes within practice environments, some researchers found no differences in group cohesion and job stress (Jones et al., 1993), in nurses having greater feelings of making a difference throughout departments or feelings of making a difference to patients or to the hospital (Ireson & McGillis, 1998), in the areas of autonomy, role ambiguity, role conflict, self-perceived effectiveness, job satisfaction, and nurse commitment to the organization and anticipated turnover (Kennerly, 1996). Moore and Wells (2010) reported nursing council participation had no significant effect on nurse perceptions of structural empowerment and organizational commitment.
The implementation, dissemination, and enculturation of a shared governance model can prove to be challenging (Ballard, 2010; Hess, 2004; Moore & Hutchison, 2007). Implementation requires a shift away from traditional hierarchical, centrally controlled management styles. This shift encompasses the reconfiguration of not only the organizational structure, but the relationships among nurses and management, their interactions, and their decisions (Porter-O’Grady, 2001). The leadership team must be willing and able to participate in the shifting of their roles and power. Resistance is often met by those unwilling to change, and the conceptual ambiguity of shared governance is often misunderstood (Hess, 2004).
Research shows that empowerment is positively correlated with a nurse’s improved job satisfaction, (Laschinger, 2008; Laschinger & Finegan, 2005; Laschinger, Leiter, Day, & Gilin, 2009; Leggat, Bartram, Casimir, & Stanton, 2010; Ning, Zhong, Lido, & Qiujie, 2009), increased work engagement (Kuokkanen, Leino-Kilpi, & Katajisto, 2003), decreased levels of burnout (Laschinger et al., 2009), and incivility among nurses (Laschinger et al., 2009; Smith, Andrusyszyn, & Laschinger, 2010). For health care organizations, empowerment of nurses translates into improved organizational outcomes.
In contrast to the outcomes correlating higher levels of empowerment to decreased intent to leave, Fitzpatrick, Campo, Graham, and Lavandero (2010) found that nurses with graduate degrees had higher empowerment scores, but also higher intent to leave. It was also found that African Americans (Fitzpatrick et al., 2010), younger nurses (Fitzpatrick et al., 2010; Kuokkanen et al., 2003), and men were more likely to leave their positions to pursue career advancement despite their perceived empowerment (Fitzpatrick et al., 2010).
Empowerment of nurses has also been shown to influence patient outcomes. A significant positive correlation was found between nurses’ perceptions of empowerment and patient satisfaction (Donahue, Piazza, Griffin, Dykes, & Fitzpatrick, 2008). It has also been linked to improved quality of care (Laschinger, 2008; Leggat et al., 2010) and increased perceptions of the safety climate (K. J. Armstrong & Laschinger, 2006; K. Armstrong, Laschinger, & Wong, 2009).
Control over nursing practice is a concept that describes how nurses influence decisions about their practice, and is operationalized through a shared governance structure (Kramer et al., 2009) ensuring that nurses have the responsibility and power to make decisions about rules, policies, practices, and structures (Weston, 2008). According to Kowalik and Yoder (2010), control over practice is autonomy experienced by nurses at the organizational level. Without support at the organizational level, control over nursing practice cannot be practiced at the unit level (Kramer et al., 2009).
Nurse’s DI, shared governance, empowerment, and control over nursing practice, all similar concepts and often used interchangeably, have been shown throughout the literature to positively enhance the nursing practice environment and lead to positive organizational outcomes. These improved outcomes have the potential to enhance the effort of decreasing the nursing shortage.
Purpose
The purpose of this descriptive comparative secondary data analysis (data obtained from studies from 2004 and 2010) was to describe the level of RN’s actual and preferred DI in the 2010 study and to describe the difference in the levels of actual and preferred DI between the 2004 study and the 2010 study in a rural Midwestern medical center after a new shared governance structure was implemented. The specific research questions explored in this study included the following:
Research Question 1: What is the difference in the level of RN’s actual and preferred DI in the 2010 study?
Research Question 2: What is the difference in the level of RN’s actual DI between the 2004 study and the 2010 study?
Research Question 3: What is the difference in the level of RN’s preferred DI between the 2004 study and the 2010 study?
