Since the initial IOM report, public health practice and academic organizations have focused increasingly on public health leadership development. These efforts include the development of state, regional, national, and international public health leadership institutes, the formation of a national public health leadership development network, and the development of public health leadership competency frameworks for both educational and practice settings. These efforts shape the development of public health leaders now and into the future.

To assist in the shaping of future public health leaders, the Association of Schools and Programs of Public Health (ASPPH) has developed core competencies for individuals obtaining Master of Public Health (MPH) (12) or Doctor of Public Health (DrPH) (13) degrees. Both models delineate specific competencies necessary for public health students to evince in order to achieve their leadership potential. Further, both models define leadership in terms of creating a shared vision, combined with notions of motivating others, galvanizing organizational and community resources to address public health problems, and utilizing the best strategies and practices to enhance service and solve problems. Finally, both models highlight the potential of public health graduates to lead organizations and communities, with the competence to influence others, establish a shared vision, and accomplish the mission of public health. Further, leadership competencies are not the purview of ASPPH alone. The Core Competencies for Public Health Professionals (14) include leadership and systems thinking skills. Clearly, leadership competence is on the agenda for current and future public health workforce efforts.

The overarching theme of public health leadership literature and public health leadership development efforts is the need for highly skilled and well-educated leaders capable of galvanizing organizations and communities in transformational change processes that not only ensure, but also seek to improve the health and well-being of the public. These efforts frame a vision for public health leadership that currently prefers transformational, change-agent leaders. However, as Nicola (15) has pointed out, classic management functions – planning, organizing, leading, and controlling – remain vital to assuring the performance of public health organizations. While transformational leadership qualities enable public health leaders to engage communities in efforts to improve population health, the full range of leadership qualities, including technical and managerial acumen, is necessary not only to lead change but also to effectively attend to general and regular organizational tasks and responsibilities should not be overlooked.

To that end, leadership may have discipline-specific requirements unique to public health. Rather than assume that leadership qualities, characteristics, and processes are universal to all professions, public health agencies would benefit from a better understanding of the skills and competencies required for successful public health leadership.

The purpose of this study was to examine the full range of leadership styles among local health department directors in Kentucky. Specifically, this portion of the study quantitatively explores: (1) the leadership styles of local health department directors and their perceptions of organizational outcomes, (2) the sub-components of each style contributing to the overall leadership style of local health department directors, and (3) whether there is any relationship between leadership styles and specific county health outcomes. We posited that while leadership styles would vary, the predominant leadership style, and the one most closely correlated with positive leadership outcomes, would be the transformational leadership style.

Transformational and transactional leadership are both related to the success of the group. Success, or outcomes of leadership, was measured with the MLQ through leaders’ self-reported skills at motivation, effectiveness in interacting at different levels of the organization, and perceived employee satisfaction with leaders’ methods of working with others. These include: extra effort (EE), effectiveness (EFF), and satisfaction with the leadership (SAT). EE may be defined as the extent to which leaders get others to do more than they expected to do, heighten others’ desire to succeed, and increase others’ willingness to try harder. Effective leaders are effective in meeting others’ job-related needs, in representing their group to higher authority, and in meeting organizational requirements. They lead groups that are effective. Leadership satisfaction includes: using methods of leadership that are satisfying and working with others in a satisfactory way.

The extent of transformational and transactional leadership styles and components of leadership styles reported by study participants was explored with descriptive statistics (e.g., means and SDs). Pearson bivariate correlation coefficients were calculated to measure relationships between variables of interest. Given the limited sample size, power analyses, as well as additional or more complex analyses, were not feasible. Finally, findings from the descriptive analysis were examined for any relationship to existing county-level data: Beale Codes, which measure relative population density on a rural-urban continuum, and the County Health Rankings (26), which rank counties based on health outcomes (mortality and morbidity) and health factors (health behaviors, clinical care, physical environment, and social and economic factors).

This study was approved by the University of Kentucky Institutional Review Board.

Thirteen local health directors completed the initial survey phase of the study. Complete demographics are provided in Table 2 below. Nearly two-thirds (62%) of participants were female and the majority (n = 12, 92%) were white. Participants were of varying ages. All participants had at least a bachelor’s degree. The majority (n = 11, 85%) classified their health departments as rural and the others (n = 2, 15%) classified their health departments as sub-urban.

Personal leadership development was at least a moderate priority for participants. About a third of participants (n = 4, 31%) indicated that their own leadership development was a moderate priority. Just over two-thirds (n = 9, 69%) felt it was a high priority. Similarly, developing the leadership abilities of staff was also at least a moderate priority for participants. The results were identical. About a third of participants (n = 4, 31%) indicated that their own leadership development was a moderate priority. Just over two-thirds (n = 9, 69%) felt it was a high priority. Finally, participants felt their leadership at least had considerable influence on the performance of the organization. Just over two-thirds of participants (n = 9, 69%) believed their leadership had considerable influence on the performance of the organization; while about a third (n = 4, 31%) indicated that their own leadership had significant influence on organizational performance.

Table 3 provides a reference to the reader on the major leadership styles and their sub-style components.

