RESEARCH ARTICLE

The need for strong clinical leaders –

Transformational and transactional

leadership as a framework for resident

leadership training

Barbara Saravo1, Janine Netzel2, Jan Kiesewetter1*

1 Institut für Didaktik und Ausbildungsforschung in der Medizin, Klinikum der Ludwig-Maximilians-Universität

München, Munich, Bavaria, Germany, 2 Center for Leadership and People Management, Ludwig-

Maximilians-Universität München, Munich, Bavaria, Germany

* jan.kiesewetter@med.lmu.de

Abstract

Background

For the purpose of providing excellent patient care, residents need to be strong, effective

leaders. The lack of clinical leadership is alarming given the detrimental effects on patient

safety. The objective of the study was to assess whether a leadership training addressing

transactional and transformational leadership enhances leadership skills in residents.

Methods

A volunteer sample of 57 residents from postgraduate year one to four was recruited across

a range of medical specialties. The residents took part in an interventional controlled trial.

The four-week IMPACT leadership training provided specific strategies for leadership in the

clinical environment, addressing transactional (e.g. active control, contingent reward) and

transformational leadership skills (e.g. appreciation, inspirational motivation).

Transactional and transformational leadership skill performance was rated (1) on the Per-

formance Scale by an external evaluator blinded to the study design and (2) self-assessed

transformational and transactional leadership skills. Both measures contained items of the

Multifactor Leadership Questionnaire, with higher scores indicating greater leadership skills.

Results

Both scores were significantly different between the IMPACT group and the control group.

In the IMPACT group, the Performance Scale increased 15% in transactional leadership

skill performance (2.10 to 2.86) (intervention effect, 0.76; 95% CI, 0.40 to 1.13; p < .001, eta2 = 0.31) and 14% in transformational leadership skill performance (2.26 to 2.94) (inter-

vention effect, 0.68; 95% CI, 0.27 to 1.09; p < .001, eta2 = 0.22). The self-assessed transac- tional skills revealed a 4% increase (3.83 to 4.03) (intervention effect, 0.20; 95% CI, 0.08 to

0.33; p < .001, eta2 = 0.18) and a 6% increase in transformational leadership skills (3.54 to 3.86) (intervention effect, 0.31; 95% CI, 0.02 to 0.40; p< .001, eta2 = 0.53).

PLOS ONE | https://doi.org/10.1371/journal.pone.0183019 August 25, 2017 1 / 13

a1111111111

a1111111111

a1111111111

a1111111111

a1111111111

OPENACCESS

Citation: Saravo B, Netzel J, Kiesewetter J (2017)

The need for strong clinical leaders –

Transformational and transactional leadership as a

framework for resident leadership training. PLoS

ONE 12(8): e0183019. https://doi.org/10.1371/

journal.pone.0183019

Editor: Mirjam Körner, University of Freiburg,

GERMANY

Received: October 6, 2016

Accepted: July 28, 2017

Published: August 25, 2017

Copyright: © 2017 Saravo et al. This is an open access article distributed under the terms of the

Creative Commons Attribution License, which

permits unrestricted use, distribution, and

reproduction in any medium, provided the original

author and source are credited.

Data Availability Statement: The dataset

underlying our results is publicly available from the

Open Data LMU Repository. The DOI for our data is

10.5282/ubm/data.109.

Funding: This work was funded by the

Förderprogramm für Forschung und Lehre

(FöFoLe) to JK by a grant to the Klinikum of the

Ludwig-Maximilians-Universität Munich. The

funder had no role in study design, data collection

and analysis, decision to publish, or preparation of

the manuscript.

Discussion and conclusions

These findings support the use of the transactional and transformational leadership frame-

work for graduate leadership training. Future studies should incorporate time-latent post-

tests, evaluating the stability of the behavioral performance increase.

Introduction

There is currently a consensus that young physicians are in need of training in how to be effec-

tive leaders[1–5]. Medical residents take on various leadership responsibilities in their daily

clinical work. For instance, they negotiate care plans, teach medical trainees, balance diverging

perspectives in multiprofessional teams, while providing effective, safe delivery of care[5, 6].

Residents’ leadership skills are not facilitated enough [5]. This is surprising, given the body of

evidence that highlights the association between effective leadership and the improvement of

medical care in fields such as teamwork[7–9], communication[3] and patient safety[10, 11].

Current perspectives on clinical leadership are not precisely conceptualized. Existing con-

cepts mainly reflect traditional understandings of the leader-follower relationship, focusing on

individual behaviors and positional power[12]. As physicians in training usually do not hold

formal authority and their leadership roles are not clearly defined[13], a distinct approach in

framing the concept of leadership is needed[6]. Through a more precise concept of leadership,

specific leadership behaviors could be identified and trained.

A vast number of leadership theories have been established in the organizational literature

[14]. However, the medical community is facing the challenge of identifying appropriate con-

cepts, and of adjusting them to the complex clinical environments in order to design target-

oriented training programs. For graduate medical education, it has been recommended to

base leadership training programs on established best practices[5, 15].

A few attempts in developing resident leadership training have been made. However, as has

been found in a current review on leadership training in undergraduate medical education

[15], most programs vary greatly in aligning the curricula with competencies. Also, existing

studies rarely associate outcomes to the interventions[16, 17]. Leadership programs for resi-

dents mostly adhere to broad dimensions, such as confidence[18], communication skills[3],

emotional intelligence[19], or postgraduate careers[20] and organizational leadership[21]. To

date, there is no systematic, evidence-based knowledge about effective strategies to cultivate

outstanding, strong leaders in residency.

