The suboptimal performance of the U.S. health care system has led to large-scale efforts to improve care processes to yield better outcomes. However, organizations often struggle implementing improvement programs (Alexander, Weiner, Shortell, Baker, & Becker, 2006). A key reason for the lack of success is likely flawed implementation (Klein, Conn, & Sorra, 2001).
Overcoming barriers to successful improvement efforts requires several factors, such as sustained support from senior leadership, extensive training for frontline employees, robust measurement and data systems, realigned incentives, and cultural receptivity to change (Alexander & Hearld, 2011). Of these facilitating factors, employee commitment to change is believed to be one of the most important factors in achieving implementation success (Herscovitch & Meyer, 2002). Commitment to organizational change, defined as a mindset that binds an individual to a course of action deemed necessary for the successful implementation of a change initiative, has been postulated as a key psychological mechanism linking organizational efforts to implement planned change to behaviors of employees (Herscovitch & Meyer, 2002).
Prior literature has focused on the important role of senior manager commitment to the successful implementation of improvement programs (Taylor & Wright, 2003). In particular, when senior managers engage in transformative leadership behaviors, such as providing followers with a clear sense of purpose, expressing confidence, and articulating a compelling vision, it increases employees’ commitment to the organizational change (Aarons et al., 2016 ; Birken et al., 2015 ; Herold, Fedor, Caldwell, & Liu, 2008). Perhaps, as a result of the central importance of senior managers’ role in improvement efforts, the contribution of middle managers has been less studied (Birken, Lee, Weiner, Chin, & Schaefer, 2013 ; Birken, Shoou-Yih, & Weiner, 2012 ; Engle et al., 2017). We define middle managers as individuals who directly supervise the work of frontline employees and who themselves report to a manager at least one hierarchical layer below the chief executive officer (Birken et al., 2012). Senior managers, in contrast, report directly to the chief executive officer and are not directly responsible for supervising employees who deliver service to customers at the frontlines of the organization.
Middle managers may play a distinct and particularly important role in improvement programs because of their strategic location between executive leaders and frontline workers. Their position enables them to rally frontline workers to embrace the process changes typically required by an improvement effort. Their position also enables them to bridge informational gaps related to the implementation. For example, middle managers can provide frontline workers with information regarding implementation, making it relevant to them; give them tools necessary for implementation; and encourage them to use those tools (Birken et al., 2013).
A few scholars have begun investigating the role of middle managers in improvement efforts (Birken et al., 2012, 2013 ; Engle et al., 2017). However, much remains to be learned as there are scant studies that investigate antecedents and consequences of middle manager commitment to improvement efforts (Birken et al., 2013). For example, current studies do not test specific factors that may influence middle manager commitment, nor do they examine many mechanisms through which middle manager commitment influences implementation success, such as through influencing frontline support.
The objective of our study is to address these gaps by shedding light on the ways in which middle manager commitment to an improvement program influences the program’s successful implementation and uncovering organizational antecedents that foster middle manager commitment. In doing so, our study makes several contributions to the literature on implementation effectiveness. First, our study provides evidence of a positive relationship between middle manager commitment and their perceptions of successful improvement program implementation. Second, we show that middle managers’ perceptions of frontline worker support mediate this relationship. Thus, our findings suggest that middle manager commitment is an important driver of perceived implementation success, in part, because it positively influences perceived frontline worker support for the improvement program. Third, we identify a bundle of five key drivers associated with middle manager commitment: having a clear implementation plan, being held accountable for program results, having adequate financial resources for program implementation, having adequate personnel resources for program implementation, and having senior manager support to overcome implementation challenges. Collectively, our study deepens theoretical understanding of the important construct of middle manager commitment and its relationship with perceived implementation success.
