When the findings of To Err Is Human: Building a Safer Health System (Kohn, Corrigan, & Donaldson, 2000) were first released, they made headlines throughout the country. This report of the Institute of Medicine (IOM) Committee on the Quality of Health Care in America drew public attention to the widespread incidence of medical error, particularly to the estimate that as many as 98,000 Americans in acute care hospitals are killed each year by preventable errors.

Coming from a highly respected body like the IOM, this report focused national concern on a problem that a cadre of researchers and analysts had been studying and discussing for several years. A growing body of work, reflected in research literature as well as in the work of several organizations, foundations, health care leaders, and others, had already begun to establish priorities in describing, reporting, and preventing medical errors.

Culture of Safety: A Systems Approach

In addition to estimating the incidence of medical error, the IOM report made several recommendations for preventing error and promoting patient safety. Many of these recommendations highlighted priorities that had been identified previously by the growing patient safety movement. Among these, a key priority (arguably, the key priority) is to replace what the report calls the "culture of blame" with a "culture of safety." (Kohn, Corrigan, & Donaldson, 2000).

The "culture of blame" refers to the traditional approach that health care providers and others have taken toward medical errors. That approach focuses on identifying which individual or individuals to blame for the error, and on taking punitive or remedial action against them, such as termination, suspension, professional discipline, litigation, removal from patient care, or re-education. This approach views errors as resulting from individual incompetence, carelessness, and/or poor judgment.

The approach taken by the IOM panel, as by other patient safety experts, focuses away from individual blame and on systems failures as the root of medical errors. Any individual, no matter how competent and careful, can make a mistake. As long as humans are involved in patient care, the potential exists for committing errors. Systems factors beyond the control of individual health care workers can increase the likelihood of error. Health systems can also provide structures and processes aimed toward preventing error, catching errors before they reach the patient, or minimizing patient harm when errors occur.

The traditional "shame and blame" approach to health care error, i.e., an emphasis on individual culpability, not only focuses away from the central role of systems failures in health care error, but also works to discourage individuals from reporting errors, since being identified as having committed an error can lead to a range of adverse consequences. Consistent underreporting of errors has not only made it difficult to accurately gauge the extent of many types of error; it has also arguably made it more difficult, in many instances, to identify systems "fixes" that might help prevent future instances of many types of errors.

The shift from individual to systems accountability for medical errors also facilitates another shift: from a primary focus on post-hoc faultfinding to a focus on preventing errors. In an optimally safe health care system, one that minimizes the incidence of error and strives to eliminate it completely, blame and fault would be progressively less relevant Preventing errors in health care obviates the need to determine or apportion fault.

This article focuses on health care systems’ accountability for error in health care. For the purposes of this article, our focus is specific to health care delivery organizations and systems. We recognize that there are other important components of the health care system which have significant roles in medical errors and (at least potentially) in their prevention, including, for example, the pharmaceutical industry, medical device manufacturers, and educational institutions. Our narrower focus for this discussion is taken for purposes of scope, clarity and length, and not to negate the important role that other parts of the broader health care system have to play.

Traditional Mechanisms

Some mechanisms for addressing organizations’ accountability for patient safety and for preventing error have existed for some time. However, questions about their effectiveness have been raised by, among other things, the fact that error remains endemic to the health care system.

The Tort System

The tort system provides a means to address some types of health care error, particularly those attributable to negligence. This system is significant in that it provides a mechanism for injured patients or their families to receive monetary compensation—to be "made whole"—when errors lead to death or serious injury. This system, however, is crude and inconsistent as a means of enforcing accountability for error. Not all errors meet the criteria for negligence. For instance, the act or omission involved must depart from commonly accepted practice (the "standard of care,") and the injury that results must be reasonably foreseeable. There are some important exceptions, such as some instances of defective products, where victims may seek damages based on strict liability, in which case negligence does not need to be proved.