In this article, we introduce the Full Range Leadership Model (FRLM)[22] into resident lead- ership training. It is the best empirically studied and most significant leadership framework in

organizational literature and has been successfully implemented in several fields of application

[23–27]. First efforts have been made to empirically test the model in medicine, however these

have primarily addressed nursing leadership[28, 29] and hospital management[30–32].

According to the authors of the FRLM[33], a specific set of leadership components is neces-

sary for effective leadership: a large portion of transformational leadership, higher levels of

transactional leadership and a minimum of passive leadership. Transformational leadership refers to leaders with an appealing vision for their team who intellectually stimulate others in a

way that is demanding and appreciative of the individual needs of the team members[14].

Transactional leaders exert influence on followers based on exchanging benefits for outstand- ing performance and response to their self-interests when they have achieved defined goals

[34]. In contrast to transformational and transactional leadership, Bass[35] defines leaders

Transformational and transactional leadership training for residents

PLOS ONE | https://doi.org/10.1371/journal.pone.0183019 August 25, 2017 2 / 13

Competing interests: The authors have declared

that no competing interests exist.

who do not take charge of their leadership role as passive leaders. These three components of leadership are conceptualized as different levels of activity a leader can display, with passive

leadership as the least active form of leadership[36]. For example, passive leaders avoid inter-

vening when mistakes are made and do not execute managerial functions[37]; especially in

high-stakes organizations like medical care, passive leadership can have harmful consequences

[38]. On the contrary, transactional leaders reinforce their leadership by exercising active con-

trol when problems arise. Given the detrimental effects of passive leadership, we consider

transactional and transformational leadership a crucial part of clinical leadership, where

patient safety is among the highest of priorities. Transformational and transactional leadership

are conceptualized as two distinct, yet interrelated components of leadership behavior[39].

Therefore, leadership programs should address both components alike.

We argue that, for ensuring high-quality delivery of care and for maximizing clinical pro-

ductivity, physicians are expected to formulate clear expectations, set high standards and moti-

vate team members to make strides to meet specified requirements. If, for example, followers

get a feeling of involvement and are rewarded for making good efforts, they are more likely to

be eager to achieve the goals that have been set. A clinical leader should both have the capacity

to be transformational and transactional, but always be able to exert active control when

needed.

Prior research has found positive effects of transformational and transactional leadership

on several outcomes, such as enhanced satisfaction[40], the willingness of followers to generate

extra effort[41], and increased performance[27].

While prior research showed that transformational, transactional and passive leadership are

applicable in evaluating leadership styles in residents[42] and senior physicians[43], so far, no

study has examined whether the model is suitable to guide resident leadership training and

advance clinical leadership.

Gabel[13] particularly calls for training programs for informal leaders addressing transfor-

mational leadership. In our four-week IMPACT leadership training for residents, we explicitly

tied those transformational and transactional leadership skills to the curriculum that are most

relevant for everyday clinical practice. The training curriculum included the acquisition of key

leadership knowledge, application of practical leadership skills, and simulation-based role-

plays representing real performance situations of inpatient teams.

Objectives

We hypothesize that over the course of the IMPACT training, (1) residents’ performance of

transformational and transactional leadership skills as rated by an external evaluator will

improve, (2) self-assessed transformational and transactional leadership skills will increase,

and (3) residents’ knowledge on leadership will expand.

Methods

Sample

For organizational reasons we split the IMPACT training group into four cohorts, with each

cohort consisting of 10–15 persons. Cohort one and two took part in the IMPACT leadership

training program in February/March 2015 and cohort three and four in August/September

2015. All residents at our institution were eligible for enrollment in the study. Participation

was voluntary and free of charge. For inclusion in the study, residents had to be 1) affiliated to

one of the clinics of the university hospital, 2) in residency training for up to four years, and 3)

willing to participate in all four consecutive training sessions. 57 residents were included in the

study, representing a range of specialties: internal medicine, pediatrics, surgery, psychiatry,

Transformational and transactional leadership training for residents

PLOS ONE | https://doi.org/10.1371/journal.pone.0183019 August 25, 2017 3 / 13

anesthesiology, neurology, radiology, gynecology, dermatology and ophthalmology. Exclusion

criteria were 1) affiliation to an institution other than a university hospital, 2) in residency

training for more than four years, 3) not committing to participation in all training sessions.

According to these criteria, six applicants had to be excluded from study entry. Participants of

the control group (n = 23) were recruited via email listings between cohort two and three, after all available positions for the IMPACT group had been assigned.

Training procedure

The training design and procedure was based on and adapted from a leadership training for

final year medical students[44, 45]. The IMPACT leadership training was conducted over four

consecutive weeks, with two-and-a-half hour sessions once a week after clinical duties. To

ensure instructional training efficiency, we relied on the same three instructors across all train-

ing cohorts. Instructors either came from a leadership training background (JN) or a medical

education background (BS, JK).

We designed the training in four modules. Module one introduced leadership theory,

focusing on transactional and transformational leadership and reflection on the residents’

leadership role within their clinical team. Module two tested the participants’ leadership

behavior in one of four standardized five-minute scenarios in a simulation-based environ-

ment. The scenarios were carried out with professional actresses who were specially trained

and functioned as nurses within the role-plays. Female actresses were chosen for role-plays

representing daily practice in German hospitals where the majority of nurses is female[46].

