Engaging Health System Boards of Trustees in Quality and Safety: Six Must-Know Guidelines

Around the turn of the 21^st century, the National Academy of Medicine (NAM), formerly the Institute of Medicine (IOM), called for profound transformation to improve the culture of patient safety in two landmark reports: To Err Is Human (1999) and Crossing the Quality Chasm (2001). These reports accelerated the pace at which health systems began to understand and implement changes to improve the quality and safety of care.

Several impactful movements and influential studies followed the NAM reports, all continuing the appeal for better, safer care and placing accountability for that change on healthcare leaders. Dr. Donald Berwick, former president and CEO of the IHI, added urgency to the need for transformation when he called 2007 the “Year of Governance,” in The Joint Commission Journal on Patient Quality and Safety, placing responsibility for the proper stewardship of healthcare with health system boards of trustees. Leaders needed to own governance, Berwick explained, for improvement to take hold.

Healthcare governing bodies responded to the NAM and Berwick appeals, as well as appeals and arguments from other governing bodies, with regulations to incentivize good health system governance and crack down on inadequate leadership. This article looks at the repercussions for health systems of falling short on governance measures as well as how organizations can engage their boards around quality and safety measures to better meet their communities’ needs.

Regulators Take a Hard Line on Governance

After Berwick’s call to healthcare boards, regulators soon proved they would take a hard line on governance. For example, in 2008, Modern Healthcare reported that a midsized regional medical center faced intense regulatory scrutiny. The organization’s leadership and board had failed to transparently present and review clinical information and data that would have exposed serious safety issues, resulting in harm to multiple patients. By not understanding their fiduciary responsibilities and requirement to hold leadership responsible for unsafe conditions, the board failed the community it served.

Following patient complaints, state regulators and the CMS cited the medical center for compliance issues surrounding patient care and governance. In May 2008, the state health department discovered that seven leg wound infections occurred among open-heart patients at this hospital; the state then shut down the open-heart program.

In addition, regulators cited the organization for deficiencies in five areas required for participation in Medicare:

1. Governance.
2. Infection control.
3. Medical staff.
4. Surgical services.
5. Quality assurance.

The Significant Risks and Repercussions of Bad Governance

With the regulatory crackdown on governance, risks and repercussions of bad governance have significant, costly, and long-term impacts. For example, after the state released the findings of its report on the medical center above, the hospital CEO abruptly resigned, one nearby hospital suspended referrals of cardiac patients, and the hospital’s owner sent staff over 2,000 miles from the corporate office to provide administrative oversight. Such serious consequences placed greater pressure on health system boards to ensure good governance.

Part of the medical center’s correction plan was to ensure that “the [hospital] board is fulfilling its corporate oversight.” The organization clearly hadn’t taken heed of the growing calls within healthcare to improve quality and patient safety. It failed to establish sound governance practices, despite a growing body of evidence that embracing a culture of patient safety was fundamental for healthcare organizations.

Addressing Urgent Demands for Quality and Patient Safety

While To Err Is Human demonstrated how much needless harm and death occurred in healthcare settings, Crossing the Quality Chasm provided a framework for conceptualizing and defining healthcare quality. This framework, often referred to as the STEEEP framework, laid out six aims for improvement: to provide safe, timely, effective, efficient, equitable, and patient-centered care. Both reports have continued to inform what effective healthcare governance looks like for health system boards.

Although the healthcare industry has improved since these reports’ publication, a 2018 IHI analysis demonstrated that many health systems still lack sound governance practices and that work to improve hospital governance has moved slowly. Governing bodies continue to rest transformation of care delivery firmly on health system boards.

Notable campaigns since the NAM reports, such as the following, outline board responsibility and continue to serve as guidelines towards better governance:

A Call to Responsibility

When the National Quality Forum (NQF) issued Hospital Governing Boards and Quality of Care: A Call to Responsibility in 2004,the NQF strongly encouraged “hospital governing boards to become actively engaged in quality improvement” to place emphasis on the relationship between governance and quality of care.

Getting Boards on Board

In 2006, the IHI launched the 5 Million Lives Campaign, which included a call to health systems to join in “getting boards on board.” The campaign recommended that hospital boards get data and hear stories about safety. The IHI also set the expectation that boards “select and review progress toward safer care as the first agenda item at every board meeting, grounded in transparency, and putting a ‘human face’ on harm data.”

The IHI campaign developed six key steps for improving governance:

1. Setting aims.
2. Getting data and hearing stories.
3. Establishing and monitoring system-level measures.
4. Changing the environment, policies, and culture.
5. Learning (starting with the board).
6. Establishing executive accountability.