Employee Commitment to Organizational Change and the Important Role of Middle Manager Commitment in Improvement Program Implementation Success
Employee commitment to change is believed to be a vital factor in achieving implementation success within organizations (Herscovitch & Meyer, 2002). We define commitment to organizational change as a mindset that binds an individual to a course of action deemed necessary for the successful implementation of a change initiative (Herscovitch & Meyer, 2002). Herscovitch and Meyer (2002) investigate commitment to organizational change and show that employee commitment to change is a strong and significant predictor of behavioral support for the change. Their research has shown that commitment to change can reflect three different beliefs: (a) a desire to provide support for the change based on a belief in its inherent benefits (“affective” commitment to change), (b) a sense of obligation to provide support for the change (“normative” commitment to change), or (c) a recognition that there are costs associated with failure to provide support for the change (“continuance” commitment to change). Their studies show that these three components of commitment to change are distinct from each other and all relate to an employee’s self-reported level of behavioral support for change but that the affective dimension of commitment to change has the strongest positive correlation with desirable work behaviors (Herscovitch & Meyer, 2002). Affective commitment may relate to engaging in activities that go beyond meeting job requirements, which may in turn promote implementation success. Consequently, we focus on the affective dimension of employee commitment to change and the antecedents that may strengthen employee affective commitment. For brevity, we refer to affective commitment as commitment from this point forward.
Whereas robust literatures from both health services research and manufacturing settings provide strong support for the central role that senior manager commitment plays in successful organizational change and the implementation of improvement programs (Aarons et al., 2016 ; Lukas et al., 2007), little empirical research has been conducted on the role of middle manager commitment to the successful implementation of improvement programs (Birken et al., 2013).
Middle managers can positively impact the implementation of improvement programs for several reasons. They are close to day-to-day operations and frontline employees, relative to senior managers, and therefore may be better positioned to recognize where potential implementation challenges may arise as well as to see how the overall implementation is progressing (Tucker, Nembhard, & Edmonson, 2007). They influence unit level culture, which in turn influences frontline workers’ willingness to engage in implementation efforts (Zohar & Luria, 2010). They may engage in transformative leadership behaviors, which inform, motivate, and direct frontline workers with regard to program implementation (McFadden, Henagan, & Gowen, 2009). They may seek resources or span boundaries to address implementation challenges when they arise (Birken et al., 2012). Middle managers are also closer to senior managers than frontline workers and are thus privy to and may help influence organizational strategy and policies that are intended to shape implementation practices (Ashmos, Huonker, & McDaniel, 1998).
A nascent body of evidence in health services research on the importance of middle manager commitment in the organizational change process has recently emerged. Birken and colleagues (2012) have put forth a theory of how middle managers’ expression of commitment is a key factor in determining implementation effectiveness. They suggest that middle manager commitment to innovation implementation is essential in facilitating shared perceptions of the extent to which innovation implementation is rewarded, supported, and expected in the organization, which in turn promotes implementation effectiveness. While Birken et al.’s theory focuses on middle managers’ behavioral manifestations of commitment and how they relate to implementation effectiveness, we choose to focus on the affective dimension of commitment given its strong and positive correlation with behavioral support for organizational change (Herscovitch & Meyer, 2002). We hypothesize that stronger middle manager affective commitment to a change initiative is associated with greater implementation success (Figure 1). More formally, we hypothesize the following:
Hypothesis 1 (H1): Higher levels of middle manager affective commitment to a quality improvement program are associated with higher levels of program implementation success.
Middle Manager Commitment May Increase Frontline Worker Support
We build upon Birken and colleagues’ (2012) theory of how middle manager commitment relates to implementation success in a number of important ways. First, the management and strategy literatures suggest that an important way middle managers may influence organizational change is through enabling the success of frontline workers, yet the health services literature has not established a quantitative link between middle manager commitment to a quality improvement program and frontline worker support for that program. Middle manager commitment may impact implementation success indirectly through its impact on frontline employees’ support for the improvement program. Researchers emphasize that implementation success depends on the ability and willingness of frontline employees to use the new processes in their daily work routines (Shortell, Bennett, & Byck, 1998 ; Tucker et al., 2007). Tucker and colleagues (2007), for example, find that frontline staff need to adapt the new (and better) practices developed outside their organization for those practices to be successful in the unique context of their neonatal intensive care unit. Middle managers can facilitate this adaptation by establishing a culture that values improvement (Singer, Hayes, Gray, & Kiang, 2015), encouraging frontline workers to be involved with improvement efforts, and overseeing the project teams’ implementation efforts. Furthermore, middle manager commitment to change has been theorized as a key link between an organization’s change initiative and employees’ willingness to accept the new initiative (Herscovitch & Meyer, 2002 ; Meyer, Srinivas, Lal, & Topolnytsky, 2007). Collectively, these studies suggest that strong middle manager commitment to an improvement program has a positive effect on implementation success via its ability to increase frontline worker support for the program. Thus, we hypothesize the following:
Hypothesis 2 (H2): The positive relationship between middle manager affective commitment to a quality improvement program and the program’s implementation success is mediated by frontline worker support for the program.
