Transformational Leadership and Intent to Stay: Moving Research Into Nurse Leaders' Practice

Research Background: High turnover of staff nurses is a persistent problem for healthcare systems in the United States and globally. It negatively impacts patient outcomes and has high organizational costs estimated at $1.4-2.1 billion per year based on 2006 RN salaries (Brewer, Kovner, Greene, Tukov-Shuser, & Djukic, 2012). Although turnover is a complex phenomenon with many precursors at play, poor management by nurse leaders is one of the top reasons staff nurses leave their employers (Kovner et al., 2007).In a 2011 systematic review of 53 studies, (Cowden, Cummings, & Profetto-Mcgrath, 2011)  only three studies showed that a transformational leadership style had a direct positive significant relationship with intent to stay. Since then Laschinger et. al (Laschinger, Wong, Grau, Read, & Pineau Stam, 2012) found a similar relationship, however Abualrub (Abualrub & Alghamdi, 2012) found no relationship.  

Objective: The purpose of our study was to examine the relationship between transformational leadership style and intent to stay, a direct antecedent of actual turnover, in a nationally representative sample of early-career nurses from the U.S. using multiple covariates.

Methods: This was a cross-sectional, correlational study of nurses who had been licensed for 7.5 to 8.5 years. We used data collected in 2013 as part of a large longitudinal study begun in 2006. Our nationally representative sample of RNs licensed between August 1, 2004 and July 31, 2005 is from 51 metropolitan statistical areas and 9 rural areas in 34 U.S. States and the District of Columbia. The analytic sample in 2013 was 1,375. The response rate for the sample used here was 69%. After removing nurses who did not work with physicians, and those who did not work in a job that required an RN license or in a health related field, the analytic sample was 1,198 RNs. After removing cases in which critical data were missing, we analyzed data from a sample of 1,173. We collected data using a survey with a $5 incentive and a maximum of five contacts for non-responders. Data were collected from both a mailed survey and an internet based survey. We used an expanded Price Model of Turnover (Price, 2001)to guide inclusion of survey questions, which comprised four domains: demographics (e.g. age, race), 2) work setting (e.g. position, shift), 3) attitudes about work (e.g. satisfaction, work group cohesion), and 4) availability of work in other organizations.  Attitudes were measured using instruments that had good reliability and validity and that had been used in previous published research. Bass (Bass & Steidlmeier, 1999, p.11) notes “Transformational leadership refers to the leader moving the follower beyond immediate self-interests through idealized influence (charisma), inspiration, intellectual stimulation, or individualized consideration. It elevates the follower’s level of maturity and ideals as well as concerns for achievement, self-actualization, and the well-being of others, the organization, and society.” TL was measured with a 6-item scale (e.g. “My manager develops a team attitude and spirit in the staff in the department”) with possible response options ranging from strongly disagree to strongly agree and mean scores ranging from 1 to 7 (Hammer, Ommen, Rottger, & Pfaff, 2012). ITS was measured using a scale with 4 items (e.g. I plan to stay with my present employer as long as possible.) Mean scores ranged from 1 to 5.

Covariates included: job tenure, months worked as RN, work status, setting, unit type, overtime, highest nursing degree, age, gender, ethnicity, children at home and marital status, autonomy, MD-RN collaboration, job-family conflict, organizational constraints, promotional opportunities, quantitative workload, and local and nonlocal job opportunities, all of which have been reported in the literature and/or found in our own research to be related to ITS.  We used ordered probit multivariate regression analyses (Greene, 2008) and STATA to model the relationship between TL and ITL.

Principal Findings: The nurses in our sample were married (74%), white (86%), women (90%) with a mean age of 40 who had worked on average 7.5 years. Transformational leadership was not significantly related to intent to stay (p<0.275). It was correlated with supervisor support  (.801, p<.000) and mentor support (.583, p<.000). Neither TL nor supervisory support, entered separately into the models, directly impacted ITS. Both job satisfaction (p<0.000) and organizational commitment (p<0.000) were positive, significant predictors of ITS.

Conclusions: Transformational leadership is not directly related to RNs’ ITS, nor is a related concept, supervisory support. Our findings are in contrast to what others have found for transformational leadership. Differences include using a different sample, including different covariates, and measurement differences. Because of our large sample size we are confident that our study was sufficiently powered to find a relationship if one existed. It may be that a “fully specified” model (or one that has a sample large enough to include many covariates) provides more accurate findings than those with fewer covariates and/or smaller samples. Further, our measurement of transformational leadership may be different from how others have measured the concept. We treated ITS as an ordinal variable, while others have treated it as a ratio or interval variable.

Implications for Policy, Delivery, or Practice: Having nurse leaders who are knowledgeable about evidence-based practices for staff retention is critical to ensure a stable nursing workforce and optimal patient outcomes. Leadership development is costly; therefore leadership education and practice must be informed by research in order to be effective. Our study does not lend support for transformational leadership style as an important competency for leaders in reducing ITS. Yet, organizations invest in TL training. Based on our findings, leaders who want to increase ITS need to focus on job satisfaction and creating good work environment. Because our findings are not consistent with many other TL findings reported in the literature, further study of both transformational leadership and other concepts of leadership is warranted.