The scenarios originated from a critical incident study[44], were randomly assigned to the

residents and were recorded on video for later evaluation. Given the positive effects of feed-

back on training transfer[26, 47], we integrated a half-hour one-on-one feedback session

between modules two and three; based on the recorded role-plays, physicians reflected their

leadership performance together with their instructor. Module three comprised practicing

communication techniques explicitly tied to transactional and transformational leadership.

Module four tested the participants in another standardized scenario within the simulation-

based environment. Actresses and scenarios were evenly distributed over module two and

module four.

At intervals of four weeks, participants of the control group filled out two online question-

naires containing the same self-assessment scales as those of the IMPACT group and received

a manuscript regarding clinical leadership. All participants of the control group received €40 in compensation after completion.

Study design and setting

This study was a single-institution repeated measurement controlled trial at a large university

hospital during the year 2015 involving several different clinics of various medical specialties.

The primary outcome was transactional and transformational leadership skill performance

as assessed by an external evaluator, the secondary outcome self-evaluated transformational

and transactional leadership skills, and leadership knowledge. In the training group (IMPACT

group), the primary and secondary outcomes were tested in a pre-post design. Measurement

took place before the training and four weeks after the first training session.

In the control group, only the secondary outcome was assessed pre-to-post-test within a

four week interval, as residents of the control group did not receive any role-play intervention.

Ethical approval for the study was obtained from the ethical committee of the Ludwig-Max-

imilians-University (LMU) Munich. Prior to the training, all residents gave their written

informed consent in study participation.

Transformational and transactional leadership training for residents

PLOS ONE | https://doi.org/10.1371/journal.pone.0183019 August 25, 2017 4 / 13

Outcome measures

For the 12-item Performance Scale (measuring the primary outcome), we designed items that

represented target behaviors for transactional and transformational clinical leadership. For

instance, target behaviors for transactional leadership were tested by statements such as ‘The

physician gave positive feedback for good performance’; transformational leadership was

assessed by items like ‘The physician talked about the goals that have been set in an encourag-

ing way.’. An external evaluator, who was blinded to the assignment and specially trained to

assess (i.e. video coding), rated the primary outcome. The evaluator rated the recorded role-

plays on a five-point Likert scale (range, 1 = strongly disagree to 5 = strongly agree). Examples

for the ratings and corresponding leadership skills (i.e. target behaviors) are presented in

Table 1. Rating quality was ensured by a ten percent inter-rating (ICC = 0.92). The 40-item Leadership Scale (assessing the secondary outcome) includes three subscales:

transactional, transformational and passive leadership, containing items of the German ver-

sion of the Multifactor Leadership Scale[48], a valid tool to evaluate the FRLM. Before and

after four weeks of training, participants stated on a five-point Likert scale (range, 1 = not at all

to 5 = frequently, if not always) the frequency with which they exerted a certain leadership

behavior.

The nine-item multiple choice (measuring the secondary outcome) knowledge test was

developed to examine leadership knowledge regarding transformational and transactional

leadership. The test yields a composite mean score of 34.

A-priori, all scales were validated in a pilot study. Residents of different fields of application

(e.g. anesthesia, surgery, radiology) were interviewed one-by-one after filling out the scales of

the subjective measurement. Based on their ratings and comments, we revised the Leadership

Scale as well as the multiple choice knowledge test thoroughly and made changes where

necessary.

Data collection

Prior to the training, participants received a random pseudonym as an identifier on data

sheets. Residents participating in the IMPACT group filled out the Leadership Scale as well as

the knowledge test on-site before and after the training, and were supervised by one of the

trainers at any time. Within an interval of four weeks, participants of the control group filled

out these two tests, as well. However, data of the control group were only obtained via online

assessment. As participants of the control group filled out the tests at home or at their work-

place, supervision could not be established. The Performance Scale was assessed by an inde-

pendent evaluator after modules two and four. The dataset underlying our results is publicly

available from the data repository Open Data LMU (DOI: https://doi.org/10.5282/ubm/data.

109).

Table 1. Example leadership skills of the performance scale by leadership component.

Leadership component Leadership skill

Transactional leadership The resident. . .

• gave positive feedback for good efforts.

• made clear what the nurse can expect when she performs well.

• clarified who is responsible for defined tasks.

Transformational

leadership

• treated the nurse respectfully.

• encouraged the nurse to engage in overall goals of the clinical team.

• formulated an appealing vision of what shall be achieved to improve

patient care.

https://doi.org/10.1371/journal.pone.0183019.t001

Transformational and transactional leadership training for residents

PLOS ONE | https://doi.org/10.1371/journal.pone.0183019 August 25, 2017 5 / 13

Statistical analysis

A sample size calculation was performed and it was found that a sample size of 60 was needed

to provide 80% power to detect medium to large effects. Data was entered into SPSS (version

23.0, SPSS Inc. Chicago, Illinois) for further analysis. We used repeated measures ANOVA to

test for (1) the progression of externally rated leadership skills and self-assessment of the lead-

ership scale as well as of the knowledge test, and (2) group between control and IMPACT

group for the leadership knowledge test and self-assessment of the leadership-scale. All analy-

ses were based on a 5% level of significance.

Results

Of 57 residents, 50 (88%) completed the training (m = 29.98 years; SD = 2.60), with 25 (50%) female participants; 40 residents (70%) performed both role-plays. Reasons for missing ses-

sions included clinical emergencies, unexpected changes in rotation schedules, clinical exami-

nation of incoming refugees at the central station, or illness. 23 residents participated in the

control-group (n = 23; m = 29.13 years; SD = 2.53), with 18 (78%) female residents. All out- come variables showed sufficient reliability, with a pre-test Cronbach’s alpha of 0.55 (knowl-

edge test), 0.64 (transactional leadership) and 0.85 (transformational leadership) and a post-

test Cronbach’s alpha of 0.70 (transactional leadership), 0.81 (knowledge test) and 0.84 (trans-

formational leadership).