How Organizational Support for an Improvement Program May Increase Middle Manager Commitment and In Turn Influence Improvement Program Implementation Success
A second important contribution we offer to Birken and colleagues’ theory is drawing upon implementation research that has illustrated the importance of organizational support in helping to bring about the successful implementation (Klein et al., 2001 ; Weiner, Lewis, & Linnan, 2009). Organizational support can take many forms, including structures, plans, policies, financial and personnel support, training and education, integrated data systems, and clinical integration (Alexander et al., 2006 ; Damschroder et al., 2009 ; Klein et al., 2001). Transformational leadership has been associated with these types of organizational support (Herold et al., 2008). Studies suggest that organizational support enables the full implementation of a new process or technology (Alexander et al., 2006 ; Damschroder et al., 2009). Literature suggests that, if an organization provides structures, plans, and policies that support middle managers’ implementation efforts, middle manager commitment to that change will increase (Harrington & Williams, 2004).
We draw on these studies to theorize that an organization’s support for implementation (e.g., financial and personnel resources, incentives, implementation plans) fosters middle manager commitment to the change, which, in turn, leads to higher rates of implementation success. In other words, organizational support is important for implementation success because it is an antecedent to middle manager commitment. More formally, we hypothesize:
Hypothesis 3 (H3): The positive relationship between an organization’s support for a quality improvement program and the program’s implementation success is mediated by manager affective commitment to the program.
Study Design and Context
We conduct an exploratory cross-sectional survey of all nurse managers from 30 U.S. hospitals. Nursing is an ideal context to study the middle manager role in health care improvement implementation because of the clearly delineated hierarchy of managerial roles within the profession. At the top of the hospital nursing hierarchy is a chief nursing officer, the highest-ranking nurse executive, who is responsible for overseeing and coordinating a hospital’s nursing department and its daily operations. The chief nursing officer directly oversees associate chief nurses (sometimes called nursing directors), who are senior managers in charge of a hospital service line. Each associate chief nurse, in turn, supervises nurse managers, who typically oversee one to two hospital units and the frontline nurses who work on those units.
Our survey was administered as part of a larger study of nursing leadership across 30 hospital sites. Fourteen hospitals belonged to a large hospital system in Texas, 13 hospitals belonged to a large hospital system in New York, and the remaining three hospitals were medium-to-large medical centers located in Florida, North Carolina, and Connecticut. Sample hospitals were larger than the average hospital in the United States (15 hospitals had 250+ beds). Internal review board approval was received by all participating hospitals and investigators.
Online survey data were collected as a module in a larger, multipurpose leadership survey of nurse leaders (1,569) in participating hospitals between September 2014 and January 2015. Invitations to participate in the study were sent out via email by each hospital’s chief nursing officer and included a link to the electronic survey. Three email reminders were sent to request survey participation. We received 569 responses to the larger leadership survey (36% response rate).
Of those participants who accessed the larger survey, 53% (304) received the module with additional questions related to the implementation of their hospital’s falls prevention improvement program after indicating that the hospital units that they oversaw were currently participating in a falls reduction improvement program. While we collected data across 246 nurse leaders, our final analytic sample included 67 nurse middle managers. Response rates by hospital do not correlate significantly with respondents’ perceptions of falls prevention program implementation success.
Variables and Measures
We operationalize the domains in our conceptual framework using established measures that we adapt to a health care context. All items are taken from existing literature with exception of the items representing organizational support for the program and behavioral commitment to the program, which we identified in consultation with the literature and qualitative interviews with a convenience sample of nurse managers (N = 12) as part of survey development.
All survey questions pertain specifically to the implementation of a falls prevention improvement program currently underway in the hospital (see Supplemental Digital Content 1, http://links.lww.com/HCMR/A29, for a complete list of survey items). We accomplished this by specifying in the introductory text and in the stem of the survey items that our questions related to the respondent’s unit falls prevention improvement program. Only nurse leaders who managed units with a falls prevention improvement program currently underway received our survey questions. Note that we were not implementing a specific falls reduction program but instead studied units that had implemented any type of falls reduction program.