In theory, fear of being sued may lead to more careful behavior by health care professionals. But the tort system’s success in deterring negligent behavior has been widely questioned and debated. (Mello & Brennan, 2002). At best, the tort system is an imprecise mechanism of incentivizing "safe" care, since, among other things, it presupposes knowledge of safe practices that may not yet have been established, and since it can also lead to adoption of practices that may have little clinical value (such as ordering unnecessary diagnostic tests).

One of the main reasons why the tort system has proved an imprecise and ineffective method for enforcing patient safety is that patients’ decisions to sue are poorly correlated to incidents of medical negligence: not only do many patients sue when negligence has not occurred; most victims of negligence do not sue (Brennan, et al., 1991; Mello & Brennan, 2002; Studdert, Thomas, Burstin, Zbar, Orav, & Brennan, 2000). This apparently poor match between the presence or absence of negligence and the decision whether or not to sue has led some to believe, as Liang and Cullen (1999) suggest, that "one major goal of the medical malpractice system—to provide efficient and appropriate (clinician) incentives to render nonnegligent care to minimize patient injury—seems to be unfulfilled by the traditional tort system."

Government and Accreditation Standards

Organizations and systems have been held to broad quality requirements set by external agencies, including Medicare Conditions of Participation, state licensing regulations, and Joint Commission on Accreditation of Healthcare Organizations (JCAHO) accreditation standards.

Traditionally, few if any of these spoke specifically to error prevention. In fact, a series of high-profile errors in the mid-1990s helped bring the issue of medical errors to the fore, and even raised questions as to the effectiveness of JCAHO in assuring quality care. Such criticisms were likely appropriate in that a major purpose of JCAHO and its accreditation programs is to attest to accredited health care organizations’ ability to provide quality care. However, very few experts had yet focused on errors as a major quality issue, and the JCAHO standards and policies mainly reflected this overall lack.

Evolving Approaches to Patient Safety

A growing focus on health care error, and the failure of traditional mechanisms to prevent error, has helped to spur new approaches to accountability for patient safety. These include initiatives undertaken by JCAHO, as well as a number of efforts to develop systems for reporting health care errors.

JCAHO Policies, Standards, and Goals

JCAHO initiated its Sentinel Events Policy in 1996. JCAHO defines a Sentinel Event as:

...an unexpected occurrence involving death or serious physical or psychological injury, or the risk thereof. Serious injury specifically includes loss of limb or function. The phrase, ‘or the risk thereof’ includes any process variation for which a recurrence would carry a significant chance of a serious adverse outcome for a recipient of care. Such events are called 'sentinel' because they signal the need for immediate investigation and response.(JCAHO, 2000).

This policy encourages hospitals to report such occurrences to JCAHO and to perform a root cause analysis, in which the hospital identifies and discusses the factors that gave rise to the incident and outlines steps it will take to avoid such future incidents. While hospitals are not required by JCAHO to report incidents, submitting an acceptable root cause analysis can prevent sanctions or other threats to continued accreditation. JCAHO regularly reports on findings from Sentinel Events reports and analyses in its Sentinel Events Alert, which enables other health care organizations to learn from these incidents and to initiate preventive measures. (The Sentinel Events Alert can be accessed at www.jcaho.org/).

In 2001, JCAHO revised its hospital accreditation standards in a number of areas to encourage activities related to prevention of errors. In 2002, it announced six new Patient Safety Goals, which went into effect on January 1, 2003. These goals were derived from reviewing patient safety advisories that had been published in the Sentinel Events Alert. A total of 44 recommendations were identified as being evidence-based or expert-based, cost-effective, and practical. (JCAHO, 2003). From this list of 44, the JCAHO Board of Commissioners adopted six goals and accompanying recommendations. Those six goals are:

• Improve the accuracy of patient identification.
• Improve the effectiveness of communication among caregivers.
• Improve the safety of using high-alert medications.
• Eliminate wrong-site, wrong-patient, wrong-procedure surgery.
• Improve the safety of using infusion pumps.
• Improve the effectiveness of clinical alarm systems.