Performance scale

As hypothesized, after four weeks of training, the Performance Scale increased 15% in transac-

tional leadership skill performance (2.10 to 2.86) (intervention effect, 0.76; 95% CI, 0.40 to

1.13; P< .001, eta2 = 0.31) and 14% in transformational leadership skill performance (2.26 to 2.94) (intervention effect, 0.68; 95% CI, 0.27 to 1.09; P< .001, eta2 = 0.22). Table 2 presents pre- and post-test means, standard deviations and mean changes for leadership skill perfor-

mance. Graph C illustrates this effect in Fig 1.

Leadership scale

Expectedly, we found a significant increase in self-assessed transactional and transforma-

tional leadership, as well as a significant interaction between point of measurement and group,

indicating a substantial gain in both leadership components only for the IMPACT group

(F(1,73) = 5.63, P< .02, eta2 = 0.07). The Leadership Scale revealed a 4% increase in self-

Table 2. Descriptive statistics and mean changes for leadership skill performance, divided into transactional and transformational leadership.

Performance Scale External Rater n m SD mean change (95% CI)

Transactional leadershipa

Pre-test to post-test 40 0.76 (0.40, 1.13)

Pre-test 40 2.10 0.75

Post-test 40 2.86 0.99

Transformational leadershipa

Pre-test to post-test 40 0.68 (0.27, 1.09)

Pre-test 40 2.26 0.88

Post-test 40 2.94 1.13

a Scale: 1 = strongly disagree, 2 = disagree, 3 = neutral, 4 = agree, 5 = strongly agree; m = mean; SD = standard deviaton; n = sample size; CI = confidence

interval

https://doi.org/10.1371/journal.pone.0183019.t002

Transformational and transactional leadership training for residents

PLOS ONE | https://doi.org/10.1371/journal.pone.0183019 August 25, 2017 6 / 13

assessed transactional skills (3.83 to 4.03) (intervention effect, 0.20; 95% CI, 0.08 to 0.33; P< .001, eta2 = 0.18) and a 6% increase in transformational leadership skills (3.54 to 3.86) (inter-

vention effect, 0.31; 95% CI, 0.02 to 0.40; P< .001, eta2 = 0.53). The reported effects are consid- ered to be excellent[49]. Mean scores for passive leadership were low at baseline, both in the

IMPACT group (m = 1.87, SD = 0.55) and the control group (m = 1.94, SD = 0.51). Graph A and B illustrate this effect in Fig 1.

Knowledge test

Two cases had to be eliminated from further analysis as data sets were not complete. Groups

differed significantly in a-priori (F(1,69) = 17.17, P< .001, eta2 = 0.20) and post-test leadership knowledge (F(1,69) = 15.51, P< .001, eta2 = 0.19). Testing for group differences for the leader- ship knowledge test resulted in a significant effect for group (F(1,69) = 22.26, P< .001, eta2 = 0.25), yet there was no significant increase in leadership knowledge in any of the groups. In

Table 3, pre- to post-test findings of transactional, transformational leadership, as well as lead-

ership knowledge are summarized.

Correlations between the scales

In line with prior studies[25, 50], the two subscales of the leadership scale, transactional and

transformational leadership, correlated significantly (r = .60; P< .05). Accordingly, passive leadership showed negative, significant correlations to transformational (r = -.49; P< .001) and transactional leadership (r = -.52; P< .001).

Fig 1. Overview of the effect for the self assessed leadership scale (graph A and B) and the performance

scale (graph C).

https://doi.org/10.1371/journal.pone.0183019.g001

Transformational and transactional leadership training for residents

PLOS ONE | https://doi.org/10.1371/journal.pone.0183019 August 25, 2017 7 / 13

Discussion

In this study, the four-week IMPACT leadership training was designed to increase transforma-

tional and transactional leadership skills in residents of various specialties. Transformational

leadership refers to the leader’s ability to motivate team members to commit to a common

goal. Transactional leadership involves the practice of exchanging benefits for excellent perfor-

mance. Taking into account that a comprehensive set of leadership skills is necessary to meet

the complex demands of daily clinical practice[5], we based our training on the Full Range

Leadership Model[22].

Previous studies on leadership trainings have mainly been conducted in the organizational

setting, primarily focusing on the transformational leadership component. The study by Abrell

et al.[26] can be considered as one of the most comprehensive effort to train transformational

leadership in a long-term study design. In their study, transformational leadership was as-

sessed by subordinates and leadership performance was rated by leaders’ supervisors, showing

a significant improvement over time[26]. Abrell et al.[26] incorporated feedback mechanisms

into their curriculum as well as theoretical sessions giving an in-depth review of different

transformational leadership styles, such as ‘individual consideration’ or ‘inspirational motiva-

tion’. To the best of our knowledge, there is no program training transformational and transac-

tional leadership alike, neither in the organizational nor in the medical field of application.

Existing leadership programs in graduate medical education, such as the one by Awad et al.

[3], focus on broader communication skills. Awad et al.[3] implemented a leadership training

for surgical residents over the course of 6 months. They aimed at improving collaborative lead-

ership through fostering a communication style that is regarded less commanding. Before and

after completion of the training, residents assessed self-perceived alignment of the team, com-

munication and integrity. The authors were able to demonstrate significant increases in these

Table 3. Improvement of transactional and transformational leadership skills in the IMPACT group and changes in leadership knowledge, as com-

pared to a control group.