We focus on a falls prevention improvement program for two reasons. First, focusing on a single type of improvement program reduces variation across respondents. Second, falls reduction programs are widely implemented across hospitals in the United States, which increases the number of managers who would be able to respond to our survey questions. The American Nurses Association, the Joint Commission, and the Institute for Healthcare Improvement, for example, recommend common protocols and strategies to reduce patient fall rates in the hospital setting. Common protocols include risk assessments for patients, patient and staff education, bedside signs and wristband alerts, footwear advice, scheduled and supervised toileting, and a medication review (Miake-Lye, Hempel, Ganz, & Shekelle, 2013). For all items, respondents use a 7-point Likert scale to indicate their level of agreement (1 = strongly disagree, 4 = neither agree nor disagree, 7 = strongly agree).
We measure our dependent variable of implementation success with a three-item survey scale developed by Noble and Mokwa (1999). They define implementation success as “the extent to which an implementation effort is considered successful by the organization.” We adapt their three-item scale to the health care context. Because we gather these data from middle managers as opposed to by using objective measures of performance, our dependent variable is more accurately termed “perceived implementation success.” Although it would be ideal to use an objective measure of falls reduction over time, we were unable to collect these data. However, prior studies using subjective scales of improvement find that staff perceptions are highly correlated with objective measures (Hansen, Williams, & Singer, 2011 ; Tucker et al., 2007).
Our analyses focus on three main predictors and how they relate to the falls prevention program’s implementation success. The first main independent variable of interest is manager affective commitment to the program. Commitment is a multidimensional construct. We chose to focus on measuring one specific type of commitment—managers’ affective commitment to change—instead of two other types of commitment identified by Herscovitch and Meyer (2002): continuance commitment and normative commitment. Affective commitment is a desire to provide support for the program based on its inherent benefits. We focus on affective commitment because this dimension of commitment has the strongest positive correlation with desirable work behaviors (Herscovitch & Meyer, 2002). To measure frontline worker support for the improvement program, we draw on Klein and colleagues’ six-item validated measure of management support for implementation, adapting survey items to assess managers’ views of frontline worker support rather than manager support (Klein et al., 2001). Again, as these data are gathered from middle managers, it can be thought of more precisely as “middle manager perceptions of frontline worker support.” Finally, we use five survey items to measure managers’ perceptions of the level of organizational support for the falls prevention program, such as whether there is a clear falls program implementation plan or whether the manager feels that he or she could get senior manager support when implementation challenges are encountered.
Using middle managers’ responses to construct both dependent and independent variables may introduce unmeasured variance to regression models. To control for this common method bias, we include in our regression model a self-reported survey item scale that is theoretically unrelated to and minimally correlated with middle manager affective commitment and perceptions of improvement program implementation success (Lindell & Whitney, 2001). The survey construct that we use for this purpose is middle managers’ attitudes of professionalism. Please see Supplemental Digital Content 1 (Supplemental Digital Content 1, http://links.lww.com/HCMR/A29) and Supplemental Digital Content 2 (Supplemental Digital Content 2, http://links.lww.com/HCMR/A30) for survey item details including correlations with our main model constructs.
Additional control variables in our regression models include age, gender (male or female), nursing education (bachelors/associates or masters/PhD), race (White or non-White), years of direct patient care experience, years of administrative experience, whether the respondent belongs to one of the two large hospital systems surveyed, the length of time the falls prevention program implementation has been underway, and the number of beds on the respondent’s nursing unit.
We provide sample characteristics in Table 1 and construct means, correlations, and standardized Cronbach’s alphas in Table 2. We also assess survey properties including item nonresponse, means, and variance. To evaluate internal consistency and construct reliability, we use the common threshold of Cronbach’s alpha greater than or equal to .70 (Nunnally, 1967). All constructs exceed the .70 threshold for internal consistency (Table 2). We average all items within a construct to create a composite variable for each respondent.
To examine the relationships between our independent variables of interest and our dependent variable, we fit two-level hierarchical linear models with hospital random effects, which account for the nesting of nurse managers within hospitals. We choose random effects because the random effect estimators have smaller variances than fixed effect estimators (Wooldridge, 2002). We conduct a robustness check using a fixed effects model and find similar results (Supplemental Digital Content 2, http://links.lww.com/HCMR/A30). A likelihood ratio test comparing linear regressions with two-level models finds significant differences, confirming the appropriateness of using a multilevel model (Snijders & Bosker, 1999).