Monitoring Performance and Setting Improvement Aims

At its core, the IHI “boards on board” campaign emphasized key elements in creating a culture of patient safety:

• Effectively deploying a measurement system to monitor performance and set performance improvement aims.
• Leveraging meaningful data that a non-clinical board could understand.

The campaign clarified that hospital boards, including non-clinical volunteer trustees, have a fiduciary responsibility to ensure high-quality clinical outcomes in their hospitals.  Simultaneously, however, the increasing pace of value-based purchasing and data transparency—largely driven by CMS, TJC, and National Committee for Quality Assurance (NCQA)—have made it difficult for boards to overlook poor quality and safety performance.

Four Steps for Appropriate Goal Setting

Boards increasingly rely on subcommittees of subject matter experts, including the trustees as well as additional independent external experts, to support the various domains of governance (e.g., finance, quality and patient safety, compliance, and strategy) and identify improvement goals. Hospital or health system staff leadership from these functional areas are then responsible to support their respective board committees.

Goal setting for an organization’s quality and patient safety performance follows four steps:

1. Appropriate measures and targets: Quality and safety leadership (e.g., the chief clinical officer, chief nursing officer, or patient safety officer) and their teams working closely to recommend appropriate measures and targets to the executive team as well as the board quality and patient safety committee.
2. Recommendations to the full board: The safety and quality committee refers its recommendations to the full board of trustees.
3. Recommendations to the quality and safety committee: The staff constructs recommendations to the board quality and patient safety committee based on the needs of its patients, moving from microsystem to macrosystem.
4. Big dot measures: Ultimately, the board looks for “big dot” macrosystem measures (e.g., overall rate of serious adverse events, system-wide mortality versus heart-failure mortality, or overall patient satisfaction rather than satisfaction with hospital food).

The board will not want to see underlying drivers of the big dot measure, unless the organization is failing to improve. Board committees will, however, want to do deeper dives to understand these drivers. Figure 1 provides a schematic for health systems to view the measurement selection process.

Six Guidelines for Quality and Safety Measures for Engaging Healthcare Boards

Health system boards can follow six guidelines to select quality and safety measures most likely to support good governance and drive improvement for their organizations:

1. Emphasize Quality and Patient Safety Goals

Quality and patient safety goals should easily represent at least half of all measures the board reviews. Typically, an organization will track 10 to 20 objectives. As Table 1 shows, boards can use the STEEEP framework and big dot approach to guide measure selection:

STEEEP Framework	Measure Examples
Safe	Serious safety events (hospital acquired infections [HAIs], serious reportable events [SREs]) Mortality
Timely	Median time from ED arrival to ED departure of admitted ED patients
Effective	Readmissions Evidence-based measures and protocols (e.g., sepsis protocol adherence) Preventable hospitalization
Efficient	Length of stay Cost per case Per-capita cost Time to next available appointments Patient flow measures
Equitable	Timely ambulatory carePerformance stratified by race and ethnicity
Patient Centered	Patient experience

2. Leverage NQF-Endorsed Measures

The NQF itself does not develop its own measures; other organizations (e.g., the CMS, TJC, and the NCQA) develop measures of accountability for accreditation purposes. Many organizations select measures endorsed by the NQF consensus development process and link these measures to the six aims of the NAM’s Quality Chasm report or the IHI’s Triple or Quadruple Aim. Some boards of trustees have actually made it policy to only use NQF-endorsed measures for quality and patient safety objectives, when available.

Specialty societies, such as the Society of Thoracic Surgeons and the American College of Cardiology, often submit their measures to the NQF for endorsement. The NQF established committees of nationally recognized experts to review specific measure sets. Thus, NQF-endorsed measures have not only gone through an internal review and development process by the developer, but also get a second review by a NQF’s independent experts. Selecting NQF-endorsed measures reduces the internal debate over the validity of measures within health systems.

3. Use Benchmarking and Risk Adjustment to Select Targets

Meaningful targets are critical. Most boards and CEOs aspire to achieve top performance, but not all measures have a meaningful benchmark. A true benchmark looks at top performers and understands how their process led them to outstanding outcomes. Healthcare providers often simply look at top decile or top quartile performance and strive for that goal without breaking down the process changes required to achieve a better outcome.