IMPACT Group Control Group

Scale mean change

(95% CI)

m SD mean change

(95% CI)

m SD

Transactional leadershipa

Pre-test to post-testc 0.20 (0.08, 0.33) 0.07 (0.10, 0.23)

Pre-test 3.83 0.48 3.79 0.47

Post-test 4.03 0.45 3.86 0.53

Transformational leadershipa

Pre-test to post-testc 0.31 (.023, 0.40) 0.83 (0.08, 0.25)

Pre-test 3.54 0.38 3.53 0.48

Post-test 3.86 0.31 3.61 0.40

Knowledge testb

Pre-test to post-testd 0.6 (-1.17, 1.29) -1.09 (-2.96, 0.77)

Pre-test 26.48 2.67 22.95 4.32

Post-test 26.54 4.19 21.86 5.28

a Scale: 1 = not at all, 2 = once in a while, 3 = sometimes, 4 = fairly often, 5 = frequently.

b Scale: Multiple choice format, maximum achievable score = 34

c Significance level: p < .01 d Significance level: p>.30 m = Mean, SD = Standard Deviaton, CI = confidence interval.

https://doi.org/10.1371/journal.pone.0183019.t003

Transformational and transactional leadership training for residents

PLOS ONE | https://doi.org/10.1371/journal.pone.0183019 August 25, 2017 8 / 13

areas; however, training effects in terms of leadership performance such as improved team

interactions have not been evaluated.

Our results go beyond prior research in different ways: First, unlike in the study by Abrell

et al.[26], we not only tested for transformational leadership skills but also for transactional

leadership skills in a pre-post design. Second, for the first time, the two leadership components

were trained in a group of residents, extending external validity of the proposed leadership

model to the medical education area of application. Overall, our results indicate that the

FRLM is well suited for empirically testing leadership skills in residents of a wide spectrum.

Third, we built upon first attempts to test the model in the medical context[42, 43] by provid-

ing a targeted, multimethod, structured training curriculum to improve resident leadership.

Fourth, the different evaluation data modalities we applied (self-assessment, evaluation of per-

formance) expand existing studies that have not evaluated the behavioral component of

leadership.

We provided evidence that both distinct leadership components laid forth in the model are

applicable for displaying significant increases in residents’ leadership performance. For exam-

ple, at the end of the training, residents were able to show appreciation for good efforts (trans- formational leadership skills) and make clear who is responsible for specific tasks (transactional leadership skills). Interestingly, residents scored higher in self-assessed transactional leadership at baseline than in transformational leadership. They did change significantly in both leader-

ship components, yet remained higher mean scores for transactional leadership also after

training was completed. We believe this reflects the unique requirements of the clinical setting

where fostering and sustaining patient safety is among the highest of priorities. In their every-

day clinical practice, residents might feel more obliged to intervene and exert active control in

order to prevent medical errors, thus exhibiting more transactional leadership behaviors.

Our results further suggest that four weeks of training seem to be a good starting point to

effectively train leadership skills in residents across a wide range of specialties.

A control group did not increase in self-assessed leadership skills. It is remarkable that a

substantial gain in both leadership components was demonstrated by video coding of simula-

tions from an external evaluator perspective and by subjective data, as well. The increase in

leadership skills from two different, independent perspectives supports the applicability of the

leadership model for graduate medical education.

We controlled for a possible confounding effect of passive leadership at baseline, as this

most ineffective leadership component is considered to attenuate the effect of transformational

leadership on safety[38]. Consistent with previous studies[33, 42], mean scores for passive

leadership were low in both groups.

Limitations

Randomization and blinding between groups were not complete, as the participants were

aware of which group they would be assigned to when applying; a selection bias can thus not

be ruled out. However, this is somewhat ameliorated by the low and not differing scores

between IMPACT group and control group as regards the leadership scale. For the knowledge

test, there was no significant change in the two groups. Item difficulty for the pre-test measure

might have been too high, resulting in low values of cronbach’s alpha at pre-test. In contrast,

high values for cronbach’s alpha at post-test might be explained by increasing knowledge

through the training. Apart from that, high standard deviations for mean scores for pre- and

post-test might reflect a high range in leadership knowledge among the participants. A rede-

sign of our knowledge test in terms of discriminatory power analysis might be beneficial in

order to detect significant changes in residents’ leadership knowledge. In addition, both

Transformational and transactional leadership training for residents

PLOS ONE | https://doi.org/10.1371/journal.pone.0183019 August 25, 2017 9 / 13

groups differed significantly in leadership knowledge at baseline and after the training. We

traced this finding back to the different data acquisition modalities (on-site assessment versus

unsupervised online assessment) and different levels of motivation. Unsupervised data collec-

tion also poses the problem of participants potentially consulting textbooks or the internet in

order to gain better test results. For preventing potential confounding effects generated by

data collection modalities, we suggest that future research should incorporate on-site data

acquisition also for the control group. Passive leadership was only assessed prior to the training

by the participants themselves. Since self-assessment is prone to self-serving biases, future

studies should examine passive leadership also from an external perspective. As this study was

a single-institution controlled trial, the degree of transferability could be enhanced by recruit-

ing residents from different institutions. It must be noted, however, that our participants came

from a wide range of specialties, and the proportion of female to male residents was balanced,

indicating that the generalizability of our findings is not completely limited. Overall, our study

would have benefited from a larger sample size, especially in the control group. However, the

effect sizes for the group comparisons were extraordinarily high[49], raising questions as to

the benefit of larger sample sizes.