To identify the potential mediation effect of perceived frontline worker support for the improvement program on the relationship between middle manager affective commitment and perceived program implementation success (H2), as well as the potential mediation effect of middle manager affective commitment on the relationship between perceived organizational support for the improvement program and the program’s perceived implementation success (H3), we implement a mediation analysis. We follow the procedure described by Krull and MacKinnon (2001) to identify the total effect of the independent variable on the outcome variable, which equals the sum of their direct and indirect effects, using multilevel modeling. The indirect effect equals the product of the effect of the independent variable on the mediator and the effect of the mediator on the outcome variable, while controlling for the independent variable. We apply a bootstrap extension to our analyses generating 5,000 random samples to estimate standard errors and obtain a confidence interval for our point estimates (Preacher & Hayes, 2008).
For each of the analyses described, we control for all available demographic characteristics of survey respondents to account for potential differences in survey responses that may be driven by a respondent’s gender, age, race, education, years of patient care experience, years of administrative experience, manager level, professional attitudes, whether the respondent belongs to one of the two large hospital systems surveyed, and the number of beds on the respondent’s unit as well as the length of time the falls prevention program has been underway in the hospital. We perform all quantitative analyses using STATA/MP 13.1.
To ensure that our results are not driven by our random effects model, we carried out a number of sensitivity analyses using alternative model specifications and variable construction to assess the robustness of our findings. We fit a hospital fixed effects model to examine the relationship of our dependent variable and main independent variables of interest.
We also carried out a sensitivity analysis to examine middle manager behaviors that may be demonstrative of their commitment to the improvement program. Prior literature suggests the importance of studying expressions of middle manager commitment to program implementation (Birken et al., 2012, 2013). We surveyed middle manager behaviors related to program implementation including the frequency with which they engaged in one-on-one conversations with staff about the improvement program, discussed the program during unit meetings, provided feedback to individual staff members on their personal performance related to the program, modified or redesigned the program during or after program implementation, reviewed data on the unit’s performance related to the program, and shared data or feedback on the unit’s performance related to the program with staff. Survey measures can be found in Supplemental Digital Content 1 (see Supplemental Digital Content 1, http://links.lww.com/HCMR/A29).
The characteristics of respondents are shown in Table 1. Most respondents are female (87%), which is typical for the nursing profession. More than half of the respondents have at least a nursing master’s or doctoral degree. Years of direct patient care average 15, and years of administrative experience average 9 among respondents.
Table 2 reports the summary statistics, Cronbach’s alphas, and zero-order correlations for all of our constructs. All constructs are positively and significantly correlated with one another (p < .05) but are .69 or less, which is below the threshold of .80 where multicollinearity would be a concern (Chatterjee & Hadi, 1986). To further test for multicollinearity, we also compute the variance inflation factors for all regression equations, and all values are below 5, again indicating that multicollinearity is not a concern (Chatterjee & Hadi, 1986). Table 2 presents construct means and standard deviations for middle managers.
Relationship of Perceived Program Implementation Success and Middle Manager Affective Commitment, Frontline Worker Support, and Organizational Support
We present the results of our hierarchical linear model testing the independent effects of middle manager affective commitment to the falls program, their perceptions of frontline worker support for the falls program and organizational support for the falls program, and their perceptions of implementation success of the program in Table 3. Regression results suggest that all three independent variables of interest are independently, significantly, and positively associated with middle manager perceptions of program implementation success (p < .001) after controlling for respondent demographics and organizational characteristics. A 1-point higher score of manager affective commitment is associated with a 0.40 point higher implementation success score for middle managers (p < .001), independent of frontline worker support or organizational support. These results offer support for H1.
Frontline Worker Support as a Mediator of Middle Manager Affective Commitment and Perceived Falls Improvement Program Implementation Success
Results of mediation analyses suggest that the total effect of middle manager affective commitment on perceptions of implementation success is positive and statistically significant (p < .001), as are each of the two component effects comprising the indirect effect (the effect of affective commitment on frontline worker support [p < .001] and the effect of frontline worker support on implementation success [p < .001]; Table 4). Analyses suggest that the indirect effect is also positive and statistically significant (p < .001), which suggests that frontline worker support is a mediator of the relationship between manager affective commitment and program implementation success. Because the direct effect of affective commitment on implementation success when controlling for frontline worker support is still positive and significant (p < .001), we conclude that frontline worker support only partly explains the effect of manager affective commitment on implementation success. These results offer support for H2.