Method
Design
A descriptive comparative secondary data analysis was used for this study. Data were collected for the 2004 study with a basic shared governance structure in place. In January 2009, a new shared governance structure was established at this rural Midwestern medical center with follow-up data collected for the 2010 study in August/September 2010. This secondary analysis received approval from the Winona State University’s Institutional Review Board.
Sample/Setting
The 2004 study consisted of a convenience sample of 337 RNs within a rural Midwestern medical center. The 2010 study consisted of a convenience sample of 140 RNs in the same Midwestern medical center. Nurses were required to be staff/charge/lead RNs (2004, n = 290; 2010, n = 111) with clinical responsibilities and to have completed the Decisional Involvement Scale (DIS) to be included in this secondary analysis sample.
The organization had a basic shared governance structure in 2004 which consisted of three councils: the Education Council, the Nurse Practice Council, and the Coordinating Council. The Education Council and the Nurse Practice Council reported to the Coordinating Council for work activity and if there were concerns that needed broader consideration or direction. The composition of the Education Council and Nurse Practice Council was staff nurses from each of the inpatient nursing units. The Coordinating Council included the leaders of the Education Council, nurse leaders, the chief nursing officer (CNO), and representatives from other areas such as outpatient and home care settings. The Coordinating Council was led by the Director of Nursing Practice.
In 2010, the shared governance structure was changed so that more nurses could become involved in making decisions about their practice. The structure was changed to unit/practice location councils called Affinity Councils. The unit-based Affinity Councils comprised staff nurses from unit/practice locations. The practice locations represent the entire spectrum of care for the organization, all inpatient units, outpatient, clinics, home care, and hospice.
Staff nurses from the Affinity Councils serve on the division wide Operational Councils. These councils include: Education, Practice Standards, Resource, Informatics, Quality & Patient Safety, and Patient Care Experience. The staff leaders of each of the Operational Councils comprise the Coordinating Council that is cochaired by an elected staff member and the CNO.
Data Collection
Data for the descriptive comparative secondary analysis were collected in 2004 and 2010. The 2004 study data were collected through mailed questionnaires sent to RNs at a rural Midwestern medical center. Due to a less-than-desirable return rate, a second mailing was also completed following modification of the survey and the results of the two surveys were combined (Scherb et al., 2011). Data from the 2010 study were collected in August and September after computer training sessions or via an email request in the form of an online survey at the same rural Midwestern medical center. The data pertinent to the secondary analysis were obtained from the principal investigators and exported to SPSS® for analysis and were kept on a password-protected computer.
Instrument
The instrument used in this secondary analysis was the DIS (Havens & Vasey, 2003, 2005). The DIS comprised 21 items that measure the actual and preferred DI of nurses. The DIS also measures nurse involvement in decisions associated with six subscales: unit staffing; quality of professional practice; professional recruitment; unit governance, and leadership; quality of support staff practice; and collaboration/liaison activities (Havens & Vasey, 2003, 2005). The DIS uses a 5-point scale to indicate the degree to which the participant perceives decisions are the responsibility of staff nurses and administration/management on the nursing unit (Havens & Vasey, 2003, 2005). Responses to questions regarding actual and preferred levels of DI are as follows: 1 = administration/management only; 2 = primarily administration/management—some staff nurse input; 3 = equally shared by administration/management and staff nurses; 4 = primarily staff nurses—some administration/management; and 5 = staff nurses only (Havens & Vasey, 2003). The score ranges from 21 to 105; a high score indicates a high degree of nursing involvement, while a lower degree indicates little staff nurse involvement (Havens & Vasey, 2003).
Content experts rated the tool as receiving 1.0 as a content validity index score and additional analysis provided evidence of construct validity. The Cronbach’s alpha for the DIS was .91 to .95 and the subscales had reliabilities of .68 to .85 (Havens & Vasey, 2005). Cronbach’s alpha for the overall DIS for the 2004 study was .933 and .858 for the 2010 study.