Conclusions

Our study is the first to establish and design a training format for the clinical setting based on

the transactional and transformational leadership approach, going beyond previous research

in a number of ways. First, we illustrated the feasibility of the proposed leadership framework

for the clinical environment by providing evidence for support of this model. Second, we

included a number of strong design elements, such as the use of a control group, and an out-

comes assessment based on the performance rating by an external evaluator, as opposed to

self-assessment. Third, in the following we provide best practice strategies for leadership pro-

grams specifically tailored for residents: As it is known that transactional and transformational

leadership can have both a protective impact on patient safety[10, 11] and a positive effect on

teamwork[7–9], medical institutions should establish resident leadership training drawn to the

transactional and transformational framework. To ensure behavioral change, curricula should

embed simulation-based practices, addressing transactional and transformational behaviors

alike. Programs including one-on-one feedback can guide the way to individual high-quality

leadership performance. Future studies should examine the implementation of the behavioral

changes in daily clinical work, potentially incorporating leadership training as a starting point

for mentoring programs within specialties[51]. We recommend building upon the target

behaviors which we tied to transactional and transformational leadership skills in the perfor-

mance assessment. We suggest to evaluate the stability of the increased leadership performance

by conducting a time-latent post-test of the behavioral component in further studies. In order

to expand upon the promising findings of our study, a research network for clinical leadership

could prove beneficial for researchers to catalyze the design and evaluation of programs.

Acknowledgments

We are grateful to the following individuals, who participated and were compensated for par-

ticipation in this study:

Research assistants: Sabrina Reif (Ludwig-Maximilians-Universität Munich)

Actresses: Eva-Maria Kerp, Tamara Mayer (Klinikum der Universität)

Independent evaluators: Christian Strobel (Ludwig-Maximilians-Universität Munich),

Hannah Freienstein (Ludwig-Maximilians-Universität Munich)

Transformational and transactional leadership training for residents

PLOS ONE | https://doi.org/10.1371/journal.pone.0183019 August 25, 2017 10 / 13

We thank all resident physicians who participated in the study. Participants of the control

group receives honoraria for their contributions.

Author Contributions

Conceptualization: Barbara Saravo, Janine Netzel, Jan Kiesewetter.

Data curation: Barbara Saravo, Janine Netzel, Jan Kiesewetter.

Formal analysis: Barbara Saravo, Jan Kiesewetter.

Funding acquisition: Barbara Saravo, Jan Kiesewetter.

Investigation: Barbara Saravo, Janine Netzel, Jan Kiesewetter.

Methodology: Barbara Saravo, Janine Netzel, Jan Kiesewetter.

Project administration: Barbara Saravo.

Supervision: Barbara Saravo, Jan Kiesewetter.

Validation: Barbara Saravo, Janine Netzel, Jan Kiesewetter.

Writing – original draft: Barbara Saravo, Janine Netzel, Jan Kiesewetter.

Writing – review & editing: Barbara Saravo, Janine Netzel, Jan Kiesewetter.

References

1. Angood P. Cultivating Physician Leadership Is Necessary for Lasting and Meaningful Change. Aca-

demic Medicine. 2015; 90(10):1290. https://doi.org/10.1097/acm.0000000000000880 00001888-

201510000-00002. PMID: 27002874

2. Jardine D, Correa R, Schultz H, Nobis A, Lanser BJ, Ahmad I, et al. The Need for a Leadership Curricu-

lum for Residents. Journal of graduate medical education. 2015; 7(2):307–9. https://doi.org/10.4300/

JGME-07-02-31 PMID: 26221472; PubMed Central PMCID: PMCPMC4512827.

3. Awad SS, Hayley B, Fagan SP, Berger DH, Brunicardi FC. The impact of a novel resident leadership

training curriculum. American journal of surgery. 2004; 188(5):481–4. Epub 2004/11/18. https://doi.org/

10.1016/j.amjsurg.2004.07.024 PMID: 15546554.

4. Collins-Nakai R. Leadership in medicine. Mcgill J Med. 2006; 9(1):68–73. PMID: 19529813; PubMed

Central PMCID: PMCPMC2687901.

5. Blumenthal DM, Bernard K, Bohnen J, Bohmer R. Addressing the leadership gap in medicine: Resi-

dents’ need for systematic leadership development training. Academic Medicine. 2012; 87(4):513–22.

https://doi.org/10.1097/ACM.0b013e31824a0c47 2012-13134-016. PMID: 22361800

6. Dhaliwal G, Sehgal NL. Demystify Leadership in Order to Cultivate It. Academic Medicine. 2014; 89

(11):1441. https://doi.org/10.1097/ACM.0000000000000489 00001888-201411000-00015. PMID:

25350336

7. Curry LA, Spatz E, Cherlin E, Thompson JW, Berg D, Ting HH, et al. What Distinguishes Top-Perform-

ing Hospitals in Acute Myocardial Infarction Mortality Rates? Annals of Internal Medicine. 2011; 154

(6):384-w.130. PMID: 59521266.

8. Nembhard IM, Edmondson AC. Making it safe: The effects of leader inclusiveness and professional sta-

tus on psychological safety and improvement efforts in health care teams. Journal of Organizational

Behavior. 2006; 27(7):941–66. https://doi.org/10.1002/job.413 2006-20524-008.