Each goal includes two specific recommendations. For instance, recommendations accompanying Goal 3 (Improve the safety of using high-alert medications) are "Remove concentrated electrolytes (including, but not limited to, potassium chloride, potassium phosphate, sodium chloride >0.9 percent) from patient care units," and "standardize and limit the number of drug concentrations available in the organization." Recommendations for each of the 6 goals may be found at: www.jcaho.org/. Organizations may also propose alternative approaches (other than the listed recommendations) for achieving one or more of the goals. These approaches must be formally submitted for review and approval by JCAHO for use by the submitting organization.

The development of the JCAHO patient safety goals is significant, since the goals include specific actions that hospitals must take in order to maintain accreditation and which have been linked to reduction in preventable medical errors. The process for establishing these and future goals also puts more teeth into the Sentinel Events Policy, expanding its use beyond retrospective reporting. The process takes some of the more prominent lessons that health care organizations have learned through previous errors and analysis, generalizes them, and uses them as the basis not only for organizational goals, but also for specific, measurable actions with which organizations must comply.

Internal Reporting Systems

Reporting systems provide another potential route for requiring accountability by health systems for medical errors. Traditionally, most health care organizations have had internal mechanisms in place for reporting errors. These have focused primarily on individual nurses, physicians, and other health care workers and have relied on these individuals to report their own errors. A number of factors have typically discouraged health care workers from reporting their errors. The traditional focus on individual blame and punishment has created a significant disincentive for many individuals. The usual processes for filing reports of "incidents" or "unusual occurrences" are bureaucratic, time-consuming and, perhaps most significantly, often fail to result in any visible action to prevent future occurrences. Evidence suggests that such internal reporting systems have resulted in significant under-reporting of errors (Cullen, Bates, Small, Cooper, Nemeskal, & Leape, 1995), frustrating efforts to gain an accurate picture of error rates and/or to gauge the effectiveness of error prevention efforts.

Newer voluntary approaches allow for anonymous reporting and provide a focus on identifying systems failures and means to prevent future errors, rather than on assigning blame. US Pharmacopoeia (USP) and the Institute for Safe Medication Practices operate a Medication Error Reporting (MER) program, which allows nurses, physicians, pharmacists, and students to report medication errors (including errors in ordering, dispensing, and administration) anonymously via mail, fax, Internet, or telephone to USP offices. USP also operates the MedMARX error reporting system. MedMARX allows for anonymous reporting of medication errors collected through participating health care organizations (which enables organizations to track and analyze their errors more carefully).

In 2002, the Veterans Health Administration (VHA) launched its Patient Safety Reporting System (PSRS). The PSRS is closely modeled on the Aviation Safety Reporting System (ASRS), which has been widely credited with improving air safety. Under the PSRS, VHA personnel may file reports of safety-related events, such as errors, other adverse events, close calls, and safety suggestions, confidentially with the National Aeronautics and Space Administration (NASA), which operates the PSRS under contract with the VHA. (NASA also operates the ASRS). NASA contacts the reporter to seek additional information or clarification, but does not retain a record of the reporter’s identity. The purpose of the ASRS is to collect data on safety-related incidents and to identify means to prevent adverse incidents. It is designed as a compliment to existing VHA reporting structures, including its internal reporting system, the Patient Safety Information System. Information on the PSRS can be accessed at the program’s website, http://psrs.arc.nasa.gov/.

External Reporting Systems

External error reporting mechanisms (where they exist) vary considerably in terms of the kind of information that is collected and how it is categorized and used, including whether or not data on medical errors (individual or aggregate) are made available to the public. The IOM report noted three main purposes for mandatory reporting systems: (a) to provide the public with a minimum level of protection by assuring that the most serious errors are reported and investigated and appropriate follow-up action is taken, (b) to provide an incentive to health care organizations to improve patient safety in order to avoid the potential penalties and public exposure, and (c) to require all health care organizations to make some level of investment in patient safety, creating a more level playing field (Kohn, Corrigan, & Donaldson, 2000, p.86).

