Hospitals and health care systems are facing increasing pressure to provide high-quality, cost-effective care (Porter & Lee, 2013). With the passage of the Patient Protection and Affordable Care Act and a shift toward value-based payment models such as Medicare’s Comprehensive Care for Joint Replacement model, providers will need to focus on improving the value of care, defined as patient outcomes achieved per health care dollar spent (Friedberg et al., 2015). Although general strategies for promoting value in health care organizations are well described (Porter, 2009), there has been less focus on how organizational structure and leadership drive value (Glickman, Baggett, Krubert, Peterson, & Schulman, 2007; Gunderman & Kanter, 2009; Hearld, Alexander, Fraser, & Jiang, 2008).
Although most hospital and health care leadership positions are currently held by nonphysicians (Gunderman & Kanter, 2009), there is a belief that physicians’ clinical perspective and operational understanding of medicine may position them to more optimally run a hospital or health system in the era of value-based health care (Dwyer, 2010). Goodall described the “Theory of Expert Leadership,” which states that expert leadership is directly dependent on knowledge of the core business activity, industry experience, and general leadership capabilities (Goodall, 2012). Multiple studies have shown a relationship between the success of a leader and his or her skill level/experience with the core business (Goodall, Kahn, & Oswald, 2011; Goodall & Pogrebna, 2012). The Theory of Expert Leadership implies that management experience alone is not sufficient for expert leadership in health care but that experience in clinical medicine is required as well.
Multiple experts have called for a shift to clinician governance (Chaudry, Jain, McKenzie, & Schwartz, 2008; McAlearney, Fisher, Heiser, Robbins, & Kelleher, 2005); however, data on the effect of clinician leadership on hospital performance are limited. Some evidence suggests that physician-led hospitals have better clinical outcomes (Dorgan et al., 2010; Imison & Giordano, 2009) and financial performance (Goes & Zhan, 1995; Molinari, Alexander, Morlock, & Lyles, 1995). Goodall (2011) studied the quality ratings of three specialties in the 100 top rated hospitals in the U.S. News and World Report (USNWR) and found that hospitals led by physician CEOs had higher average quality ratings. Similar studies have shown higher quality ratings and lower morbidity rates in hospitals with physician board members (Jiang, Lockee, Bass, & Fraser, 2009; Veronesi, Kirkpatrick, & Vallascas, 2013) and with physicians in multiple levels of management (Bloom, Sadun, &Van Reenen, 2014). Other evidence suggests that hospitals with physician board members and leaders have better financial performance as indicated by lower operating costs and higher margins (Goes & Zhan, 1995). Although this evidence suggests that a broad view of hospital leadership may be taken when assessing physician involvement (Hambrick & Mason, 1982), the CEO role is particularly significant in that they are singularly responsible for much of the conceptualization and implementation of changes within the system (Leggat, 2007).
To our knowledge, few studies have analyzed hospital performance with a specific focus on the correlation between physician leadership of the overall health system and hospital performance. Hospitals functioning within a larger health system must operate under the mission and regulations of their parent organization and are therefore inherently affected by the overarching leadership of the health system (Kimberly & Evanisko, 1981). The highest-level manager, whether leading one individual hospital or a system of hospitals, is responsible for making many decisions about the strategic direction and operational priorities of the organization. The purpose of this study was to analyze the impact of senior physician leadership on hospital performance—specifically, quality, operating efficiency, revenue, and profitability—of the 115 largest hospitals and health systems in the United States. Our hypothesis was that physician leadership is associated with the delivery of higher-value care. The end points used in the study were selected to emphasize cost and quality—the subcomponents of Michael Porter’s value framework (Porter & Lee, 2013). We selected USNWR quality ratings because of the comprehensive assessment across specialties (Chen, Radford, Wang, Marciniak, & Krumholz, 1999) and to allow for comparability with previous studies (Goodall, 2011).
