Assessment of the Association of Leadership Behaviors of Supervising Physicians With Personal-Organizational Values Alignment Among Staff Physicians

Key Points Question Is physicians’ perception of alignment between personal and organizational values associated with their evaluation of the leadership behaviors of their immediate supervisors? Findings In this survey study of 1285 physicians and physician leaders, each 1-point increase in the aggregate leadership behavior score of physicians’ immediate supervisor was associated with a 0.65-point increase in the personal-organizational values alignment score for the physicians in their work unit. The aggregate leader behavior score of each leader as assessed by all physicians they lead explained 21.6% of the variation in personal-organizational values alignment scores between work units. Meaning The leadership behavior of each physician’s immediate supervisor was associated with the variation in physicians’ perception of values alignment with their organization overall.


Introduction
Physicians increasingly are employed by large organizations. 1 This arrangement works best when physicians and the organizations in which they work have a shared sense of mission and purpose. [2][3][4] It would be intuitive to infer that such alignment of values would come naturally. Fundamentally, health care organizations should exist to provide high-quality medical care for the patients in their community, which is a mission that aligns well with the professional values of physicians.
Unfortunately, experience suggests such alignment does not come easily. [5][6][7][8][9][10][11] Physicians often feel that their organizations view them as a commodity or unit of production. 5,6,12 They frequently project that administrative leaders care more about payer mix, profitable service lines, and net operating income than providing compassionate, expert, and individualized care for the patients served. 8,12 This perception is promulgated by organizational reliance on financial incentives that focus on relative value unit generation (or throughput) as a strategy to increase physician productivity. 13 At academic medical centers, many physicians believe that their organization values research activity more than clinical care. 7, 14,15 Collectively, these factors often result in physicians feeling as if their institution views them as a replaceable part rather than a professional. 5,6 Although such misalignment may stem from fundamental differences in purpose, strategy, and goals, evidence suggests that this is an oversimplification. 16 Ultimately, the perceived misalignment of values between physicians and their organization breeds distrust, division, and reciprocal scapegoating (physicians blame administrators, administrators blame physicians), which undermine the ability of the organization to achieve its mission. 16 Although perceived misalignment has been well chronicled, the array of factors that contribute to it are poorly understood. Perhaps more importantly, there is a need to identify specific, targetable interventions that senior leaders and health care organizations can focus on to improve perceived alignment. In the present study, we (1)

Participants and Survey Measures
As previously reported, 17 Stanford Medicine conducted a survey of its clinical faculty and affiliated physicians in the spring of 2019 to help guide organizational efforts to improve professional fulfillment. The current analysis focused on the 1924 faculty physicians invited to participate in the online survey between April 1, 2019, and May 13, 2019. Medical staff faculty, hereafter referred to as physicians, are overwhelmingly physicians but also include a limited number of other PhD-level clinicians (eg, psychologists). Participation in the survey was voluntary. The response rate was determined by the American Association for Public Opinion Research (AAPOR) reporting guideline for studies of internet surveys of named persons. 18 This study was reviewed by the Stanford University institutional review board and deemed exempt because it involved retrospective analysis of administratively collected data using a completely anonymized data set.
The survey included questions exploring basic demographic characteristics (age, gender) as well as professional characteristics (faculty track, academic rank, percentage work effort dedicated to clinical work, and mean weekly work hours). Gender concordance between a physician and their leader was also determined and recorded for each physician (concordant: male leader and male physician or female leader and female physician; not concordant: male leader and female physician or female leader and male physician).