This IOM report (Kohn, Corrigan, & Donaldson, 2000) identified the existence of inconsistent, sporadic state requirements for error reporting, and offered a proposal for a more consistent and (it argued) useful approach. It proposed:

A nationwide mandatory reporting system should be established that provides for the collection of standardized information by state governments about adverse events that result in death or serious harm. Reporting should initially be required of hospitals and eventually be required of other institutional and ambulatory care delivery systems (pp. 87-88).

The IOM committee also supported expansion of voluntary reporting systems for other, less serious errors.

The IOM proposal represented a compromise of sorts, by focusing on "serious" errors. In theory, it might be most helpful to have information on all errors and near misses. The same systems failures that lead to a harmless error, such as the non-injurious administration of an antacid to the wrong patient, could, in many instances, lead to more injurious, even fatal errors. And the fact that an error was narrowly averted does not reduce the harm that could be caused if, under similar circumstances, the same error is not caught in time. To the extent that a focus on actual occurrences of serious errors represents a compromise, however, it is one born of necessity and practicality; the sheer number of errors (not to mention near misses) would overwhelm a mandatory reporting system and present considerable concerns about both cost and administrative burden.

However, if the IOM committee’s proposal was intended to be narrow or modest, it was not received as such by much of the provider community. Both the American Hospital Association and the American Medical Association objected to the proposal for a national mandatory error reporting system, citing concerns with increased risk of litigation and the potential that errors might be further driven "underground" (Leape, 2002). AMA subsequently modified its stance, rejecting outright opposition to mandatory reporting and focusing on ensuring that reporting systems are non-punitive, and data-based, and that they include legal protections for reported data. (Prager, 2000). While some states have made changes in their error reporting systems since the release of the IOM report, and others have created new systems, there has been no substantial movement toward the IOM proposal of a nationwide, standardized reporting system.

What Should Accountability Mean?

In characterizing mandatory reporting systems, the IOM report makes one important point clear: accountability must be accompanied by some means of enforcement—some consequence. Clearly, the emphasis of efforts to create a safe health care system must be on prevention of errors, rather than on punishment. Over the last several years, health professionals, health care executives, and managers have made significant strides toward improving patient safety. Their efforts have been guided by several factors, including the desire to provide safe care for their patients.

At the systems level, however, improvements in patient safety must often compete for resources, financial and otherwise, with other priorities. As JCAHO President Dennis O’Leary explains:

With operational resources already strained in many organizations, potential investments in patient safety compete every day against other basic needs such as staff recruitment, maintenance of the physical plant, clinical technology upgrades simply to meet the standard of care and other investments to respond to community needs. Further, investments in patient safety--while a moral obligation--usually provide financial benefits to payors and purchasers rather than to the organization, a point not lost on stressed organization leaders (O’Leary, 2003).

Making Systems Accountability Real

The fact that patient safety errors must often compete for resources with other priorities makes it all the more important that the concept of systems accountability be made as concrete and specific as possible. Identifying what systems are accountable for and what consequences attach to that accountability can provide incentives for systems to prioritize patient safety measures.

The JCAHO new Patient Safety Goals represent a step toward providing specificity to systems’ accountability for patient safety, but they also raise some important questions as well. JCAHO has identified a series of clear actions, centered on ensuring the use of established structures and processes for error prevention in six areas, and now requires health care organizations to employ them. Non-compliance puts an organization’s accreditation status at risk, providing a clear incentive and a clear enforcement mechanism for accountability.

Some of the recommendations represent relatively simple actions that have long been known as effective means to prevent injury and death, such as removal of concentrated potassium chloride from patient care units, and marking surgical sites to prevent wrong-site surgery. Notably, the first issue of the JCAHO (1998a) Sentinel Events Alert included a discussion of patient deaths caused by mistaken administration of potassium chloride, and specifically recommended that "health care organizations NOT make concentrated KCl available outside of the pharmacy unless appropriate specific safeguards are in place" (paragraph 5). Issue #6 (JCAHO, 1998b) was devoted to wrong-site surgeries and to specific recommendations for preventing them.