9. Stockwell DC, Slonim AD, Pollack MM. Physician team management affects goal achievement in the

intensive care unit. Pediatr Crit Care Med. 2007; 8(6):540–5. https://doi.org/10.1097/01.PCC.

0000288709.15113.8A PMID: 17906596.

10. Baker D, Gustafson S, Beaubien J, Salas E, Barach P. Medical teamwork and patient safety: the evi-

dence-based relation. Rockville, MD: Agency for Healthcare Research and Quality; 2005.

11. Sutcliffe KM, Lewton E, Rosenthal MM. Communication failures: an insidious contributor to medical mis-

haps. Academic medicine: journal of the Association of American Medical Colleges. 2004; 79(2):186–

94. PMID: 14744724.

Transformational and transactional leadership training for residents

PLOS ONE | https://doi.org/10.1371/journal.pone.0183019 August 25, 2017 11 / 13

12. Gordon LJ, Rees CE, Ker JS, Cleland J. Dimensions, discourses and differences: trainees conceptual-

ising health care leadership and followership. Medical education. 2015; 49(12):1248–62. https://doi.org/

10.1111/medu.12832 PMID: 26611190

13. Gabel S. Expanding the Scope of Leadership Training in Medicine. Academic Medicine. 2014; 89(6).

14. Yukl G. Leadership in Organizations. Edinburgh: Pearson; 2013.

15. Webb AMB, Tsipis NE, McClellan TR, McNeil MJ, Xu M, Doty JP, et al. A First Step Toward Under-

standing Best Practices in Leadership Training in Undergraduate Medical Education: A Systematic

Review. Academic Medicine. 2014; 89(11).

16. Kiesewetter J, Schmidt-Huber M, Netzel J, Krohn AC, Angstwurm M, Fischer MR. Training of leadership

skills in medical education. GMS Zeitschrift fur medizinische Ausbildung. 2013; 30(4):Doc49. Epub

2013/11/28. PMID: 24282452; PubMed Central PMCID: PMC3839077.

17. Straus SE, Soobiah C, Levinson W. The impact of leadership training programs on physicians in aca-

demic medical centers: A systematic review. Academic Medicine. 2013; 88(5):710–23. https://doi.org/

10.1097/ACM.0b013e31828af493 2013-19889-015. PMID: 23524921

18. Lee MT, Tse AM, Naguwa GS. Building leadership skills in paediatric residents. Medical education.

2004; 38(5):559–60. Epub 2004/04/27. https://doi.org/10.1111/j.1365-2929.2004.01867.x PMID:

15107107.

19. Johnson J, Stern T. Teaching Residents About Emotional Intelligence and Its Impact on Leadership.

Acad Psychiatry. 2014:1–4. https://doi.org/10.1007/s40596-014-0048-4 PMID: 24493362

20. Kuo AK, Thyne SM, Chen HC, West DC, Kamei RK. An innovative residency program designed to

develop leaders to improve the health of children. Academic Medicine. 2010; 85(10):1603–8. https://

doi.org/10.1097/ACM.0b013e3181eb60f6 2010-20918-004. PMID: 20703151

21. Dickey C, Dismukes R, Topor D. Creating Opportunities for Organizational Leadership (COOL): Creat-

ing a Culture and Curriculum that Fosters Psychiatric Leadership Development and Quality Improve-

ment. Acad Psychiatry. 2014:1–5. https://doi.org/10.1007/s40596-014-0082-2 PMID: 24664604

22. Bass BM, Avolio BJ. Transformational leadership: A response to critiques. In: Chemers MM, Ayman R,

editors. Leadership theory and research: Perspectives and directions. Sydney, Australia: Academic

Press Inc.; 1993. p. 81–107.

23. Lowe KB, Galen Kroeck K. Effectiveness correlates of transformational and transactional leadership: A

meta-analytic. Leadership Quarterly. 1996; 7(3):385. PMID: 9705070871.

24. Franke F, Felfe J. How does transformational leadership impact employees’ psychologicalstrain?

Examining differentiated effects and the moderating role of affective organizational commitment. Lead-

ership 2011; 7(3):295–316.

25. Geyer A, Steyrer J. Messung und Erfolgswirksamkeit transformationaler Führung. Zeitschrift für Perso-

nalforschung. 1998; 12(4):377–401.

26. Abrell C, Rowold J, Weibler J, Moenninghoff M. Evaluation of a Long-term Transformational Leader-

ship. Development Program. 2011; 25(3):205–24.

27. Bass BM, Avolio BJ, Jung DI, Berson Y. Predicting unit performance by assessing transformational and

transactional leadership. The Journal of applied psychology. 2003; 88(2):207–18. Epub 2003/05/07.

PMID: 12731705.

28. Prenkert F, Ehnfors M. A measure of organizational effectiveness in nursing management in relation to

transactional and transformational leadership: a study in a Swedish county hospital. Journal of nursing

management. 1997; 5(5):279–87. Epub 1997/12/31. PMID: 9348842.

29. Wang X, Chontawan R, Nantsupawat R. Transformational leadership: effect on the job satisfaction of

Registered Nurses in a hospital in China. Journal of advanced nursing. 2012; 68(2):444–51. Epub

2011/07/21. https://doi.org/10.1111/j.1365-2648.2011.05762.x PMID: 21771039.

30. Al-Mailam FF. Transactional versus transformational style of leadership—employee perception of lead-

ership efficacy in public and private hospitals in Kuwait. Quality management in health care. 2004; 13

(4):278–84. PMID: 15532520.