Hospital Inclusion Criteria
A retrospective review of 2014 Medicare Cost Reports was used to identify the 115 largest hospitals in the United States by staffed bed size as reported (Cost Report worksheet S-3; American Hospital Directory [AHD], 2015), including general medical/surgical beds and special care beds. Staffed beds were defined as “the number of beds available for use by patients at the end of the cost reporting period” (AHD, 2015). The list was cross-referenced with Becker’s 100 Largest Hospitals in America, which reported hospitals of equal size on one line (Gordon, 2014). This totaled 115 individual hospitals, all of which were included in the study to maximize statistical power.
Hospital and Health System Leadership Data
Data on leadership positions were collected manually through available online resources or the hospital or health system Web site or by contacting the institution directly. For each hospital, the highest governing body was identified, and the CEO of that organization was recorded. In many instances, the hospital was part of a larger health care network, and in these cases, the CEO of the health care network was recorded as the leader for that hospital. If the hospital was not determined to be part of a larger health system, the CEO of the hospital was recorded as the leader. Once the CEO was identified, they were classified as either a physician leader if they held a Doctor of Medicine degree or as a nonphysician leader. These data were accessed from December 2014 to February 2015.
Quality data for each hospital were obtained from the USNWR 2015 rating system, which drew from a set of 4,716 facilities and rated hospitals by subspecialty quality from 0 to 100 (Olmstead et al., 2015). To be included in the USNWR data set, a hospital must be a community hospital registered in the American Hospital Association and meet the following criteria. First, the hospital must either (a) be a teaching hospital, (b) be affiliated with a medical school, (c) have at least 200 beds, or (d) have at least 100 beds and have four or more medical technologies out of eight deemed significant for the pertinent patient population. To be qualified for rating in a certain specialty, the hospital needed a minimum number of specialty-specific discharge diagnoses in the previous 3 years combined. Hospitals could additionally be eligible if they had a reputation score in the top 1% of hospitals as reported by physician survey. Individual quality ratings were identified for each of 16 specific specialties. Twelve of the 16 specialties were rated using quantitative Index of Hospital Quality scores: cancer; diabetes; endocrine disorders; digestive disorders; ear, nose, and throat; geriatric care; gynecology; heart and heart surgery; kidney disorders; neurology and neurosurgery; orthopedics; respiratory disorders; and urology. These scores were composed of structural (30.0%), process (27.5%), outcomes (32.5%), and patient safety (10.0%) measures (Olmstead et al., 2015). Structure scores were based on the hospital environment, with an emphasis on patient volumes, staffing ratios, and available technology. Process scores were created from physician surveys that focused on reputation of processes and safety and outcomes scores from mortality rates and safety scores. All ratings controlled for case mix index and were modified for each specialty. The other four specialties were ranked by physician reputation only and were not included in this study: ophthalmology, psychiatry, rehabilitation, and rheumatology. This analysis utilized the quantitative USNWR quality ratings for each of the 12 specialties in our predefined list of hospitals. We did not adjust and modify the ratings for our analysis. The data used were acquired through publicly available sources such as USNWR’s Web site and from hospital Web sites, so no additional ethics approval was necessary for this study (Chaudry et al., 2008).
Financial data were taken from the 2013/2014 Medicare Cost Report, including gross patient revenue, nonpatient revenue, and net income (AHD, 2015). The Medicare Cost Report was also used to identify the number of patient days for each hospital.
Statistical analysis was conducted using SAS (Version 9.3) with a two-tailed alpha of .05. Bivariate analysis was conducted to compare USNWR quality ratings, hospital volume, and financial performance between physician-led and non-physician-led hospitals. This was performed using Student t test after testing for normality and equal variance. Relationships between hospital leadership, percent margin, inpatient days per bed, and the average hospital quality rating were modeled using multivariate logistic regression for discreet variables and multivariable linear regression for continuous variables. All variables were assessed for confounding and interaction, where appropriate.
Analysis included 34 physician-led hospitals and 81 non-physician-led hospitals. Bivariate analysis revealed that physician-led hospitals had significantly higher USNWR quality ratings in all specialties than did hospitals in manager-led networks by an average of 8.5 points (p < .001; Table 1). There were no differences in the number of staffed beds, inpatient days, total revenue, or profit margins between the two cohorts; however, physician-led hospitals had a higher average number of inpatient days per hospital bed (280.4) than did non-physician-led hospitals (259.5; p = .02; Table 1).