What is noteworthy is not so much that these recommendations are included, but that they have to be. While many health care organizations had taken these steps already, at least some organizations need specific consequences for failure to adopt tested, established, and simple measures such as these.

Should organizations’ accountability be limited to instituting standards that have been set by external agencies and that consist of the most firmly established, best tested practices? The JCAHO Patient Safety Goals consist of 6 out of the 44 goals initially identified for possible inclusion. Granted, JCAHO had to start somewhere; issuing a list of 44 goals with accompanying recommendations would have been overwhelming, and JCAHO plans to adopt additional goals over time. But this helps to emphasize the question: is it sufficient for organizations to wait until more Patient Safety Goals and action recommendations have been added to the list? Or should health care organizations be responsible for taking into account the other 38 goals and for initiating processes to identify latent errors within their own organizations and to take steps to prevent them from occurring?

Closely connected to these questions is the issue of health care organizations’ accountability for identifying and correcting factors within their practice environments that may contribute to errors. For instance, a growing body of literature links quality of care, and particularly prevent of adverse events, with hospitals’ registered nurse staffing levels (Aiken, Clarke, Sloane, Sochalski, & Silber, 2002; Cho, Ketefian, Barkauskas, & Smith 2003; Kovner, Jones, Zhan, Gergen, & Basu 2002; Needleman, Buerhaus, Mattke, Stewart, & Zelevinsky, 2002). JCAHO’s own findings suggest that 24 percent of Sentinel Events reported to it from 1995 through March 2002 were attributable to staffing levels. (JCAHO, 2002). A representation of root causes of all sentinel events from 1995-2002 may be found at www.jcaho.org/). If staffing is closely linked to patient safety, what kind of accountability for appropriate staffing should attach to organizations’ decisions regarding staffing levels and mix? This question may be particularly important in light of O’Leary’s observations that patient safety must often compete with other priorities within health care organizations, since nurse staffing, recruitment, and retention must often compete with other demands on organizational resources.

Identifying Consequences

Answers to many questions about accountability, such as those raised in the previous section, require some consideration of the kinds of consequences that are used to enforce it. There is, of course, a very wide range of possible consequences that could be brought to bear in enforcing accountability for patient safety, and particularly for serious medical errors. This range covers a continuum from relatively mild measures, e.g., a small fine, to harsher penalties such as loss of accreditation, large monetary damages through fines or lawsuits, or even criminal penalties.

Where errors result from an organizations’ failure to adopt well established, recognized preventive practices, such as those embodied in the JCAHO Patient Safety Goals, significant consequences (such as loss of accreditation) may be entirely appropriate. Enforcing accountability for most errors, however, call for less severe consequences.

Mandatory Reporting

In this respect, the IOM proposal for mandatory reporting of serious errors continues to hold some appeal as a mechanism for enforcing health care organizations’ accountability. Such a system could operate as a compliment to other mechanisms that focus on accountability. Public reporting provides a kind of "strict accountability," in the sense that consequences are not limited to errors for which the cause is clearly attributable to specific or identifiable systems failures, or to violations of standards or recommendations derived from an established list. The onus for identifying causes and solutions is placed on the health care organization, which does not have the option of waiting passively until experts have identified and agreed on preventive strategies for common errors. At the same time, the consequences are less severe: public disclosure, a possible diminution of reputation, and an increase in reported error rates are undesirable, but less so than loss of accreditation or licensure, for instance.

As reactions to the IOM recommendation for mandatory reporting make clear, there are a number of details about how a nationwide reporting system would work that need to be carefully discussed and analyzed before a workable proposal for a nationwide reporting system could be made. Questions such as whether or not a reporting system should in fact be mandatory, or whether a mandatory system would be undermined by underreporting, require more discussion. Other issues include: how data would be reported; whether or not individual instances of serious errors would be reported, or if data would be reported in the aggregate; how data might be broken down to be of optimal use; and what kind of risk adjustment mechanisms would be appropriate.