Results
The convenience sample of RNs with clinical responsibilities from the 2004 study consisted of 290 RNs and the convenience sample from the 2010 study consisted of 111 RNs. The participants of the 2004 study were mostly female (97.6%) with an average age of 42.85 (SD = 10.54) and a majority having an education level less than a Bachelor’s of Science (BSN) degree (71.7%). The 2010 study also consisted mainly of females (95.5%) with an average age of 40.65 (SD = 11.24), who primarily held less than a BSN degree (70.3%; Table 1).
Demographics.
The first research question studied the difference in the level of RN’s actual and preferred DI in the 2010 study. A paired-sample t-test was used to determine the differences in the means of actual and preferred DI from the 2010 study. A statistically significant difference (p < .001; t = −13.61; df = 110) was found when comparing the actual and preferred DI as evidenced with an actual mean score of 1.93 (SD = 0.44) and a preferred mean score of 2.59 (SD = 0.41).
The second research question investigated the difference in the level of RN’s actual DI between the 2004 study and the 2010 study. An independent-sample t-test was used to determine the differences in the means of actual DI between the 2004 and the 2010 studies. The t-test revealed that the mean score of 1.93 (SD = 0.44) from the 2010 study decreased from the mean score of 2.10 (SD = 0.58) of the 2004 study with a statistically significant difference (p = .001; Table 2).
Independent-Sample t-Test of Actual and Preferred Decisional Involvement Between the 2004 and 2010 Studies.
The final research question examined the difference in the level of RN’s preferred DI between the 2004 study and the 2010 study. An independent-sample t-test was used to determine the differences in the means of preferred DI between the 2004 and 2010 studies. The t-test revealed that the mean score of 2.59 (SD = 0.41) from the 2010 study decreased from the mean score of 2.79 (SD = 0.52) of the 2004 study with a statistically significant difference (p < .001; Table 2).
Discussion
In reviewing the results of this study, it was found that a dissonance exists between the actual and preferred DI of RNs as well as lower mean differences in actual and preferred DI of nurses in 2010 as compared with 2004 following the implementation of a new shared governance structure. The results of this study were found to contradict the majority of research that highlights positive outcomes following shared governance implementation. In light of these findings, a discussion is presented on the difference found between actual and preferred DI and the potential factors that may have contributed to the unexpected decrease in DI following shared governance implementation.
The first research question identified the difference between the actual and preferred levels of DI among RNs in the 2010 study. The results of this study showed a statistically significant difference between the actual and preferred DI of RNs, indicating that nurses would prefer more DI than what actually exists. DI requires participation from nurses; this increased participation ultimately increases actual and preferred levels of decision-making (Kramer et al., 2009). However, what is not known is how active, willing, or confident the nurses in the study are to participate in the decision-making process, and how well the leadership team has acclimated to the role-sharing required of them for successful shared governance integration. Consistent with Hess’ (2004) assertion, the challenge arises in changing the roles and attitudes of participants. Some nurses may find it difficult to voice opinions and concerns due to fear of over stepping traditional boundaries regarding decision-making. In addition, leadership teams may not be willing or able to participate in the shifting of their roles and power to nurses.
The second and third research questions identified the difference in the level of RN’s actual and preferred DI between the 2004 study and the 2010 study. The results of this study revealed statistically significant lower means in the actual (p = .001) and preferred (p < .001) DI following the implementation of a new shared governance structure. While these actual (2004, M = 2.10; 2010, M = 1.93) and preferred (2004, M = 2.79; 2010, M = 2.59) mean differences were statistically significant, the 0.17 (2004) and 0.2 (2010) differences in means may not have clinical significance. Factors that could influence error within a study include small sample sizes, instruments that are not precise, and multiple interacting variables in which small differences can be overlooked (Burns & Grove, 2009). The results of the 2010 study were unexpected due to examination of previous research that highlighted numerous positive effects of having a shared governance structure in place. Several factors may have contributed to these results and they include: level of RN education (Donahue et al., 2008; Fitzpatrick et al., 2010; Ning et al., 2009; Zurmehly, Martin, & Fitzpatrick, 2009), participant’s age (Donahue et al., 2008; Ning et al., 2009), length of time that shared governance had been instituted (Kennerly, 1996; Moore & Wells, 2010), process of implementation (Ballard, 2010), and nursing acceptance of shared governance (Hess, 2004; Porter-O’Grady, 2001). Participation in the 2004 study was not a prerequisite for participation in the 2010 study resulting in a different population of nurses with differing perceptions.