In multivariate analysis controlling for physician versus nonphysician CEO, inpatient days per bed, and hospital percent income, physician leadership was the only factor correlated with a higher average quality rating (9.66 points, p < .001; Table 2). Similar analysis of inpatient days per bed revealed that physician leadership was independently associated with greater inpatient days per bed (22.07, p = .02). None of the factors among physician versus nonphysician CEO, inpatient days per bed, or average quality rating were independent predictors of hospital margin. Higher average hospital quality rating (odds ratio = 1.07, p < .001) and a higher number of inpatient days per bed (odds ratio = 1.02, p = .02) were related to a higher likelihood of physician leadership (Table 3).
Health care costs have reached an unprecedented high, resulting in a variety of policy initiatives aimed at reducing costs and improving access to care. Accountable care organizations, bundled payments, and other pay-for-performance models place greater responsibility on hospitals to achieve better outcomes at a lower cost (Porter, 2009). Several studies have suggested performance improvement interventions (e.g., care redesign activities in bundled payment programs) to address these concerns, but little research has investigated how organizational structure affects a health system’s ability to provide higher-value care. Our study aimed to answer this question by analyzing the impact of physician leadership at the most senior level on quality, operational efficiency, and profitability of the 115 largest hospitals in the United States.
Previous studies have indicated better clinical outcomes with physician-managed hospitals. A study of approximately 1,200 hospitals across six countries from 2010 (Dorgan et al., 2010) and 2014 (Bloom et al., 2014) showed that higher management scores were associated with a lower risk-adjusted rate of 30-day acute myocardial infarction mortality (a proxy for hospital quality; Kessler & McClellan, 2000) as well as physician leadership of the hospital (as opposed to manager led). Goodall (2011) performed one of the few available studies evaluating the correlation between physician leadership and hospital quality with a similar methodology to our study. The study compared 2009 USNWR quality ratings of the top 100 ranked hospitals for three specialties (digestive disorders, cancer, and heart and heart surgery) and found that, within each specialty, physician-led hospitals were associated with an 8- to 9-point higher quality rating. Our study analyzed all 12 USNWR quantitatively rated specialties in 2015 and found that all ratings were higher for hospitals in physician-led health systems, with an average difference of 8.5 quality points for hospital systems with physician CEOs (p < .001). Analysis controlling for hospital margin, inpatient days per bed, and physician leadership variables indicated that physician leadership is independently associated with an average difference of 9.66 quality points (p < .001). The reverse relationship holds true, with higher rated hospitals being 7% more likely to be in a physician-led system. Our results correlate closely with the findings from Goodall and indicate that the association between USNWR quality ratings and clinical leadership holds true across specialties and at the highest level of health system governance—an association that is likely multifactorial.
Whereas hospital managers are more directly involved with the day-to-day processes that affect care delivery, overarching health system leaders create the regulations and governance structure that the hospitals operate under. Some authors have suggested that clinically trained leaders may lead to higher quality through more effective communication with staff, clinically informed and patient-centered decisions, and greater employee buy-in than nonclinician leaders (Bloom et al., 2014). Others believe that clinical leaders are better able to evaluate performance (including “course correcting” when necessary), formulate optimal strategic visions and translate them into practice, and create an ideal work environment (Goodall, Bastiampillai, Nance, Roeger, & Allison, 2015). Our hypothesis is that medical training allows physician leaders to better analyze the clinical impact of management decisions, leading to more patient-centric care and better outcomes.
There are much less data regarding the effect of physician versus nonphysician leaders on operational efficiency and financial performance of hospitals and health systems. Goes and Zhan (1995) analyzed multiple hospital-physician integration strategies and identified significant correlations between physician governance and higher average occupancy rate, greater operating margin, and lower operating costs. In an analysis of California hospitals, Molinari et al. (1995) similarly found a positive correlation between having physician board members and achievement of higher operating margins. In contrast, one study (Kaissi & Begun, 2008) found that CEO and board participation in strategic planning efforts around financial performance improved financial performance and that physician involvement alone had no effect. This conflicting evidence displays the lack of consensus in research on this topic to date. In our study, the number of inpatient days per year per staffed bed (analogous to the occupancy rate) was considered a proxy for operational efficiency. Our results indicate a comparably strong correlation between occupancy rate and physician leadership, although we did not identify differences in net income between physician-led and non-physician-led hospitals. Our data further support existing evidence of improved operating performance without an impact on financial performance with physician leadership. It is unclear why operating performance does not lead to improved financial performance. One explanation is that the greater operational efficiency (in this case, more care provided in the form of bed days) generates greater costs as well as revenues. As an example, one study showed that physician management leads to higher staff-to-patient ratios, which is associated with greater cost (Kuntz & Scholtes, 2013), thus potentially neutralizing any increase in revenue. Another explanation is that modern hospital financial teams have developed skills allowing the institution to maintain financial proficiency relatively independent of operational decisions.
There is much variation in the methodology and results of the various hospital rating systems in use today, making it difficult to rely on any one system. One study compared the rankings between five well-known ranking systems—Hospital Compare, Health Grades, Leapfrog Group, USNWR, and Massachusetts Healthcare Quality and Cost—and found minimal consistency between the ranking systems, even for outcome measures focusing on a single diagnosis (Rothberg, Morsi, Benjamin, Pekow, & Lindenauer, 2008). Other studies observed similar discordance between ranking systems, most frequently due to use of different outcome measures, subjective score components, and differences in hospital exclusion criteria (Halasyamani & Davis, 2007; Schauffler & Mordavsky, 2001). It has also been shown that there is a high correlation between USNWR ratings and hospital reputation (Sehgal, 2010), which is a weakness when trying to deduce the objective quality of care. Nevertheless, USNWR is considered the most widely accepted hospital rating system because of its broad implementation, early establishment (Chen et al., 1999), and transparent rating metrics (Austin et al., 2015). Although the incremental value of each quality point with regard to specific health outcomes remains to be elucidated, research has shown a correlation between hospitals rated highly in this system and disease-specific outcomes (Chen et al., 1999). For these reasons, USNWR was chosen for assessing quality in this study.
An important limitation of our findings is that there may be other explanations for the correlation between leadership and hospital quality. For example, although our analysis of the publicly available USNWR rating methodology (Olmstead et al., 2015) indicated no clear bias toward physician leadership, there may be inherent favoritism affecting the “hospital reputation” component of the process score described previously. To assess this theory, analyzing the relationship between hospital leadership and each component of the overall rating score would be helpful, although accessing the data to this level of granularity may not be feasible. Another very real possibility is that physician CEOs are more motivated to pursue higher USNWR ratings for marketing or negotiation purposes or that hospital systems that have hired physician CEOs are making other strategic choices to ensure better scores on the metrics included in the USNWR. Veronesi et al. (2013) addressed this question by analyzing hospitals before and after integrating physicians into their governance structure and found a strong relationship between physician board membership and higher quality. To our knowledge, similar before-and-after studies have not been performed within one hospital or health care network, but the findings in our study add to the literature of hospital quality ratings being correlated with physician leadership.
Additional limitations in this study arise from the difficulty in accessing other characteristics about the physician and nonphysician leaders in our analysis to control for confounders. Although leaders were categorized based on their medical degree, we did not assess other potentially relevant characteristics such as prior health care administration education, work experience, other advanced degrees (such as MBA or MHA), or tenure in the health care industry. This information was not included because of inconsistent availability of the data and an inability to verify the information from nonbiased sources. Some studies have found managerial success tied to administrative education (Garman, Goebel, Gentry, Butler, & Fine, 2010) or tenure at one organization (Bidwell, 2011), and these additional factors should be considered in future studies. It is also likely that the location and payer mix of a hospital have an impact on the financial performance, and future studies should attempt to control for these factors.
Finally, it must be noted that the quality data in our study were obtained from 2015, but the most recent financial data available were from 2013 to 2014. There may have been changes in financial or quality performance across these 1–2 years that obscured any correlation in our analysis.
Understanding drivers of higher-value health care delivery, including organizational structure and physician leadership, will be critical for hospitals to succeed in a value-driven payment landscape.
Our study indicates that physician leadership of a hospital or health system is associated with higher quality ratings, which may support increasing physician presence in hospital leadership positions (Darzi, 2008; Rowling, 2012). Our analysis adds to the growing set of evidence that clinical knowledge and experience in an administrative manager positively influence hospital performance (Goodall, 2012; Stoller, Goodall, & Baker, 2016). When placed in context of the value framework, our findings of higher quality ratings without a negative impact on financial performance suggest that a higher value of care is under physician CEOs. As health care payments become increasingly tied to hospital quality and efficiency to emphasize value, we may eventually be able to better characterize the specific factors that influence the correlation between physician leadership and financial performance that could not be elucidated here.
This study also underscores the importance of reliable health system performance data. The applications of our analysis are limited by the reliability and reproducibility of hospital quality scores across reporting systems as well as the relationship between these scores and specific health outcomes. As we continue to emphasize data-driven decisions in health care, we must develop accurate and pertinent hospital reporting systems to guide our analyses, and eventually, decisions, in hospital management.
Future studies should seek to identify key leadership traits that drive successful health care organizations (Chaudry et al., 2008; Keroack et al., 2007) so that hospital leadership development programs may develop important skill sets in physician and nonphysician managers. These findings could also be applied earlier in clinical training to medical educational programs (including residencies and fellowships) to create the next generation of physician leaders (McAlearney et al., 2005) focused on improving the value of care we deliver to our patients.
Austin J. M., Jha A. K., Romano P. S., Singer S. J., Vogus T. J., Wachter R. M., Pronovost P. J. (2015). National hospital ratings systems share few common scores and may generate confusion instead of clarity. Health Affairs
, 34(3), 423–430.
Bidwell M. (2011). Paying more to get less: The effects of external hiring versus internal mobility. Administrative Science Quarterly
, 56(3), 369–407.
Bloom N., Sadun R., Van Reenen J. (2014). Does management matter in healthcare?
Stanford, CA: Stanford University.
Carsen S., Xia C. (2006). The physician as leader. McGill Journal of Medicine
, 9(1), 1–2.
Chaudry J., Jain A., McKenzie S., Schwartz R. W. (2008). Physician leadership
: The competencies of change. Journal of Surgical Education
, 65(3), 213–220.
Chen J., Radford M. J., Wang Y., Marciniak T. A., Krumholz H. M. (1999). Do “America’s Best Hospitals” perform better for acute myocardial infarction? New England Journal of Medicine
, 340(4), 286–292.
Darzi A. (2008). High quality care for all: NHS next stage review final report
. Norwich, England: TSO.
Dorgan S., Layton D., Bloom N., Homkes R., Sadun R., Van Reenan J. (2010). Management in healthcare, why good practice really matters
. London, UK: McKinsey and Company/London School of Economics.
Dwyer A. J. (2010). Medical managers in contemporary healthcare organisations: A consideration of the literature. Australian Health Review
, 34(4), 514–522.
Friedberg M. W., Chen P. G., White C., Jung O., Raaen L., Hirshman S., Lipinski L. (2015). Effects of health care payment models on physician practice in the United States
. Santa Monica, CA: RAND Corporation.
Garman A., Goebel L., Gentry D., Butler P., Fine D. (2010). Healthcare leadership “outliers”: An analysis of senior administrators from the top US hospitals. Journal of Health Administration Education
, 27(2), 87–97.
Glickman S. W., Baggett K. A., Krubert C. G., Peterson E. D., Schulman K. A. (2007). Promoting quality
: The health-care organization from a management perspective. International Journal for Quality in Health Care
, 19(6), 341–348.
Goes J. B., Zhan C. (1995). The effects of hospital-physician integration strategies on hospital financial performance. Health Services Research
, 30(4), 507–530.
Goodall A. H. (2011). Physician-leaders and hospital performance: Is there an association? Social Science and Medicine
, 73(4), 535–539.
Goodall A. H. (2012). A theory of expert leadership. IZA Institute for the Study of Labor
, 6566, 1–43.
Goodall A. H., Bastiampillai T., Nance M., Roeger L., Allison S. (2015). Expert leadership: Doctors versus managers for the executive leadership of Australian mental health. The Australian and New Zealand Journal of Psychiatry
, 49(5), 409–411.
Goodall A. H., Kahn L. M., Oswald A. J. (2011). Why do leaders matter? A study of expert knowledge in a superstar setting. Journal of Economic Behavior and Organization
, 77, 265–284.
Goodall A. H., Pogrebna G. (2012). Expert leaders in a fast moving setting. IZA Institute for the Study of Labor
, 6715, 1–52.
Gunderman R., Kanter S. L. (2009). Perspective: Educating physicians to lead hospitals. Academic Medicine
, 84(10), 1348–1351.
Halasyamani L. K., Davis M. M. (2007). Conflicting measures of hospital quality
: Ratings from “Hospital Compare” versus “Best Hospitals”. Journal of Hospital Medicine
, 2(3), 128–134.
Hambrick D. C., Mason P. A. (1982). The organization as a reflection of its top managers. Academy of Management Proceedings
, 1982(1), 12–16.
Hearld L. R., Alexander J. A., Fraser I., Jiang H. J. (2008). Review: How do hospital organizational structure and processes affect quality
of care?: A critical review of research methods. Medical Care Research and Review
, 65(3), 259–299.
Imison C., Giordano R. W. (2009). Doctors as leaders. BMJ
, 338, b1555.
Jiang H. J., Lockee C., Bass K., Fraser I. (2009). Board oversight of quality
: Any differences in process of care and mortality? Journal of Healthcare Management
, 54(1), 15–29.
Kaissi A. A., Begun J. W. (2008). Strategic planning processes and hospital financial performance. Journal of Healthcare Management
, 53(3), 197–209.
Keroack M. A., Youngberg B. J., Cerese J. L., Krsek C., Prellwitz L. W., Trevelyan E. W. (2007). Organizational factors associated with high performance in quality
and safety in academic medical centers. Academic Medicine
, 82(12), 1178–1186.
Kessler D. P., McClellan M. B. (2000). Is hospital competition socially wasteful? The Quarterly Journal of Economics
, 115(2), 577–615.
Kimberly J. R., Evanisko M. J. (1981). Organizational innovation: The influence of individual, organizational, and contextual factors on hospital adoption of technological and administrative innovations. Academy of Management Journal
, 24(4), 689–713.
Kuntz L., Scholtes S. (2013). Physicians in leadership: the association between medical director involvement and staff-to-patient ratios. Health Care Management Science
, 16(2), 129–138.
Leggat S. G. (2007). Effective healthcare teams require effective team members: Defining teamwork competencies. BMC Health Services Research
, 7, 17.
McAlearney A. S., Fisher D., Heiser K., Robbins D., Kelleher K. (2005). Developing effective physician leaders: Changing cultures and transforming organizations. Hospital Topics
, 83(2), 11–18.
Molinari C., Alexander J., Morlock L., Lyles A. C. (1995). Does the hospital board need a doctor? The influence of physician board participation on hospital financial performance. Medical Care
, 33(2), 170–185.
Porter M. E. (2009). A strategy for health care reform—Toward a value-based system. New England Journal of Medicine
, 361(2), 109–112.
Rothberg M. B., Morsi E., Benjamin E. M., Pekow P. S., Lindenauer P. K. (2008). Choosing the best hospital: The limitations of public quality
reporting. Health Affairs
, 27(6), 1680–1687.
Rowling E. (2012). Leadership and engagement for improvement in the NHS: Together we can: Report from the King’s Fund leadership review 2012
. London, England: King’s Fund.
Schauffler H. H., Mordavsky J. K. (2001). Consumer reports in health care: Do they make a difference? Annual Review of Public Health
, 22, 69–89.
Sehgal A. R. (2010). The role of reputation in U.S. News & World Report’s rankings of the top 50 American hospitals. Annals of Internal Medicine
, 152(8), 521–525.
Veronesi G., Kirkpatrick I., Vallascas F. (2013). Clinicians on the board: What difference does it make? Social Science and Medicine
, 77, 147–155.
Keywords:Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved
governance; hospital management; patient outcomes; physician leadership; quality