Middle Manager Affective Commitment as a Mediator of Organizational Support and Perceived Falls Improvement Program Implementation Success
The total effect of middle manager perception of organizational support on perceived implementation success is positive and statistically significant (p < .001), as are each of the two component effects comprising the indirect effect (the effect of perceived organizational support on manager affective commitment [p < .001] and the effect of perceived organizational support on perceived implementation success [p < .001]; Table 4). The test of mediation indicates that the indirect effect is also positive and statistically significant (p < .001), which suggests that manager affective commitment is a mediator of the relationship between perceived organizational support for the falls program and perceived program implementation success. These results support H3. Because the direct effect of perceived organizational support for the falls program on perceived implementation success when controlling for manager affective commitment is still positive and significant (p < .001), we conclude that manager affective commitment only partly explains the effect of perceived organizational support on perceived implementation success. This result further underscores the central role middle manager affective commitment has in mediating the relationship between perceived organizational support for the falls program and perceived implementation success.
Regression analyses are robust to alternative model specifications. Hospital fixed effects model results are available upon request. Regression results that control for middle manager behaviors that may be demonstrative of their commitment to the falls reduction improvement program suggest that all three independent variables of interest (middle manager affective commitment, perceptions of frontline worker support, and organizational support for the falls improvement program) are independently, significantly, and positively associated with middle manager perceptions of program implementation success (p < .001; Supplemental Digital Content 3, http://links.lww.com/HCMR/A31).
We find a positive and significant association between middle manager affective commitment to an improvement program and middle manager perception of program implementation success. Their perception of frontline worker support for the falls program partially mediates this relationship. Finally, we find a bundle of five organizational variables that are associated with middle manager affective commitment. The five variables are as follows: having a clear implementation plan, being held accountable for program results, having adequate financial resources for program implementation, having adequate personnel resources for program implementation, and having senior manager support to overcome implementation challenges.
Contributions to Theory
This study makes a number of important contributions to the theory of organizational change and implementation. First, to our knowledge, this is one of the first studies to provide strong quantitative support of the significant role that middle manager affective commitment plays in bringing about successful implementation of an improvement program. The only prior quantitative study that we know of that attempts to explore the role of middle manager commitment and implementation success is that of Birken and colleagues (2013). Their empirical analyses found weak evidence (not significant at the 5% level) for a relationship between behavioral manifestations of middle manager commitment and improvement implementation effectiveness. One reason for their weak finding may be that their study considered implementation leaders for teams involved in a health care collaborative as middle managers. We empirically assess middle manager commitment and additional correlates to improvement program implementation success within an enduring managerial hierarchy, in which middle managers have direct and continuous oversight of frontline workers. Our finding is important because prior implementation literature has emphasized the role of senior managers rather than that of middle managers.
Another valuable theoretical contribution is that our research findings highlight the role that frontline worker support and organizational support each has in influencing improvement program implementation success. Mediation analyses suggest that an important way in which middle manager affective commitment influences improvement program implementation success may be through facilitating increased frontline worker support for the improvement program. Mediation analyses also suggest that a key pathway through which organizational factors in support of an improvement program influence implementation success is by fostering increased levels of affective commitment, particularly among middle managers.
Study Limitations and Future Research
Our results should be considered in light of study limitations. First, this study examined only associations and not causal relationships because of the cross-sectional nature of our data. Moreover, we cannot rule out the possibility that the associations we see are due to reverse causality (i.e., middle managers are more committed to an improvement program that was successful) or confirmation bias, which is the tendency to seek, favor, and recall information in a way that confirms one’s preexisting beliefs or hypotheses, while giving disproportionately less consideration to alternative possibilities. Lawler, Thye, and Yoon (2008), for example, illustrate in a laboratory experiment that an individual’s affective dimension of commitment is strengthened by positive feelings and perceptions related to the outcome of tasks (Lawler et al., 2008). It may be therefore that positive feelings that result from an improvement program that is proceeding well will trigger a higher level of commitment in a middle manager. To address this shortcoming, future research could assess middle manager commitment at the start of an improvement project and then gather performance measures at the end of the implementation. Such a design would allow for examinations of causal relationships. Furthermore, repeated measures of both commitment and performance could shed light on the dynamic nature of the relationship between the two constructs.