There are a few rules to keep in mind:

• Process measures do not require risk adjustment: There is no way to adjust for the failure of a health system to deliver an evidence-based practice based on current guidelines. Variation in performance is not attributed to an underlying risk of a patient population. Examples include the timely reperfusion (therapy to restore blood flow) of a patient arriving with a heart attack or appropriately prescribing discharge medications.  Widely available national performance levels cover virtually all CMS publicly reported measures for comparative purposes. These can help a provider understand performance gaps and prioritize improvement work.
• There is no benchmark for patient harm: Setting a target for an acceptable level of harm is like an airline setting a target for the minimum number of plane crashes per year or a manufacturer setting a target for an acceptable number of employee deaths. Again, CMS and the CDC provide national performance for all measures of harm used for public reporting that are helpful for prioritizing improvement work, but the goal has to be zero harm.
• In contrast, not all mortality, readmissions, or admissions are avoidable: Here, benchmarking against the best in the industry can help drive process improvement. It is important to use appropriate risk adjustment among the multiple models for adjustment available. Ideally, a health system benchmarks against best in class and develops a deep understanding of how a top performer’s processes set it apart from the rest of the industry. Realistically, however, top-performing providers do not share enough detail for organizations to benchmark against. Instead, healthcare ranks providers based on risk-adjusted outcomes data against normative performance of the larger population of providers. The industry assumes that those with the best outcomes are high performers. Yet, 20 years of research on care delivery have demonstrated that better may not be best; when healthcare settles on comparing a single provider/hospital/health system against national performance or database performance at the top levels, it’s creating proxies for best practices. However, given those limitations, some benchmarks and risk adjustment models are better than others. A disproportionate number of benchmarking systems depend only on administrative or claims data. Benchmarking can, and will, improve as the industry increasingly uses more clinical information and artificial intelligence (AI) to proactively identify where an organization is performing well and where it’s underperforming.

4. Access Data Beyond the EHR

A report for the board quality and patient safety committee and the board of trustees requires data from multiple source systems. Although the dramatic growth in EHR usage has yielded an abundance of rich clinical data, an EHR cannot provide all the information a board of trustees needs to see. For example, patient safety data is often collected in separate event or incidence reporting systems; infection data may reside in a distinct surveillance system, and data as part of the revenue cycle process, as well as collected from claims, is also important.

Boards also expect patient experience of care (satisfaction) from multiple settings:

• Hospital.
• Surgical centers.
• Emergency departments.
• Skilled nursing facilities.
• Home care agencies.
• Physician clinics.

The analytics team needs a robust enterprise data warehouse (EDW) or an even more sophisticated data platform, such as the Health Catalyst® Data Operating System (DOS™) to meet the needs of the board. Otherwise, the analytics team will spend much of its time hunting for data and producing reports out of different systems. An integrated data platform will increase the timeliness and consistency of data to provide accurate reporting from across the enterprise.

5. Provide Data and Information for Multiple Organizational Levels

Health system analysts and leadership need to present information to multiple levels in the organization. To determine if its objectives are trending in the correct direction, the full board will typically focus on a single report that includes each of the system’s year-to-date performance for the system’s or hospital’s annual objectives, along with a trend line, spark line, or control chart. Staff will also need to provide an interpretive narrative of the findings for each aim they analyze.

While the board will focus on big dot measures, board committees may also want to understand the organization’s outcomes improvement strategy, especially if the organization is falling short of goals. In a multifacility organization, the subcommittee will also want to determine if certain hospitals, clinics, or other entities are not performing at the same level of the system. Reports that summarize information across multiple facilities run the risk of masking a single facility’s poor performance, so reports must drill down from the system level.

Figure 2 shows a dashboard suitable for board-level reporting, as well as for the executives who will need to speak to performance:

6. Develop a Board-Specific Measurement and Presentation Strategy

Providing information to a board of trustees or a board committee requires a different measurement and presentation strategy than what a service line or department needs to support its operations and performance improvement projects. The most valuable board resource is their time. The trustees consist of a mix of volunteer community members, often with no clinical experience, as well as clinicians from the community. The trustees typically conduct their deliberations with the CEO in attendance as well as other health system leadership. Leadership must present healthcare information clearly so that the board understands performance against meaningful targets but also in a way that empowers the board to raise tough questions about opportunities for improvement. Beyond the staff providing honest and meaningful information, while continuously disclosing failures, a strong board most have sufficient subject matter expertise and independence to hold providers accountable. Presented data must demonstrate that the hospital or health system is placing the safety of the patient first and is looking at the right measures of care. Targets and benchmarks must be clear enough to trustees to encourage them challenge each hospital or health system to improve care and eliminate all patient harm.

The Ongoing, Central Role of the Healthcare Board in Quality and Safety Improvement

Despite the launch of a quality and patient safety movement in the early 21^st century, too many hospitals and health systems still lack the resources to give their boards the information required to meet their fiduciary responsibilities. To support safety and quality progress, the industry must accelerate its ability to collect data from an ever-increasing quantity of sources, as well as transform that data into meaningful information for the board of trustees to digest.

Additional Reading

Would you like to learn more about this topic? Here are some articles we suggest:

1. A Guide to Applying Quality Improvement to Healthcare: Five Principles
2. Introducing the Health Catalyst Monitor™ Patient Safety Suite: Surveillance Module
3. The Top Five 2019 Healthcare Trends