Evaluation of Professional Fulfillment and Burnout
The Stanford Professional Fulfillment Index is a well-validated tool to assess professional fulfillment and burnout that is used by organizations across the US and the world. 19,20 The professional fulfillment domain includes 6 items, whereas the burnout domain includes 10 items (4 items assessing work exhaustion and 6 items assessing interpersonal disengagement). All items are scored on a 5-point Likert scale with options ranging from not at all to extremely for the burnout items and not at all true to completely true for items about professional fulfillment. Aggregate burnout and professional fulfillment scores were determined using the published approach. 19

Leadership Behaviors
Each participant was asked to evaluate the leadership behavior of their immediate supervisor by name. Participants began by selecting the name of their supervisor from a dropdown menu that listed the names of their department chair, division chief, and medical directors. These individuals were then evaluated using the Mayo Clinic Participatory Management Leadership Index (eTable 1 in the Supplement). 21 This instrument was designed to evaluate dimensions of leadership that drive engagement and discretionary effort among team members. The original 12-item instrument was subsequently revised to a 9-item instrument as previously reported. 17 Each item is scored on a 5-point Likert scale (1 = strongly agree; 5 = strongly agree) and scores are summed to yield a total score. The aggregate leadership behavior score for each leader is calculated by the composite evaluations of all responding physicians they supervise.

Statistical Analysis
All statistical analyses were performed using R statistical software version 3.6.0 (R Core Team, 2019) from May to December 2020. Statistical significance was set at 2-tailed P < .05. All items in the survey were scored using the published approach. 19 Burnout, professional fulfillment, and the Participatory Management Leadership index were scaled to 0 to 10 to simplify the interpretation of results. Descriptive summary statistics were used to characterize physician participants. Values Alignment scores among different age, gender, faculty track, academic rank, work hours, and percentage clinical time categories were compared using one-way analysis of variation (ANOVA).

JAMA Network Open | Occupational Health
Pearson correlation coefficients were used to assess the association between each physician's personal-organizational values alignment score and their professional fulfillment and burnout scores.
Pearson correlation was also used to examine the correlation between each supervisor's aggregate leadership score and the mean personal-organizational values alignment score of physicians reporting to them. A random-effects ANOVA-to account for the nested data structure of multiple physicians reporting to individual leaders-was specified with leadership score as the dependent variable and with a random effect for leader (eAppendix in the Supplement). Mixed-effects models were specified to test the associations between a physician's evaluation of their immediate supervisor and their values alignment score, with and without adjustment for gender, age, faculty track, academic rank, work hours, physician-leader gender concordance, and percentage clinical time. Ordinary least squares regression was used to estimate the portion of variance in physicians' values alignment score attributable to their evaluation of their supervisors.

Results
As previously reported, 17  In aggregate, the mean (SD) personal-organizational values alignment score on the 0 to 12 scale was 6.19 (3.21) (Figure 1A). Variation in values alignment scores by age, gender, work hours, faculty track, academic rank, and amount of time dedicated to clinical work is shown in As a continuous variable, values alignment score demonstrated a significant correlation with both occupational burnout (r = −0.39; P < .001) and professional fulfillment (r = 0.52; P < .001). The association between values alignment score and burnout is shown in Figure 1B. Having a higher personal-organizational values alignment score was associated with lower burnout scores. In an inverse manner, having a higher personal-organizational values alignment was associated with higher professional fulfillment scores ( Figure 1C).

JAMA Network Open | Occupational Health
Perception of Leaders' Behavior and Personal-Organizational Values Alignment Among Physicians Professional fulfillment score, mean (SD) b 6.51 (1.99) a The score range is 0 to 12.
b The score range is 0 to 10.

Leadership Behavior of Immediate Supervisor and Personal-Organizational Values Alignment in the Work Unit
Leadership behavior scores for the 66 physician leaders in the analysis ranged from 0 to 10 with a mean (SD) score of 7.57 (2.31). The leadership scale demonstrated high internal consistency (Chronbach α = 0.96). The association between the aggregate leader behavior score (the score for each of the 66 leaders based on mean rating from all the physicians they supervise) and the aggregate personal-organizational values alignment score for all participating individuals in their unit is shown in Figure 2A. Professional fulfillment score vs personal-organization value alignment scores C In panels B and C, circles denote medians, and error bars denote 95% CIs. vs 6.01 [3.11]; P = .04) on univariable analysis, gender concordance between a physician and their leader was not associated with values alignment in the adjusted model.