Second, there is concern for selection bias. Because the hospital chief nursing officer distributed our survey, we are unable to track nonrespondents. Consequently, we cannot conduct nonresponse analyses. It is possible that nurse managers who were more actively engaged in improvement efforts in their hospital were more likely to respond to our survey. This may have resulted in more positive responses, which may lead to overestimation of our results. However, this bias may also have resulted in less variance in our measures, which could have reduced our ability to detect significant relationships. Finally, our sample was homogenously White and limited to particular geographic regions. Results thus may not be generalizable to nurse managers with other racial backgrounds and from other geographic areas.
Finally, it is important to acknowledge that our findings may not generalize to all improvement programs because falls prevention programs are primarily nursing-specific activities (Miake-Lye et al., 2013). Other types of improvement programs (e.g., sepsis mortality reduction) that require considerable collaboration with or input from other professional subgroups, such as physicians, may be less impacted by nurse middle manager affective commitment. Consequently, future research could continue exploring the role of middle manager commitment by examining projects that are more cross-functional in scope.
Our results raise several important questions that should be addressed in future research. First, given that our findings underscore the importance of manager affective commitment in influencing perceived implementation success, further research should be conducted to explore additional ways through which organizations can increase levels of affective commitment among middle managers. Second, research should explore the extent to which middle manager commitment may influence implementation success in the absence of senior manager commitment. Can a committed middle manager overcome lack of senior manager commitment and influence implementation success? Similarly, future research could investigate the dynamic interactions between senior manager, middle manager, and frontline worker affective commitment. It may be that middle manager commitment is less important in organizations where senior managers have created highly empowered frontline workers.
Our study findings also have important implications for health care practice. Results lend strong support for the positive association between middle manager affective comitment to an improvement project and their perception of implementation success, suggesting that a practical strategy for implementation effectiveness may be to foster affective commitment among middle managers. Findings point to a bundle of five key organizational variables that collectively may increase levels of manager affective commitment.
Results also highlight the important role middle managers have in fostering improvement program implementation support among frontline workers. Mediation analyses suggest that middle managers’ affective commitment influences perceived implementation success, in part, by increasing their perceptions of frontline worker support for the improvement program. Strategies to enhance implementation success among health care organizations should include focused efforts on fostering middle managers’ affective commitment to the program being implemented, as increased levels should, in turn, facilitate increased frontline worker support for the program and, through that, program implementation success.
The authors are grateful to Jeffrey Adams for his support and for allowing them to partner with him on survey administration.
Aarons G. A., Green A. E., Trott E., Willging C. E., Torres E. M., Ehrhart M. G., Roesch S. C. (2016). The roles of system and organizational leadership in system-wide evidence-based intervention sustainment: A mixed-method study. Administration and Policy in Mental Health and Mental Health Services Research
, 43(6), 991–1008.
Alexander J. A., Hearld L. R. (2011). The science of quality improvement implementation: Developing capacity to make a difference. Medical Care
, 49, S6–S20.
Alexander J. A., Weiner B. J., Shortell S. M., Baker L. C., Becker M. P. (2006). The role of organizational infrastructure in implementation of hospitals’ quality improvement. Hospital Topics
, 84(1), 11–20.
Ashmos D. P., Huonker J. W., McDaniel R. R. Jr. (1998). Participation as a complicating mechanism: The effect of clinical professional and middle manager participation on hospital performance. Health Care Management Review
, 23(4), 7–20.
Birken S. A., Lee S. Y., Weiner B. J., Chin M. H., Chiu M., Schaefer C. T. (2015). From strategy to action: How top managers’ support increases middle managers
to innovation implementation in healthcare organizations. Health Care Management Review
, 40(2), 159–168.
Birken S. A., Lee S. Y., Weiner B. J., Chin M. H., Schaefer C. T. (2013). Improving the effectiveness of health care innovation implementation: Middle managers
as change agents. Medical Care Research and Review
, 70(1), 29–45.
Birken S. A., Shoou-Yih D. L., Weiner B. J. (2012). Uncovering middle managers
’ role in healthcare innovation implementation. Implementation Science
, 7(1), 28.
Chatterjee S., Hadi A. S. (1986). Influential observations, high leverage points, and outliers in linear regression. Statistical Science
, 1(3), 379–416.
Damschroder L. J., Aron D. C., Keith R. E., Kirsh S. R., Alexander J. A., Lowery J. C. (2009). Fostering implementation of health services research findings into practice: A consolidated framework for advancing implementation science. Implementation Science
, 4(1), 50.