The level of RN education and age plays an integral role in actual and preferred levels of DI. In studies by Donahue et al. (2008) and Ning et al. (2009), it was found that younger nurses and nurses with higher levels of education were more engaged at work and perceived higher levels of empowerment on the job. Nurses with a BSN degree perceived a greater sense of empowerment than those without a BSN, while nurses with a Master’s degree perceived the highest level of empowerment along with a better understanding and appreciation for shared governance (Donahue et al., 2008; Fitzpatrick et al., 2010; Zurmehly et al., 2009). Although the percentages of nurses in this study who had less than a BSN degree did decrease from the 2004 study (71.7%) to the 2010 study (70.3%), both studies had a significant percentage of nurses with less than a BSN degree. The large number of nurses that held less than a BSN degree may have contributed to the decreased results of the 2010 study through a decreased commitment and understanding of the shared governance process.
As there are a large number of RNs with less than a BSN degree in this study, greater time and commitment must be made in the training and development of nurses during shared governance implementation. Role confusion, lack of support, lack of recognition, and staff apathy are the most often reported barriers to shared governance (Ballard, 2010); therefore, management must be willing to clarify roles, offer added support and recognition, and work toward empowering and motivating their nurses. Offering reassurance to the nurses that their opinions are important to the topics being discussed within meetings and councils is crucial for management to foster continued and future participation of nurses within the decision-making process.
The length of time that shared governance was in place at the time of data collection may also have contributed to the decrease in means of actual and preferred DI. The 2010 study was conducted 1.5 years after the new shared governance was implemented within the rural Midwestern medical center. The results of the 2010 study more closely resembles the results found by Kennerly (1996) and Moore and Wells (2010) which revealed no significant changes 1 year to 18 months following shared governance implementation. In fact, the 2010 study revealed a statistically significant decrease (p = .001) in the level of actual DI from the 2004 study. One has to consider whether completing the 2010 study at a later time may have yielded different results regarding the DI levels of the RNs within this health care system. Research conducted by Jones et al. (1993) found more positive changes in organizational and professional areas 2 years post initiation of shared governance. The time frame of 1.5 years may not have allowed adequate time for the shared governance model to be accepted and understood by the participants within the system. Ballard (2010) stated that strategic implementation along with ongoing maintenance of processes will lead to the success of shared governance models.
The process used to introduce the new model of shared governance in 2009 to leadership and nurses within the rural Midwestern medical center is unknown. This is an area that should be examined to further investigate why nurses perceive they have less and also prefer less DI in 2010 than in 2004. If multiple organizational changes were occurring at the same time the new shared governance model was being introduced, these changes may have affected the participants’ responses in the 2010 study.
Examining the length of time for education and how the concept of shared governance and DI were explained to leadership and nurses is also important to consider (Hess, 2004; Moore & Wells, 2010). If nurses did not fully understand their role within the decision-making process, lacked the confidence to make decisions, or perceived DI as additional work, this may have led to indifference or resistance to the shared governance model. As a result, this may have caused nurses to prefer a lower level of DI, thus accounting for the decrease in the preferred level of DI from 2004 to 2010.
The acceptance of shared governance within the organization is another vital factor that may have affected the results of this study. Shifting from a traditional management style to a shared governance model requires the reconfiguration of not only the organizational structure, but the relationships among nurses and management, their interactions, and their decisions (Porter-O’Grady, 2001). Leadership must be willing to participate within their shifting roles and power (Hess, 2004). This shift can be difficult for those in leadership positions who are comfortable making the final decisions on unit or organizational problems having little-to-no nurse input. If leadership is unwilling to share their role of decision-making, the decision-making opportunities and abilities of the nurses will be stifled, thus causing a decrease in their actual DI.