Discussion
A shared sense of purpose and alignment of values with health care professionals is critical for health care organizations to achieve their mission. In the present study, we observed an association between the leadership behavior score of physicians' immediate supervisor and physicians' perception of values alignment with their organization overall. Our results suggested that 21.6% of the variation in physicians' perceived values alignment with the organization could be explained by the behavior of the work unit leader. This association persisted after adjusting for age, gender, work hours, faculty track, physician-leader gender concordance, and amount of time devoted to clinical work. These findings suggest the magnitude to which people experience and judge their organization through their work unit leader. Importantly, they also provide a potential target for institutional efforts to advance a sense of shared purpose and values alignment through development of firstline leaders. Personal-organizational values alignment was also positively associated with professional fulfillment and inversely correlated with occupational burnout. These findings are consistent with previous studies demonstrating an association between values alignment and physician well-being. 8, 10,23,24 Previous studies also provide evidence that perceived values alignment between physicians and their leaders is associated with objective measures of the quality of care provided by those physicians. 8 These studies further demonstrated that clinics with better values alignment at baseline were more likely to report improvements in the practice environment over time. 25 Women physicians in our study reported lower perceived values alignment with the organization than their male colleagues, a difference that persisted after adjusting for age, work hours, physician-leader gender concordance, faculty track, academic rank, and the amount of time dedicated to clinical work.
This finding suggests that perceived differences in personal-organizational values alignment is yet another factor that may be associated with professional fulfillment of physicians differently on the basis of gender. [26][27][28][29][30][31] It is also notable that, in this sample of academic physicians, values alignment decreased as the amount of professional work effort devoted to clinical care increased. This observation is consistent with previous studies 7, 14,15 suggesting that physicians practicing at medical centers engaged in research often believe their organization values research more highly than clinical care.
Practicing medicine within a large organization represents a departure from many of the practice models of the past. 1  Cultivating engagement, seeking input, and building consensus among a group of physicians with strong opinions and who care deeply about their work and practice environment is not easy. 4,35 Doing so is particularly challenging for first-line leaders who typically have limited experience with these skills and are typically not provided training and support to develop them. Helping set the vision with the team and giving team members the opportunity to help develop the plan on how to achieve it is a critical skill for first-line leaders. It is also essential that these leaders help transmit appreciation and gratitude for the work that physicians do. Our findings add to the growing body of literature on the critical role of leadership in the professional well-being of today's physicians. Evidence suggests that the behavior of immediate supervisors is associated with physician burnout and explains a large proportion of their professional fulfillment. 21 Across the broad universe of leadership skills, keeping people informed, seeking their input in decision-making, understanding their personal passions, facilitating career development, and recognizing their accomplishments appear to be some of the most critical to the well-being of team members. Recent evidence also indicates that leaders' own well-being impacts their independently rated effectiveness in these domains. 17 Organizations can help leaders develop these skills through regular assessment, feedback, training programs, and professional coaching. With respect to the leadership skills associated with the well-being of team members, training and application coaching related to emotional intelligence, creating psychologic safety, humble inquiry, understanding intrinsic motivators, building consensus, appreciative inquiry, improvement methods, and leading through influence may be particularly helpful. 36,37 Limitations Our study has a number of important limitations. First, it is based on analysis of physicians at a single academic medical center. Although it seems unlikely that the association of local leaders' behavior with physicians' perception of values alignment with their organization is an institution-specific effect, the degree of perceived values alignment likely varies across institutions. It is also likely that some findings, such as the association between time dedicated to clinical work and values alignment scores, could be specific to academic medical centers. The findings may also be less relevant for physicians without a clear supervisor working in loosely organized medical groups. Second, our data are cross-sectional and we are unable to determine causality or the potential direction of effect for some associations. Sixth, although the survey response rate was high, response bias is nonetheless a potential limitation.