Engle R. L., Lopez E. R., Gormley K. E., Chan J. A., Charns M. P., Lukas C. V. (2017). What roles do middle managers
play in implementation of innovative practices? Health Care Management Review
, 42(1), 14.
Hansen L. O., Williams M. V., Singer S. J. (2011). Perceptions of hospital safety climate and incidence of readmission. Health Services Research
, 46(2), 596–616.
Harrington D., Williams B. (2004). Moving the quality effort forward—The emerging role of the middle manager. Managing Service Quality: An International Journal
, 14(4), 297–306.
Herold D. M., Fedor D. B., Caldwell S., Liu Y. (2008). The effects of transformational and change leadership on employees’ commitment
to a change: A multilevel study. Journal of Applied Psychology
, 93(2), 346.
Herscovitch L., Meyer J. P. (2002). Commitment
to organizational change: Extension of a three-component model. Journal of Applied Psychology
, 87(3), 474.
Klein K. J., Conn A. B., Sorra J. S. (2001). Implementing computerized technology: An organizational analysis. Journal of Applied Psychology
, 86(5), 811.
Krull J. L., MacKinnon D. P. (2001). Multilevel modeling of individual and group level mediated effects. Multivariate Behavioral Research
, 36(2), 249–277.
Lawler E. J., Thye S. R., Yoon J. (2008). Social exchange and micro social order. American Sociological Review
, 73(4), 519–542.
Lindell M. K., Whitney D. J. (2001). Accounting for common method variance in cross-sectional research designs. Journal of Applied Psychology
, 86(1), 114–121.
Lukas C. V., Holmes S. K., Cohen A. B., Restuccia J., Cramer I. E., Shwartz M., Charns M. P. (2007). Transformational change in health care systems: An organizational model. Health Care Management Review
, 32(4), 309–320.
McFadden K. L., Henagan S. C., Gowen C. R. (2009). The patient safety chain: Transformational leadership’s effect on patient safety culture, initiatives, and outcomes. Journal of Operations Management
, 27(5), 390–404.
Meyer J. P., Srinivas E. S., Lal J. B., Topolnytsky L. (2007). Employee commitment
and support for an organizational change: Test of the three-component model in two cultures. Journal of Occupational and Organizational Psychology
, 80(2), 185–211.
Miake-Lye I. M., Hempel S., Ganz D. A., Shekelle P. G. (2013). Inpatient fall prevention programs as a patient safety strategy: A systematic review. Annals of Internal Medicine
, 158(5, Pt. 2), 390–396.
Noble C. H., Mokwa M. P. (1999). Implementing marketing strategies: Developing and testing a managerial theory. The Journal of Marketing
, 63(4), 57–73.
Nunnally J. C. (1967). Psychometric theory
. New York, NY: McGraw-Hill.
Preacher K. J., Hayes A. F. (2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods
, 40(3), 879–891.
Shortell S. M., Bennett C. L., Byck G. R. (1998). Assessing the impact of continuous quality improvement on clinical practice: What it will take to accelerate progress. Milbank Quarterly
, 76(4), 593–624.
Singer S. J., Hayes J. E., Gray G. C., Kiang M. V. (2015). Making time for learning-oriented leadership in multidisciplinary hospital management groups. Health Care Management Review
, 40(4), 300–312.
Snijders T. A., Bosker R. J. (1999). Introduction to multilevel analysis
. London, England: Sage.
Taylor W. A., Wright G. H. (2003). A longitudinal study of TQM implementation: Factors influencing success and failure. Omega
, 31(2), 97–111.
Tucker A. L., Nembhard I. M., Edmondson A. C. (2007). Implementing new practices: An empirical study of organizational learning in hospital intensive care units. Management Science
, 53(6), 894–907.
Weiner B. J., Lewis M. A., Linnan L. A. (2009). Using organization theory to understand the determinants of effective implementation of worksite health promotion programs. Health Education Research
, 24(2), 292–305.
Wooldridge J. M. (2002). Econometric analysis of cross section and panel data
. Cambridge, MA: The MIT Press.
Zohar D., Luria G. (2010). Group leaders as gatekeepers: Testing safety climate variations across levels of analysis. Applied Psychology
, 59(4), 647–673.
commitment; implementation success; improvement programs; middle managers
Supplemental Digital Content
Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved