547260
research-article2014
AJMXXX10.1177/1062860614547260American Journal of Medical QualitySzekendi et al
Article
Governance Practices and Performance in US Academic Medical Centers
American Journal of Medical Quality 2015, Vol. 30(6) 520–525 © The Author(s) 2014 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/1062860614547260 ajmq.sagepub.com
Marilyn Szekendi, PhD, RN1, Lawrence Prybil, PhD, LFACHE2, Daniel L. Cohen, MD, FRCPCH, FAAP3, Beth Godsey, MBA1, David W. Fardo, PhD2, and Julie Cerese, MS, RN1
Abstract Recognition of the complex nature of modern health care delivery has led to interest in investigating the ways in which various factors, including governance structures and practices, influence health care quality. In this study, the chief executive officers (CEOs) of US academic medical centers were surveyed to elicit their perceptions of board structures, activities, and attitudes reflecting 6 widely identified governance best practices; the relationship between use of these practices and organizational performance, based on the University HealthSystem Consortium’s Quality & Accountability rankings, was assessed. High-performing hospitals showed greater use of all 6 practices, but the strongest evidence supported a focus on board member education and development, the rigorous use of performance measures to guide quality improvement, and systematic board self-assessment processes. All hospitals, even those with the highest quality ratings, had major gaps in their use of best practices for CEO and board assessments. These findings can serve as the basis for developing sound board improvement plans. Keywords health care governance, quality performance, patient safety, governance best practices The uneven and unpredictable pace of improvement in health care quality and safety in the United States over the past several decades has led to intense and resolute exploration of factors that influence change. The complex, multifactorial nature of modern health care delivery systems is well recognized, and efforts are under way to better understand the effect of contributing factors, positive or negative. Recently, much attention has focused on how governing boards influence the quality of care delivered in health care organizations. Structural, processrelated, and cultural factors have been suggested as affecting a board’s ability to positively influence clinical practice and outcomes. Over the past decade, an array of consensus-based guidelines and best practices on board engagement have been issued.1-8 Foundational research has provided evidence of core features of effective governance of boards of high-performing community health systems,9 large nonprofit health care systems,10 and other acute care hospitals.11-13 This study was designed to expand this evidence base by determining whether use of these guidelines is associated with high performance in an as yet unstudied group, academic medical centers, and to identify opportunities for improvement for these medical
centers’ boards. Research conducted in 2005 and 2010 identified board involvement as a key factor associated with high performance in academic medical centers,14 but the characteristics of beneficial involvement remain unexplored. Six domains of effective governance that have been widely identified as drivers of quality and safety, along with related practice indicators, were systematically selected from the literature to be the focus of this study.1-10 A survey tool was adapted to elicit the perceptions of hospital chief executive officers (CEOs) regarding board structures, activities, and attitudes related to these 6 domains.9 For the purpose of this study, these domains are referred to as “benchmarks” (Table 1).
1
University HealthSystem Consortium, Chicago, IL University of Kentucky, Lexington, KY 3 Datix Ltd (UK), London, UK 2
Corresponding Author: Marilyn Szekendi, PhD, RN, University HealthSystem Consortium, 155 North Wacker Drive, Chicago, IL 60606. Email: [email protected]
Downloaded from ajm.sagepub.com at Duke University Libraries on November 14, 2015
521
Szekendi et al Table 1. Benchmarks of Effective Governance1-10.
Effective boards . . . 1. function in accordance with policies and structures that facilitate the performance of their designated responsibilities. 2. have members with the competencies, skills, and diversity that result in constructive, well-informed deliberation and decision making. 3. carefully establish and closely monitor hospital performance measures and goals through regular review of data, reports, and dashboards. 4. continuously improve CEO performance by setting clear expectations, conducting objective evaluation, and taking follow-up actions. 5. continuously improve board performance by setting clear expectations, conducting objective evaluation, and taking follow-up actions. 6. intentionally create a culture that nurtures engagement, mutual trust, willingness to act, and high standards of performance. Abbreviation: CEO, chief executive officer.
Methods Sample This study was designed to assess the structures, activities, and attitudes of the boards of directors of US academic medical centers and to identify how board practices and characteristics influence hospital quality performance. The study population was 105 US academic medical centers, all principal members of University HealthSystem Consortium (UHC), a not-for-profit alliance of academic medical centers and their affiliated hospitals.
Survey Development and Administration A validated survey developed by Prybil et al9 was adapted for this study to address 6 benchmarks of effective governance (Table 1). Only minor modifications were made to the survey. The survey instrument asked only for information that the CEO could reasonably be expected to answer accurately without more than very minimal investigation. Pretesting was performed by 3 UHC member CEOs to assess for clarity, face validity, and ease of completion, and minor revisions were made based on their feedback. The survey was sent via e-mail to the CEOs of 105 of UHC’s principal members in November 2013; 3 follow-up reminders were sent to nonresponders through the beginning of January 2014. A total of 58 usable responses (55%) were received and analyzed.
Analysis As part of UHC membership, organizations submit quarterly administrative data files to a comprehensive database, the UHC Clinical Data Base; the data are available to members for quality improvement and benchmarking purposes. The data also are used as the basis for UHC’s annual Quality & Accountability (Q&A) Program. The UHC Q&A Program utilizes an established scoring model developed and updated annually in close collaboration
with UHC principal members and used to rank hospital performance based on mortality, care effectiveness, efficiency, safety, and patient-centeredness measures. Q&A Program ranking is used broadly by UHC member organizations to compare year-over-year performance with that of other academic medical centers, to understand relative strengths, and to identify opportunities for improvement. For this study, the 2013 Q&A Study rank of each responding hospital was used to group respondents into 3 performance cohorts: high performers (top Q&A quartile), middle performers (middle 2 Q&A quartiles), and low performers (bottom Q&A quartile). Survey responses were scored using an adapted scoring methodology based on the set of 6 well-established benchmarks and their related practice indicators.9 Data were analyzed using χ2 and Fisher exact tests to test for differences between low, middle, and high performers. Because underrepresentation of low-performing hospitals limited the ability to detect statistically significant differences using bivariate analyses (χ2 and Fisher exact tests), the Pearson correlation coefficient also was calculated to determine correlations between the score of each of the 58 hospitals on the 2013 Q&A Study (not grouped into performance cohorts) and their scores on the governance study survey. Bivariate and correlation techniques produced nearly identical results.
Results A total of 64 of 105 CEOs who received the survey returned completed forms. Of the 64 completed surveys received, 58 (55%) were used in the study analysis; the remaining 6 could not be used either because the hospitals were not ranked in the 2013 Q&A study (4 surveys) or because the hospital from which the survey came could not be identified (2 surveys). The 2013 Q&A Study rank of each responding hospital was used to group the respondents into 3 performance cohorts: high performers (top Q&A quartile, n = 17),
Downloaded from ajm.sagepub.com at Duke University Libraries on November 14, 2015
522
American Journal of Medical Quality 30(6)
Table 2. Benchmark Scores. Benchmark (Points Possible)a
Low Performers (n = 9)
Middle Performers (n = 32)
High Performers (n = 17)
7.0 2.4 3.6
6.9 3.2 4.2
7.5 3.5 4.5
4.2 1.4 5.6 24.4
4.2 2.5 6.2 27.4
4.5 2.5 6.5 29.2
1. Policies and structures (10) 2. Members with competency, skill, diversity (6) 3. Establish and monitor performance measures and goals (5) 4. Improve CEO performance (6) 5. Improve Board performance (5) 6. Culture of engagement (7) Total (39) Abbreviation: CEO, chief executive officer. a Benchmarks with indicators reaching statistical significance are in bold.
middle performers (middle two Q&A quartiles, n = 32), and low performers (bottom Q&A quartile, n = 9). Ranks were not revised to result in a ranking of 1 to 58; instead, the rank that the hospital achieved in the 2013 study was used for grouping purposes. Low 2013 Q&A study performers are underrepresented in this sample. There were no statistically significant differences between responding and nonresponding organizations with respect to public versus private ownership, hospital size, or Q&A average score. Governance structure varies widely across organizations, especially in health care systems, where multiple boards at different levels (system, regional, and local) of the organization often exist. Because nearly all of the responding organizations are part of a health care system, it was necessary to guide the responding CEOs to identify the appropriate board to be the subject of the survey. Respondents were asked to identify which board has the principal responsibility for setting operational performance expectations and for operational oversight of the hospital’s performance. Twenty-eight hospitals (48%) identified the system-level board and responded to the survey with respect to it. Another 28 hospitals identified either the hospital-level board (n = 3) or the system- and hospital-level boards working collaboratively (n = 25); those hospitals were instructed to refer to the hospital-level board when answering the survey items. One organization replied that there was no formalized structure in place and one did not reply to the item. Overall scores for the benchmarks are shown in Table 2. Statistically significant differences were found, relating greater benchmark implementation with higher quality performance, for indicators within 3 of the 6 benchmark domains: board members having the competencies, skills, and diversity that result in constructive, wellinformed deliberation and decision making; careful establishment and monitoring of hospital performance measures and goals through regular review of data, reports, and dashboards; and continuous improvement of
board performance by setting clear expectations, conducting objective evaluation, and taking follow-up actions. Effective boards have members with the competencies, skills, and diversity that result in constructive, wellinformed deliberation and decision making. Statistically significant differences were found between low and high performers for 2 indicators for this benchmark. High performers were significantly more likely to define expertise needs and recruit new members to meet those needs (94% vs 44%; Fisher exact test, P = .0097), and to recognize the importance of ongoing Board member education (88% vs 22%; Fisher exact test, P = .0016). Fifty-nine percent of high-performing hospitals reported having a standing committee with oversight for board education and development programs, compared with 22% of low performers. High performers also were more likely to have nurses as voting members. Forty-four percent of high-performing hospital boards have at least one nurse as a voting member, compared with 11% (only 1 of 9) of low-performing hospital boards. Differences between high and low performers with respect to the percentage of female, physician, and nonwhite board members were not statistically significant. Effective boards carefully establish and closely monitor hospital performance measures and goals through regular review of data, reports, and dashboards. A statistically significant relationship was found between hospital performance and the overall scores for Benchmark 3 (Pearson correlation coefficient, r = .28430, P = .0306). However, for a related practice indicator, stating that the board regularly receives written reports on hospital performance related to established measures and standards for the quality of patient care, an upward pattern was seen (94% of
Downloaded from ajm.sagepub.com at Duke University Libraries on November 14, 2015
523
Szekendi et al high performers responded positively vs 78% of low performers), but this difference was not statistically significant. A similar upward pattern was seen with respect to patientcenteredness. Eighty-two percent of the boards of highperforming hospitals regularly hear stories about real patients’ experiences, compared with 56% of the boards of low-performing hospitals. A statistically significant correlation was found between hospital performance and reported level of prioritization of long-range strategic issues (Pearson correlation coefficient, r = .36262, P = .0052). Effective boards continuously improve board performance by setting clear expectations, conducting objective evaluation, and taking follow-up actions. The regular conduct of board self-assessment is widely accepted as a governance best practice.6,8,9,10 In the present study, a statistically significant relationship was identified for the item “The board’s core governance processes (eg, ongoing oversight of financial performance, CEO evaluation) are reviewed regularly to identify ways to improve them” (100% of high performers responded positively versus 56% of low performers; Fisher exact test, P = .0084). Of note, although 100% of the high-performing hospitals reported conducting regular self-review, only 65% of that group reported engaging in formal self-assessment, and only 6% described the assessment process as thorough and as having resulted in substantial improvement in board performance. None of the low performers reported their board assessment process to be thorough and to have resulted in substantial improvements in board performance. Overall, only 29% of respondents reported that the board assessment process had resulted in actions that substantially changed its practices over the past 2 years; this included only 24% of high performers. One upward trend was notable: 22% of low performers engage in a formal assessment every 1 to 2 years, as compared to 53% of middle performers, and 59% of high performers. Effective boards function in accordance with policies and structures that facilitate the performance of their designated responsibilities. Some board characteristics were found to be universal or nearly so. All boards have predetermined meeting schedules and nearly all have standing committees with clear oversight for quality and safety, finance, and internal and external audit functions. These fundamental fiduciary duties are well codified and widely accepted,15 and there is little room for improvement with respect to their implementation. In contrast, not all boards have a standing committee on Executive Compensation (67% of low performers,
81% of middle performers, and 94% of high performers) or on Board Education and Development (22% of low performers, 47% of middle performers, and 59% of high performers). Although not statistically significant, these figures show a notable upward pattern from low to high performers. A pattern emerged regarding CEO standing as a voting board member. Although 71% of the CEOs of high-performing organizations held voting seats, only 33% of the CEOs of low-performing hospitals were voting board members. Although this difference is not statistically significant, it identifies a distinct difference in approach to the CEO’s role in governance. Effective boards continuously improve CEO performance by setting clear expectations, conducting objective evaluation, and taking follow-up actions. The regular and rigorous evaluation of CEO performance is widely recognized as being of major importance,3,4,7,8 yet only 79% of the organizations in this study conduct regular formal evaluations of CEO performance in relation to identified expectations. Even among high performers, more than 10% reported that their CEO does not receive a regular formal performance review. Furthermore, even lower numbers of respondents report that the board has written performance expectations for the CEO (76% of all respondents and 82% of high performers). Only 43% of all respondents indicated that they believe that the CEO evaluation process provides clear expectations and assesses performance fairly. Interestingly, this percentage is higher among lower performers (56% of low performers, compared to 44% of middle performers, and only 35% of high performers). Effective boards intentionally create a culture that nurtures engagement, mutual trust, willingness to act, and high standards of performance. An atmosphere of active engagement, trust, and constructive deliberation, where high standards of performance are expected, has been widely defined as a board best practice.3,4,6,9,10 CEOs completing this survey rated levels of enthusiasm, trust, and active engagement at board meetings as being very high. The highly subjective nature of these items, all representing the CEO’s perspective on a group he or she closely identifies with, may be reflected on these scores (the highest among the 6 benchmarks; Table 2).
Discussion Statistically significant correlations were identified between governance practices associated with 3 of the 6
Downloaded from ajm.sagepub.com at Duke University Libraries on November 14, 2015
524
American Journal of Medical Quality 30(6)
study domains and high performance in the Q&A Study: board members having the competencies, skills, and diversity that result in constructive, well-informed deliberation and decision making; careful establishment and monitoring of hospital performance measures and goals through regular review of data, reports, and dashboards; and continuous improvement of board performance by setting clear expectations, conducting objective evaluation, and taking follow-up actions. The importance of board education in quality and safety is well supported in the literature and other researchers have reported similar findings.1-6,8-10 The National Quality Forum referred to this factor as the development of “quality literacy.”8 The findings of the present study highlight the importance of recognizing the necessity of appropriate education in health care quality and safety for board members, as well as recruiting new members with relevant experiential backgrounds. It is notable that 59% of high-performing hospitals reported having a standing committee with oversight for board education and development programs, compared with 22% of low performers. The importance of the health care clinician perspective in governance has come to be recognized over the past several decades, and the involvement of physicians on hospital boards has become accepted as a necessary and effective governance practice.8,16,17 Eighty-five percent of the boards in this study’s sample have physician representation, accounting for 19% of total board membership. This finding corroborates those of other studies in which physicians represent 18% of voting board members in secular systems10 and 23% of voting members in community health systems.9 Studies conducted in 2004-2005 and 2008-2009 parallel the present study finding that nurses comprise only 2% of nonprofit hospital and community health system boards.9,10,18,19 However, the extent of nursing representation rises along with hospital performance. Forty-four percent of high-performing hospital boards have at least 1 nurse as a voting member, compared with 11% (only 1 of 9) of low-performing hospital boards. And more high (41%) than low performers (11%) had nursing board representation at or above the median percentage for the entire group. As the largest proportion of the health care workforce, nurses are in a unique position to contribute meaningfully to the governance process, yet 69% of hospitals have no nurses as voting board members. The responsibility of the board to establish and monitor measures and performance goals is well established in the literature.1-6,8-11,13 The present study found that highperforming organizations were more likely to report regular board receipt of written reports on hospital quality performance than were low-performing organizations (94% of high performers as compared to 78% of low
performers). Similar findings were reported by Jha and Epstein, who reported that chairs of boards of high- performing organizations identified clinical quality as one of the top 2 priorities for board oversight more often than chairs of low-performing organizations, with the difference being highly statistically significant.11 The present study finding that 82% of the boards of high-performing hospitals regularly hear stories about real patients’ experiences, compared with 56% of the boards of low- performing hospitals, has support in the literature,2-4 as does the finding of a statistically significant correlation between hospital performance and reported level of prioritization of long-range strategic issues (Pearson correlation coefficient, r = .36262, P = .0052).2,5,12 Despite the finding that board self-assessment is highly correlated to organizational performance, it is clear that this assessment is seriously lagging, even among top performers. Overall scores on this benchmark were the lowest of all 6 (Table 2). Only 65% of the top performers reported engaging in formal self-assessment, and only 6% described the assessment process as thorough and as having resulted in substantial improvement in board performance. Only 24% of high performers reported that the board assessment process had resulted in actions that substantially changed its practices over the past 2 years. Self-assessment processes that result in actionable findings must be developed as a high priority, along with sound plans for implementation of improvement tactics. Gaps also were seen with respect to evaluation of CEO performance. Only 43% of all respondents and 35% of high performers reported that the CEO evaluation process provides clear expectations and assesses actual performance fairly. And although it may seem laudable that 88% of high performers reported that the board formally evaluates the CEO’s performance in relation to identified expectations on a regular basis, it is difficult to understand how any figure less than 100% can be seen as acceptable. There are several limitations to this study. These include the following: (1) Organizations in the bottom quartile of the 2013 Q&A study are underrepresented, making it difficult to establish meaningful comparisons among performance cohorts. For this reason, the Pearson correlation coefficient also was calculated to determine correlations between the score of each of the 58 hospitals on the 2013 Q&A Study (not grouped into performance cohorts) and their scores on the governance study survey. Use of bivariate and correlation techniques produced almost identical results. (2) This study focused on comparing board structures, practices, and cultures in relation to a selected set of established and emerging benchmarks of good governance. There are many other benchmarks that are important and warrant attention but are not
Downloaded from ajm.sagepub.com at Duke University Libraries on November 14, 2015
525
Szekendi et al addressed in this report. (3) This study examined governance practices in US academic medical centers and their applicability to other types of acute care settings is unclear. (4) Although some of the information elicited in this survey represents objectively measurable data, CEOs’ perceptions of their board’s practices and culture also were elicited. These data represent the participants’ views and are subjective in nature.
Conclusions These findings represent a broad picture of governance practices in US academic medical centers in the current health care environment and provide corroborating evidence for 3 widely accepted governance guidelines: the appropriate education and development of board members, the rigorous use of hospital performance measures to guide quality improvement, and systematic board selfassessment processes. In addition, this study found that even the highest performing organizations had gaps in the use of best practices related to effective CEO and board assessment processes. These findings can serve as the basis for developing sound board improvement plans for health care organizations working to improve the quality and safety of the care delivered. Acknowledgments The authors would like to acknowledge Priya Aggarwal, PhD, for her substantial contributions to the data management, analysis, and presentation for this study.
Declaration of Conflicting Interests The authors declared no conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding The authors received no financial support for the research, authorship, and/or publication of this article.
References 1. Corbett-Nolan A, Hazan J. What Every Healthcare Board Needs to Understand About Patient Safety. East Sussex, England: Good Governance Institute; 2010. 2. Benjamin EM. The Board’s role in quality and patient safety. Prescriptions for Excellence in Health Care. 2013;18(Winter):4-8. 3. Institute for Healthcare Improvement. Boards-on-Board, How-to Guide. Cambridge, MA: Institute for Healthcare Improvement; 2008. 4. National Business Group on Health. A toolkit for action: ensuring patient safety across health care. http://www
.businessgrouphealth.org/toolkits/et_patientsafety.cfm. Published 2008. Accessed January 14, 2013. 5. Callender AN, Hastings DA, Hemsley MC, Morris L, Peregrine MW. Corporate responsibility and health care quality: a resource for health care boards of directors. https://oig.hhs.gov/fraud/docs/complianceguidance/ CorporateResponsibilityFinal%209-4-07.pdf. Published 2007. Accessed January 14, 2013. 6. HRET Center for Healthcare Governance. Building an exceptional board: effective practices for health care governance. http://www.hscrc.state.md.us/documents/ ConsumerPublicInterest/HospitalGovernance/CtrHlth CareGovernance_2007.pdf. Published 2007. Accessed July 22, 2014. 7. Bjork D, Farley JD. Creating a culture of collaborative leadership between boards and CEOs: a practical guide for trustees. http://www.americangovernance.com/resources/ monographs/06-collaborative-leadership.shtml. Published 2006. Accessed July 22, 2014. 8. National Quality Forum. Hospital governing boards and quality of care: a call to responsibility. http://www .ihi.org/knowledge/pages/publications/hospitalgoverningboardsandqualityofcareacalltoresponsibility.aspx. Published 2004. Accessed January 14, 2013. 9. Prybil L, Levey S, Peterson R, et al. Governance in HighPerforming Community Health Systems: A Report on Trustee and CEO Views. Chicago, IL: Grant Thornton LLP; 2009. 10. Prybil L, Levey S, Killian R, et al. Governance in Large Nonprofit Health Systems: Current Profile and Emerging Patterns. Lexington, KY: Commonwealth Center for Governance Studies, Inc.; 2012. 11. Jha A, Epstein A. Hospital governance and the quality of care. Health Aff (Millwood). 2010;29:182-187. 12. McDonagh KJ. Hospital governing boards: a study of their effectiveness in relation to organizational performance. J Healthc Manag. 2006;51:377-389. 13. Jiang HJ, Lockee C, Bass K, Fraser I. Board engagement in quality: findings of a survey of hospital and system leaders. J Healthc Manag. 2008;53:121-134. 14. Keroack MA, Youngberg BJ, Cerese JL, Krsek C, Prellwitz LW, Trevelyan EW. Organizational factors associated with high performance in quality and safety in academic medical centers. Acad Med. 2007;82:1178-1186. 15. Entin F, Anderson J, O’Brien K. The Board’s Fiduciary Role: Legal Responsibilities of Health Care Governing Boards. Chicago, IL: Center for Healthcare Governance; 2006. 16. Oliva J, Totten M. A Seat at the Power Table: The Physician’s Role on the Hospital Board. Chicago, IL: Center for Health Care Governance; 2007. 17. McDonagh KJ. Good governance includes clinician participation. Trustee. November/December 2005;31-32. 18. Prybil L. Engaging nurses in governing hospitals and health systems. J Nurs Care Qual. 2009;24:5-9. 19. Prybil L. Nurses in health care governance: is the picture changing? J Nurs Care Qual. 2013;28:103-107.