Participants self-reported a wide range of leadership styles, with wide variance across all leadership characteristics.

Consistent with the range of self-reported leadership styles, participants also reported variance in leadership outcomes, especially for EFF. EE had a mean percentile of 62.7 (SD = 18.34), with a minimum percentile of 26 and a maximum percentile of 88. EFF had a mean percentile of 48.9 (SD = 20.73), with a minimum percentile of 20 and a maximum of 88. Finally, satisfaction had a mean percentile of 34.4 (SD = 29.30), with a minimum percentile of 1 and a maximum of 95. Finally, overall work outcome had a mean percentile of 34.4 (SD = 29.30), with a minimum percentile of 1 and a maximum of 95.

As shown in Table 8, neither participants’ leadership styles nor individual leadership style components were correlated with County Health Rankings (health outcomes and health factors). However, significant moderate correlations were found between county Beale code classification and both health outcomes (Pearson r = 0.401, p < 0.01) and health factors (Pearson r = 0.313, p < 0.05). Further, health outcomes were strongly and significantly correlated with health factors (Pearson r = 0.683, p < 0.01). Finally, when calculated across the counties they serve, a strong, significant, and negative correlation (Pearson r = .-0.608, p < 0.001) was found between participants’ OTF leadership and overall passive-avoidant leadership styles.

While fuller implications of both quantitative and qualitative phases of this study will be discussed in the adjoining companion paper, this quantitative portion of the study has important implications for public health practitioners. The first of these is the potential for public health leaders, including local health department directors and local boards of health, as well as public health educators, to adopt well-grounded and more holistic instrumentation such as the MLQ in their assessment of leadership styles and abilities. Such tools provide and, if used generatively, help develop greater self-awareness related to leadership styles and behaviors. Many leadership style and personality instruments have been developed, including but certainly not limited to the MLQ instrument used in this study. These instruments are available to measure a range of skills from basic management to levels of emotional intelligence. Developing a detailed leadership assessment for public health practitioners may be useful if based on empirically validated instruments, including a self-generating interpretive report which could be made readily available to interested health departments or individuals. These tools may be especially useful when orienting new supervisors or as a part of accredited higher education programs.

Assessments such as the MLQ may be helpful to state health officials and for local boards of health that may be involved in the selection and development of new local public health directors. Many members of local health boards are not trained in public health and may be appointed to fill codified positions on the local board; yet, they are tasked with hiring and supervising local directors, which may have a significant positive or negative impact on the public health agency. Boards of health should not only account for the health needs of their communities, but should also consider carefully the qualities and aspects of leaders they desire in their health director. While basic comprehension of public health services and systems may be fundamental as a qualification, if board members want a director to merely manage the books and tend to the basic tasks of clinical and environment public health, or if the health department may not have a fiscal or political environment conducive to innovation and change, they should look for a more transactional leader. If, however, the board is interested in a director who will challenge the status quo by finding creative and innovative ways to improve population health, they should focus their candidate search on more transformational leaders. Providing board members with simple measurement tools to identify these leadership styles should be helpful in this process.

A larger, more comprehensive, and randomized study utilizing 360-degree reviews by supervisors, colleagues, and subordinates could improve the measurement of public health leaders and the understanding of the qualities and characteristics that may contribute to effective public health practice. Such measures of leader attributes and skills as used in this study could be combined with more discrete measures of organizational performance to further illuminate the effect of leader style on outcomes. Further, thousands of public health leaders have been trained in state, territorial, regional, and national public health leadership institutes under the reasonable assumption that such training and development efforts improve the practice of public health. The Centers for Disease Control and Prevention (CDC), the Health Research Services Administration (HRSA), and others have invested tens of millions of dollars in these activities over the past two decades. As funding becomes increasingly limited and risks being diverted from leadership development activities, measures should be developed to objectively evaluate such activities. These measures should examine improvements in organizational financial performance, staff turnover, employee satisfaction, or other common and consistent metrics. Research into these topics should contribute substantively to the pillars of workforce research and development: enumeration, competency, and capacity.

Quantitative limitations include a small sample size and the use of self-reported data. Since responses were identifiable to the researcher and since an interpretive report was to be provided, it is plausible that participants may have been motivated to over-report what they felt were positive characteristics while under-reporting what they believed to be negative characteristics of leaders. Further, the sample of participants was randomly selected from a pre-identified pool of health department directors. It is possible that since most of the participants were white, female, and identified themselves as leading rural health departments, that these demographic factors have a significant influence on the styles of leadership practiced, preferred, or perceived to contribute to effective leadership outcomes. We did not examine how our sample compared to all US local health department directors. However, by focusing on non-urban directors and randomizing the selection, we believe our sample to be fairly representative. Sampling bias is a concern for this study, but one we accepted given the mixed-method nature of the overall study. The purpose of this study was not only to measure leadership styles, as described in this study, but also to purposefully examine transformational and transactional leaders in context. We note that caution should be taken when interpreting findings since while statistical significance was found for many measures, the small sample size and quasi-random sampling limit the generalizability of findings.