Leaders need to offer support and encouragement for nurses to take a more active role within decision-making processes (Macphee et al., 2010), and should also guide nurses in learning responsibilities and boundaries that are associated with being members within a unit or organization council. Kramer and Schmalenberg (2003) discovered that nurses felt the work they did in shared governance council or staff meetings made a difference, allowing them to have avenues for influencing practice. Offering recognition for the nurses’ participation and contribution to these meetings and councils will lead to nurses feeling empowered and more willing to assume accountability for patient, unit, and organizational outcomes.
Managers also assist in empowering their nurses by remaining highly visible, listening to concerns, and consistently offering support (K. Armstrong et al., 2009). By assuring adequate resources, promoting learning and growth, and remaining highly visible and supportive, managers can empower their nurses and assist in controlling the nursing shortage through improvements in retention and recruitment.
Respondents to the survey may also be a factor that affected the results of this study. Participants in the 2010 study were not necessarily the same participants in the 2004 study. This change in population may result in nurses having differing levels of education and years’ experience between the two groups surveyed. These factors play a role in perception and preferred level of involvement and may account for the statistical differences in the data.
There is limited research available on the concept of DI; therefore, ongoing research is needed to further the exploration of actual and preferred levels of DI among RNs. Further exploration into potential contributing variables could also aid in accomplishing equally shared DI levels.
Due to the dissonance between actual and preferred DI among nurses in the 2010 study, an effort to increase these levels to accomplish equally shared DI levels must be made. The DIS tool can be used to identify the areas with the greatest differences in decision-making levels among nurses according to the six subscales. A concerted effort can then be made on behalf of all participants, nurses, and leadership, to overcome or minimize these differences, thus increasing the actual and preferred decision-making levels among RNs.
Further research should be conducted on the relationship between demographic variables and DI. Demographic variables that may impact DI include age, years as RN, years in position, gender, and level of education.
Process-related variables recommended for future research that may impact DI include shared governance implementation, ongoing support of processes, and adequate resource allocation. Additional variables of interest may also include the sharing of power, relationships between nurses and management, nurse empowerment, autonomy, and organizational commitment. A longitudinal study of shared governance implementation could also be beneficial in future research.
Limitations of this study exist. A convenience sample was taken from one rural Midwestern medical center, making generalizations outside this research sample difficult. In addition, not having the same respondents from the 2004 study is a limitation; newly hired nurses may hold different viewpoints, and with less time worked in the shared governance environment, may lack a full awareness of the actual DI within the organization. Differences in recruiting strategies and available workforce numbers may have also been a limitation. In addition, the sample size difference between the 2004 (n = 290) and 2010 (n = 111) studies may not have provided an accurate description of the actual and preferred DI of nurses within this health care system. Another limitation is that data were collected through the use of mailed and online surveys giving participants the choice to participate; this presents the potential for a self-selection bias in the data collection process of the 2004 and 2010 studies; and because the instrument used was a survey, response fatigue could have played a role as well.
In summary, according to the literature, enhancing the DI among RNs contributes to nurse retention and recruitment, and empowers nurses to take control of their practice. DI thus acts as one variable in addressing the retention and recruitment of nurses, and assists in alleviating the nursing shortage. Shared governance, as an organizational model, has been shown to enhance the DI of nurses when implemented successfully through careful and strategic planning. The findings of this study were contrary to the literature in that they revealed a dissonance between the actual and preferred DI of RNs and also found a decrease in the actual and preferred DI of nurses following the implementation of a new shared governance model. By using the DIS tool to identify the areas with the greatest differences in decision-making levels among nurses, a concerted effort can be made to minimize these differences and increase the actual and preferred decision-making levels among RNs. For successful implementation, the enculturation of shared governance is imperative. This requires that adequate time is allowed for acceptance of the model and the shifting and sharing of decision-making roles, careful and strategic planning, and additional training and empowerment of the nurses is a must. If adequate sharing of decision-making power does not take place between leadership and nurses, actual DI may be stifled; and without portraying the implementation of shared governance positively, the preferred DI of nurses may also be stifled. Further research of actual and preferred DI within the context of shared governance is recommended as an effort to find practical solutions to assist with the retention and recruitment of nurses within organizations.
Footnotes
Declaration of Conflicting Interest
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.