Downloaded from ajm.sagepub.com at Duke University Libraries on November 14, 2015
research-article2014
AJMXXX10.1177/1062860614547260American Journal of Medical QualitySzekendi et al
Article
Governance Practices and Performance in US Academic Medical Centers
American Journal of Medical Quality 2015, Vol. 30(6) 520–525 © The Author(s) 2014 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/1062860614547260 ajmq.sagepub.com
Marilyn Szekendi, PhD, RN1, Lawrence Prybil, PhD, LFACHE2, Daniel L. Cohen, MD, FRCPCH, FAAP3, Beth Godsey, MBA1, David W. Fardo, PhD2, and Julie Cerese, MS, RN1
Abstract Recognition of the complex nature of modern health care delivery has led to interest in investigating the ways in which various factors, including governance structures and practices, influence health care quality. In this study, the chief executive officers (CEOs) of US academic medical centers were surveyed to elicit their perceptions of board structures, activities, and attitudes reflecting 6 widely identified governance best practices; the relationship between use of these practices and organizational performance, based on the University HealthSystem Consortium’s Quality & Accountability rankings, was assessed. High-performing hospitals showed greater use of all 6 practices, but the strongest evidence supported a focus on board member education and development, the rigorous use of performance measures to guide quality improvement, and systematic board self-assessment processes. All hospitals, even those with the highest quality ratings, had major gaps in their use of best practices for CEO and board assessments. These findings can serve as the basis for developing sound board improvement plans. Keywords health care governance, quality performance, patient safety, governance best practices The uneven and unpredictable pace of improvement in health care quality and safety in the United States over the past several decades has led to intense and resolute exploration of factors that influence change. The complex, multifactorial nature of modern health care delivery systems is well recognized, and efforts are under way to better understand the effect of contributing factors, positive or negative. Recently, much attention has focused on how governing boards influence the quality of care delivered in health care organizations. Structural, processrelated, and cultural factors have been suggested as affecting a board’s ability to positively influence clinical practice and outcomes. Over the past decade, an array of consensus-based guidelines and best practices on board engagement have been issued.1-8 Foundational research has provided evidence of core features of effective governance of boards of high-performing community health systems,9 large nonprofit health care systems,10 and other acute care hospitals.11-13 This study was designed to expand this evidence base by determining whether use of these guidelines is associated with high performance in an as yet unstudied group, academic medical centers, and to identify opportunities for improvement for these medical
centers’ boards. Research conducted in 2005 and 2010 identified board involvement as a key factor associated with high performance in academic medical centers,14 but the characteristics of beneficial involvement remain unexplored. Six domains of effective governance that have been widely identified as drivers of quality and safety, along with related practice indicators, were systematically selected from the literature to be the focus of this study.1-10 A survey tool was adapted to elicit the perceptions of hospital chief executive officers (CEOs) regarding board structures, activities, and attitudes related to these 6 domains.9 For the purpose of this study, these domains are referred to as “benchmarks” (Table 1).
1
University HealthSystem Consortium, Chicago, IL University of Kentucky, Lexington, KY 3 Datix Ltd (UK), London, UK 2
Corresponding Author: Marilyn Szekendi, PhD, RN, University HealthSystem Consortium, 155 North Wacker Drive, Chicago, IL 60606. Email: [email protected]
Downloaded from ajm.sagepub.com at Duke University Libraries on November 14, 2015
521
Szekendi et al Table 1. Benchmarks of Effective Governance1-10.
Effective boards . . . 1. function in accordance with policies and structures that facilitate the performance of their designated responsibilities. 2. have members with the competencies, skills, and diversity that result in constructive, well-informed deliberation and decision making. 3. carefully establish and closely monitor hospital performance measures and goals through regular review of data, reports, and dashboards. 4. continuously improve CEO performance by setting clear expectations, conducting objective evaluation, and taking follow-up actions. 5. continuously improve board performance by setting clear expectations, conducting objective evaluation, and taking follow-up actions. 6. intentionally create a culture that nurtures engagement, mutual trust, willingness to act, and high standards of performance. Abbreviation: CEO, chief executive officer.
Methods Sample This study was designed to assess the structures, activities, and attitudes of the boards of directors of US academic medical centers and to identify how board practices and characteristics influence hospital quality performance. The study population was 105 US academic medical centers, all principal members of University HealthSystem Consortium (UHC), a not-for-profit alliance of academic medical centers and their affiliated hospitals.
Survey Development and Administration A validated survey developed by Prybil et al9 was adapted for this study to address 6 benchmarks of effective governance (Table 1). Only minor modifications were made to the survey. The survey instrument asked only for information that the CEO could reasonably be expected to answer accurately without more than very minimal investigation. Pretesting was performed by 3 UHC member CEOs to assess for clarity, face validity, and ease of completion, and minor revisions were made based on their feedback. The survey was sent via e-mail to the CEOs of 105 of UHC’s principal members in November 2013; 3 follow-up reminders were sent to nonresponders through the beginning of January 2014. A total of 58 usable responses (55%) were received and analyzed.
Analysis As part of UHC membership, organizations submit quarterly administrative data files to a comprehensive database, the UHC Clinical Data Base; the data are available to members for quality improvement and benchmarking purposes. The data also are used as the basis for UHC’s annual Quality & Accountability (Q&A) Program. The UHC Q&A Program utilizes an established scoring model developed and updated annually in close collaboration
with UHC principal members and used to rank hospital performance based on mortality, care effectiveness, efficiency, safety, and patient-centeredness measures. Q&A Program ranking is used broadly by UHC member organizations to compare year-over-year performance with that of other academic medical centers, to understand relative strengths, and to identify opportunities for improvement. For this study, the 2013 Q&A Study rank of each responding hospital was used to group respondents into 3 performance cohorts: high performers (top Q&A quartile), middle performers (middle 2 Q&A quartiles), and low performers (bottom Q&A quartile). Survey responses were scored using an adapted scoring methodology based on the set of 6 well-established benchmarks and their related practice indicators.9 Data were analyzed using χ2 and Fisher exact tests to test for differences between low, middle, and high performers. Because underrepresentation of low-performing hospitals limited the ability to detect statistically significant differences using bivariate analyses (χ2 and Fisher exact tests), the Pearson correlation coefficient also was calculated to determine correlations between the score of each of the 58 hospitals on the 2013 Q&A Study (not grouped into performance cohorts) and their scores on the governance study survey. Bivariate and correlation techniques produced nearly identical results.