31. Hammer A, Ommen O, Rottger J, Pfaff H. The relationship between transformational leadership and

social capital in hospitals-a survey of medical directors of all German hospitals. Journal of public health

management and practice: JPHMP. 2012; 18(2):175–80. Epub 2012/01/31. https://doi.org/10.1097/

PHH.0b013e31823dea94 PMID: 22286287.

32. Spinelli RJ. The applicability of Bass’s model of transformational, transactional, and laissez-faire leader-

ship in the hospital administrative environment. Hospital topics. 2006; 84(2):11–8. Epub 2006/05/20.

https://doi.org/10.3200/HTPS.84.2.11-19 PMID: 16708688.

33. Bass BM, Avolio BJ. MLQ Multifactor Leadership Questionnaire: Technical report. Redwood City, CA:

Mind Garden; 1995.

Transformational and transactional leadership training for residents

PLOS ONE | https://doi.org/10.1371/journal.pone.0183019 August 25, 2017 12 / 13

34. Bass BM. Theory of transformational leadership redux. The Leadership Quarterly. 1995; 6(4):463–78.

https://doi.org/https://doi.org/10.1016/1048-9843(95)90021-7

35. Bass BM. The multifactor leadership questionnaire: Form 5. Bringhampton: State University of New

York; 1985.

36. Avolio BJ, Bass BM. Full-range of leadership development. Binghampton, NY: Bass, Avolio & Associ-

ates; 1991.

37. Bass BM. The Bass handbook of Leadership: Theory, Research & Managerial Applications. 4 ed. New

York: Free Press; 2008.

38. Mullen J, Kelloway EK, Teed M. Inconsistent style of leadership as a predictor of safety behaviour.

Work & Stress. 2011; 25(1):41–54. https://doi.org/10.1080/02678373.2011.569200 2011-08084-003.

39. Vandenberghe C, Stordeur S, D’Hoore W. Transactional and transformational leadership in nursing:

Structural validity and substantive relationships. European Journal of Psychological Assessment. 2002;

18(1):16–29. https://doi.org/10.1027//1015-5759.18.1.16 2002-00603-003.

40. Fuller JB, Patterson CEP. A quantitative review of research on charismatic leadership. Psychological

Reports. 1996; 78(1):271. PMID: 9603262961.

41. Judge TA, Piccolo RF. Transformational and transactional leadership: a meta-analytic test of their rela-

tive validity. The Journal of applied psychology. 2004; 89(5):755–68. Epub 2004/10/28. https://doi.org/

10.1037/0021-9010.89.5.755 PMID: 15506858.

42. Horwitz IB, Horwitz SK, Daram P, Brandt ML, Charles Brunicardi F, Awad SS. Transformational, Trans-

actional, and Passive-Avoidant Leadership Characteristics of a Surgical Resident Cohort: Analysis

Using the Multifactor Leadership Questionnaire and Implications for Improving Surgical Education Cur-

riculums. Journal of Surgical Research. 2008; 148(1):49–59. https://doi.org/10.1016/j.jss.2008.03.007

PMID: 18570931.

43. Schmidt-Huber M, Hörner K, Weisweiler S. Wirksames Führungsverhalten von Oberärzten und pfleger-

ischen Stationsleitungen unter der Lupe: Führen Pflegekräfte anders als Ärzte? ZFPG. 2015; 1(3): 28–

43. https://doi.org/10.17193/HNU.ZFPG 01.03.2015–07

44. Netzel J. Power and perspective taking–Analyzing factors of hierarchical and interdisciplinary coopera-

tion. Unpublished doctoral thesis. Ludwig-Maximilians-Universität München.2015.

45. Kiesewetter J, Schmidt-Huber M, Netzel J, Angstwurm M. Führungskompetenzen: Früh übt sich, wer

eine effektive und wertschätzende Zusammenarbeitskultur fördern möchte. Dtsch Arztebl. 2012; 109

(51–52):2613–4.

46. Bf Arbeit. Arbeitsmarktberichterstattung: Gesundheits- und Pflegeberufe in Deutschland. Nürnberg:

2011.

47. Barling J, Weber T, Kelloway EK. Effects of transformational leadership training on attitudinal and finan-

cial outcomes: A field experiment. Journal of Applied Psychology. 1996; 81(6):827–32. https://doi.org/

10.1037/0021-9010.81.6.827

48. Felfe J, Goihl K. Deutsche überarbeitete und ergänzte Version des “Multifactor Leadership Question-

naire” (MLQ). In: Glöckner-Rist A, editor. ZUMA-Informationsssystem Elektronisches Handbuch sozial-

wissenschaftlicher Erhebungsinstrumente Version 500. Mannheim: Zentrum für Umfragen, Methoden

und Analysen; 2002.

49. Cohen J. Statistical power analysis for the behavioral science. New York: Hillsdale; 1988.

50. Hater JJ, Bass BM. Superior’s Evaluations and Subordinates’ Perceptions of Transformational and

Transactional Leadership. Journal of Applied Psychology. 1988; 73(4):695–702. PubMed PMID:

5111775.

51. Baird DS, Soldanska M, Anderson B, Miller JJ. Current leadership training in dermatology residency

programs: a survey. J Am Acad Dermatol. 2012; 66(4):622–5. https://doi.org/10.1016/j.jaad.2009.10.

009 PMID: 22285671.

Transformational and transactional leadership training for residents

PLOS ONE | https://doi.org/10.1371/journal.pone.0183019 August 25, 2017 13 / 13

Copyright of PLoS ONE is the property of Public Library of Science and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder’s express written permission. However, users may print, download, or email articles for individual use.