Results A total of 64 of 105 CEOs who received the survey returned completed forms. Of the 64 completed surveys received, 58 (55%) were used in the study analysis; the remaining 6 could not be used either because the hospitals were not ranked in the 2013 Q&A study (4 surveys) or because the hospital from which the survey came could not be identified (2 surveys). The 2013 Q&A Study rank of each responding hospital was used to group the respondents into 3 performance cohorts: high performers (top Q&A quartile, n = 17),
Downloaded from ajm.sagepub.com at Duke University Libraries on November 14, 2015
522
American Journal of Medical Quality 30(6)
Table 2. Benchmark Scores. Benchmark (Points Possible)a
Low Performers (n = 9)
Middle Performers (n = 32)
High Performers (n = 17)
7.0 2.4 3.6
6.9 3.2 4.2
7.5 3.5 4.5
4.2 1.4 5.6 24.4
4.2 2.5 6.2 27.4
4.5 2.5 6.5 29.2
1. Policies and structures (10) 2. Members with competency, skill, diversity (6) 3. Establish and monitor performance measures and goals (5) 4. Improve CEO performance (6) 5. Improve Board performance (5) 6. Culture of engagement (7) Total (39) Abbreviation: CEO, chief executive officer. a Benchmarks with indicators reaching statistical significance are in bold.
middle performers (middle two Q&A quartiles, n = 32), and low performers (bottom Q&A quartile, n = 9). Ranks were not revised to result in a ranking of 1 to 58; instead, the rank that the hospital achieved in the 2013 study was used for grouping purposes. Low 2013 Q&A study performers are underrepresented in this sample. There were no statistically significant differences between responding and nonresponding organizations with respect to public versus private ownership, hospital size, or Q&A average score. Governance structure varies widely across organizations, especially in health care systems, where multiple boards at different levels (system, regional, and local) of the organization often exist. Because nearly all of the responding organizations are part of a health care system, it was necessary to guide the responding CEOs to identify the appropriate board to be the subject of the survey. Respondents were asked to identify which board has the principal responsibility for setting operational performance expectations and for operational oversight of the hospital’s performance. Twenty-eight hospitals (48%) identified the system-level board and responded to the survey with respect to it. Another 28 hospitals identified either the hospital-level board (n = 3) or the system- and hospital-level boards working collaboratively (n = 25); those hospitals were instructed to refer to the hospital-level board when answering the survey items. One organization replied that there was no formalized structure in place and one did not reply to the item. Overall scores for the benchmarks are shown in Table 2. Statistically significant differences were found, relating greater benchmark implementation with higher quality performance, for indicators within 3 of the 6 benchmark domains: board members having the competencies, skills, and diversity that result in constructive, wellinformed deliberation and decision making; careful establishment and monitoring of hospital performance measures and goals through regular review of data, reports, and dashboards; and continuous improvement of
board performance by setting clear expectations, conducting objective evaluation, and taking follow-up actions. Effective boards have members with the competencies, skills, and diversity that result in constructive, wellinformed deliberation and decision making. Statistically significant differences were found between low and high performers for 2 indicators for this benchmark. High performers were significantly more likely to define expertise needs and recruit new members to meet those needs (94% vs 44%; Fisher exact test, P = .0097), and to recognize the importance of ongoing Board member education (88% vs 22%; Fisher exact test, P = .0016). Fifty-nine percent of high-performing hospitals reported having a standing committee with oversight for board education and development programs, compared with 22% of low performers. High performers also were more likely to have nurses as voting members. Forty-four percent of high-performing hospital boards have at least one nurse as a voting member, compared with 11% (only 1 of 9) of low-performing hospital boards. Differences between high and low performers with respect to the percentage of female, physician, and nonwhite board members were not statistically significant. Effective boards carefully establish and closely monitor hospital performance measures and goals through regular review of data, reports, and dashboards. A statistically significant relationship was found between hospital performance and the overall scores for Benchmark 3 (Pearson correlation coefficient, r = .28430, P = .0306). However, for a related practice indicator, stating that the board regularly receives written reports on hospital performance related to established measures and standards for the quality of patient care, an upward pattern was seen (94% of
Downloaded from ajm.sagepub.com at Duke University Libraries on November 14, 2015
523
Szekendi et al high performers responded positively vs 78% of low performers), but this difference was not statistically significant. A similar upward pattern was seen with respect to patientcenteredness. Eighty-two percent of the boards of highperforming hospitals regularly hear stories about real patients’ experiences, compared with 56% of the boards of low-performing hospitals. A statistically significant correlation was found between hospital performance and reported level of prioritization of long-range strategic issues (Pearson correlation coefficient, r = .36262, P = .0052). Effective boards continuously improve board performance by setting clear expectations, conducting objective evaluation, and taking follow-up actions. The regular conduct of board self-assessment is widely accepted as a governance best practice.6,8,9,10 In the present study, a statistically significant relationship was identified for the item “The board’s core governance processes (eg, ongoing oversight of financial performance, CEO evaluation) are reviewed regularly to identify ways to improve them” (100% of high performers responded positively versus 56% of low performers; Fisher exact test, P = .0084). Of note, although 100% of the high-performing hospitals reported conducting regular self-review, only 65% of that group reported engaging in formal self-assessment, and only 6% described the assessment process as thorough and as having resulted in substantial improvement in board performance. None of the low performers reported their board assessment process to be thorough and to have resulted in substantial improvements in board performance. Overall, only 29% of respondents reported that the board assessment process had resulted in actions that substantially changed its practices over the past 2 years; this included only 24% of high performers. One upward trend was notable: 22% of low performers engage in a formal assessment every 1 to 2 years, as compared to 53% of middle performers, and 59% of high performers. Effective boards function in accordance with policies and structures that facilitate the performance of their designated responsibilities. Some board characteristics were found to be universal or nearly so. All boards have predetermined meeting schedules and nearly all have standing committees with clear oversight for quality and safety, finance, and internal and external audit functions. These fundamental fiduciary duties are well codified and widely accepted,15 and there is little room for improvement with respect to their implementation. In contrast, not all boards have a standing committee on Executive Compensation (67% of low performers,
81% of middle performers, and 94% of high performers) or on Board Education and Development (22% of low performers, 47% of middle performers, and 59% of high performers). Although not statistically significant, these figures show a notable upward pattern from low to high performers. A pattern emerged regarding CEO standing as a voting board member. Although 71% of the CEOs of high-performing organizations held voting seats, only 33% of the CEOs of low-performing hospitals were voting board members. Although this difference is not statistically significant, it identifies a distinct difference in approach to the CEO’s role in governance. Effective boards continuously improve CEO performance by setting clear expectations, conducting objective evaluation, and taking follow-up actions. The regular and rigorous evaluation of CEO performance is widely recognized as being of major importance,3,4,7,8 yet only 79% of the organizations in this study conduct regular formal evaluations of CEO performance in relation to identified expectations. Even among high performers, more than 10% reported that their CEO does not receive a regular formal performance review. Furthermore, even lower numbers of respondents report that the board has written performance expectations for the CEO (76% of all respondents and 82% of high performers). Only 43% of all respondents indicated that they believe that the CEO evaluation process provides clear expectations and assesses performance fairly. Interestingly, this percentage is higher among lower performers (56% of low performers, compared to 44% of middle performers, and only 35% of high performers). Effective boards intentionally create a culture that nurtures engagement, mutual trust, willingness to act, and high standards of performance. An atmosphere of active engagement, trust, and constructive deliberation, where high standards of performance are expected, has been widely defined as a board best practice.3,4,6,9,10 CEOs completing this survey rated levels of enthusiasm, trust, and active engagement at board meetings as being very high. The highly subjective nature of these items, all representing the CEO’s perspective on a group he or she closely identifies with, may be reflected on these scores (the highest among the 6 benchmarks; Table 2).
Discussion Statistically significant correlations were identified between governance practices associated with 3 of the 6
Downloaded from ajm.sagepub.com at Duke University Libraries on November 14, 2015
524
American Journal of Medical Quality 30(6)
study domains and high performance in the Q&A Study: board members having the competencies, skills, and diversity that result in constructive, well-informed deliberation and decision making; careful establishment and monitoring of hospital performance measures and goals through regular review of data, reports, and dashboards; and continuous improvement of board performance by setting clear expectations, conducting objective evaluation, and taking follow-up actions. The importance of board education in quality and safety is well supported in the literature and other researchers have reported similar findings.1-6,8-10 The National Quality Forum referred to this factor as the development of “quality literacy.”8 The findings of the present study highlight the importance of recognizing the necessity of appropriate education in health care quality and safety for board members, as well as recruiting new members with relevant experiential backgrounds. It is notable that 59% of high-performing hospitals reported having a standing committee with oversight for board education and development programs, compared with 22% of low performers. The importance of the health care clinician perspective in governance has come to be recognized over the past several decades, and the involvement of physicians on hospital boards has become accepted as a necessary and effective governance practice.8,16,17 Eighty-five percent of the boards in this study’s sample have physician representation, accounting for 19% of total board membership. This finding corroborates those of other studies in which physicians represent 18% of voting board members in secular systems10 and 23% of voting members in community health systems.9 Studies conducted in 2004-2005 and 2008-2009 parallel the present study finding that nurses comprise only 2% of nonprofit hospital and community health system boards.9,10,18,19 However, the extent of nursing representation rises along with hospital performance. Forty-four percent of high-performing hospital boards have at least 1 nurse as a voting member, compared with 11% (only 1 of 9) of low-performing hospital boards. And more high (41%) than low performers (11%) had nursing board representation at or above the median percentage for the entire group. As the largest proportion of the health care workforce, nurses are in a unique position to contribute meaningfully to the governance process, yet 69% of hospitals have no nurses as voting board members. The responsibility of the board to establish and monitor measures and performance goals is well established in the literature.1-6,8-11,13 The present study found that highperforming organizations were more likely to report regular board receipt of written reports on hospital quality performance than were low-performing organizations (94% of high performers as compared to 78% of low
performers). Similar findings were reported by Jha and Epstein, who reported that chairs of boards of high- performing organizations identified clinical quality as one of the top 2 priorities for board oversight more often than chairs of low-performing organizations, with the difference being highly statistically significant.11 The present study finding that 82% of the boards of high-performing hospitals regularly hear stories about real patients’ experiences, compared with 56% of the boards of low- performing hospitals, has support in the literature,2-4 as does the finding of a statistically significant correlation between hospital performance and reported level of prioritization of long-range strategic issues (Pearson correlation coefficient, r = .36262, P = .0052).2,5,12 Despite the finding that board self-assessment is highly correlated to organizational performance, it is clear that this assessment is seriously lagging, even among top performers. Overall scores on this benchmark were the lowest of all 6 (Table 2). Only 65% of the top performers reported engaging in formal self-assessment, and only 6% described the assessment process as thorough and as having resulted in substantial improvement in board performance. Only 24% of high performers reported that the board assessment process had resulted in actions that substantially changed its practices over the past 2 years. Self-assessment processes that result in actionable findings must be developed as a high priority, along with sound plans for implementation of improvement tactics. Gaps also were seen with respect to evaluation of CEO performance. Only 43% of all respondents and 35% of high performers reported that the CEO evaluation process provides clear expectations and assesses actual performance fairly. And although it may seem laudable that 88% of high performers reported that the board formally evaluates the CEO’s performance in relation to identified expectations on a regular basis, it is difficult to understand how any figure less than 100% can be seen as acceptable. There are several limitations to this study. These include the following: (1) Organizations in the bottom quartile of the 2013 Q&A study are underrepresented, making it difficult to establish meaningful comparisons among performance cohorts. For this reason, the Pearson correlation coefficient also was calculated to determine correlations between the score of each of the 58 hospitals on the 2013 Q&A Study (not grouped into performance cohorts) and their scores on the governance study survey. Use of bivariate and correlation techniques produced almost identical results. (2) This study focused on comparing board structures, practices, and cultures in relation to a selected set of established and emerging benchmarks of good governance. There are many other benchmarks that are important and warrant attention but are not
Downloaded from ajm.sagepub.com at Duke University Libraries on November 14, 2015
525
Szekendi et al addressed in this report. (3) This study examined governance practices in US academic medical centers and their applicability to other types of acute care settings is unclear. (4) Although some of the information elicited in this survey represents objectively measurable data, CEOs’ perceptions of their board’s practices and culture also were elicited. These data represent the participants’ views and are subjective in nature.
Conclusions These findings represent a broad picture of governance practices in US academic medical centers in the current health care environment and provide corroborating evidence for 3 widely accepted governance guidelines: the appropriate education and development of board members, the rigorous use of hospital performance measures to guide quality improvement, and systematic board selfassessment processes. In addition, this study found that even the highest performing organizations had gaps in the use of best practices related to effective CEO and board assessment processes. These findings can serve as the basis for developing sound board improvement plans for health care organizations working to improve the quality and safety of the care delivered. Acknowledgments The authors would like to acknowledge Priya Aggarwal, PhD, for her substantial contributions to the data management, analysis, and presentation for this study.
Declaration of Conflicting Interests The authors declared no conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding The authors received no financial support for the research, authorship, and/or publication of this article.
References 1. Corbett-Nolan A, Hazan J. What Every Healthcare Board Needs to Understand About Patient Safety. East Sussex, England: Good Governance Institute; 2010. 2. Benjamin EM. The Board’s role in quality and patient safety. Prescriptions for Excellence in Health Care. 2013;18(Winter):4-8. 3. Institute for Healthcare Improvement. Boards-on-Board, How-to Guide. Cambridge, MA: Institute for Healthcare Improvement; 2008. 4. National Business Group on Health. A toolkit for action: ensuring patient safety across health care. http://www
.businessgrouphealth.org/toolkits/et_patientsafety.cfm. Published 2008. Accessed January 14, 2013. 5. Callender AN, Hastings DA, Hemsley MC, Morris L, Peregrine MW. Corporate responsibility and health care quality: a resource for health care boards of directors. https://oig.hhs.gov/fraud/docs/complianceguidance/ CorporateResponsibilityFinal%209-4-07.pdf. Published 2007. Accessed January 14, 2013. 6. HRET Center for Healthcare Governance. Building an exceptional board: effective practices for health care governance. http://www.hscrc.state.md.us/documents/ ConsumerPublicInterest/HospitalGovernance/CtrHlth CareGovernance_2007.pdf. Published 2007. Accessed July 22, 2014. 7. Bjork D, Farley JD. Creating a culture of collaborative leadership between boards and CEOs: a practical guide for trustees. http://www.americangovernance.com/resources/ monographs/06-collaborative-leadership.shtml. Published 2006. Accessed July 22, 2014. 8. National Quality Forum. Hospital governing boards and quality of care: a call to responsibility. http://www .ihi.org/knowledge/pages/publications/hospitalgoverningboardsandqualityofcareacalltoresponsibility.aspx. Published 2004. Accessed January 14, 2013. 9. Prybil L, Levey S, Peterson R, et al. Governance in HighPerforming Community Health Systems: A Report on Trustee and CEO Views. Chicago, IL: Grant Thornton LLP; 2009. 10. Prybil L, Levey S, Killian R, et al. Governance in Large Nonprofit Health Systems: Current Profile and Emerging Patterns. Lexington, KY: Commonwealth Center for Governance Studies, Inc.; 2012. 11. Jha A, Epstein A. Hospital governance and the quality of care. Health Aff (Millwood). 2010;29:182-187. 12. McDonagh KJ. Hospital governing boards: a study of their effectiveness in relation to organizational performance. J Healthc Manag. 2006;51:377-389. 13. Jiang HJ, Lockee C, Bass K, Fraser I. Board engagement in quality: findings of a survey of hospital and system leaders. J Healthc Manag. 2008;53:121-134. 14. Keroack MA, Youngberg BJ, Cerese JL, Krsek C, Prellwitz LW, Trevelyan EW. Organizational factors associated with high performance in quality and safety in academic medical centers. Acad Med. 2007;82:1178-1186. 15. Entin F, Anderson J, O’Brien K. The Board’s Fiduciary Role: Legal Responsibilities of Health Care Governing Boards. Chicago, IL: Center for Healthcare Governance; 2006. 16. Oliva J, Totten M. A Seat at the Power Table: The Physician’s Role on the Hospital Board. Chicago, IL: Center for Health Care Governance; 2007. 17. McDonagh KJ. Good governance includes clinician participation. Trustee. November/December 2005;31-32. 18. Prybil L. Engaging nurses in governing hospitals and health systems. J Nurs Care Qual. 2009;24:5-9. 19. Prybil L. Nurses in health care governance: is the picture changing? J Nurs Care Qual. 2013;28:103-107.
Downloaded from ajm.sagepub.com at Duke University Libraries on November 14, 2015