Underreporting is likely to be a problem in any reporting system. One issue to consider is whether health care organizations should face consequences for failure to report an incident. Such an approach is not without precedent. The ASRS, for instance, provides immunity to individual reporters, but such immunity is not available when the individual fails to report the incident and the ASRS learns of it through a third party or other means. Under the JCAHO Sentinel Events Policy, health care organizations may endanger their accreditation by failing to file an acceptable root cause analysis. Earlier iterations of the policy spelled out threats to accreditation status when health care organizations failed to self-report events.

Some providers have raised concerns that reporting systems, either mandatory or voluntary, could increase their chance of being sued, and have called for immunizing or otherwise protecting information submitted as part of any such system from use in litigation. Such concerns have been raised regarding the Sentinel Events Policy (Liang & Coulson, 2002; Liang & Storti, 2000). JCAHO has made protection of root cause analyses and other information submitted as part of the Sentinel Events Policy a major legislative priority (O’Leary, 2003). So far, attempts to provide such protection have been unsuccessful in Congress. As of this writing, two separate bills are under consideration in Congress. Both HR 663, passed by the House on March 12, 2003, and S. 720, favorably reported by the Senate Health, Education Labor and Pensions Committee on July 23, 2003, propose a framework for voluntary reporting of medical errors and provide differing degrees of protection and confidentiality of reported information.

Clearly, the concerns of providers and other organizations regarding expanded risk of liability under error reporting systems should be addressed. In keeping with a shift away from a focus on individuals, and in order to remove disincentives to reporting, some protections for individual professionals should be established.

The issue of protection of health care organizations’ error-related information, while important, is more complex. Public trust and support for error-reduction efforts is critical, and efforts which are seen as enabling organizations to hide their errors may serve to undermine that trust. Despite its significant flaws, the tort system is the only mechanism by which patients and families can seek compensation for death and injury caused by negligence, including negligent medical errors. Efforts to increase clinicians’ willingness to self-report errors by limiting the use of such reports (and related documents) in court come at the expense of patients’ and families’ ability to seek compensation for their injuries. The tort system is by design an adversarial one that focuses on establishing the presence or absence of fault, and that is thus ill-suited to moving away from a "culture of blame." But in the absence of alternative mechanisms for compensation, efforts to limit lawsuits (or access to potential evidence) threaten to treat patients and families as adversaries, not partners in reducing errors. Alternative mechanisms must be explored to ensure compensation for legitimate damage claims while eliminating the focus on individual blame and thus reducing clinicians’ fear of liability. Such alternatives include the development of a no-fault system in which compensation for injury would be made from a designated fund, without making a specific determination as to whether a party was at fault in causing the injury (Mello & Brennan, 2002; Studdert & Brennan, 2001). These alternatives also include the development of some form of group liability that focuses on hospitals (Mello & Brennan, 2002; Sage, 2003) and/or other entities, such as closed-panel health maintenance organizations (Sage, 2003). One argument for a group liability approach is that the deterrent effect of malpractice liability occurs primarily at the level of health care organizations because they are sued more frequently than individual providers (Mello & Brennan, 2002).

In addition, many organizations are gaining considerable experience with disclosing errors to patients and family members. Such transparency is increasingly emphasized as important for professionals, providers, patients, and family members in coming to terms with serious errors and, particularly for health care organizations, in learning from them. The Lexington, KY Veterans Health Administration has had a policy of full disclosure of errors in place since 1987, apparently without a significant increase in malpractice liability (Kraman & Hamm, 1991). System-wide VA risk management policy now provides that when accidental or negligent injury occurs, "the medical center will inform the patient and/or the family, as appropriate, of the event, assure the[m] that medical measures have been implemented, and that additional steps are being taken to minimize disability, death, inconvenience, or financial loss to the patient or family" (Department of Veterans Affairs, 1995; Kraman & Hamm, 1991). And current JCAHO standards for hospitals require that "patients, and when appropriate, their families, are informed about the outcomes of care, including unanticipated outcomes."

Whether or not this experience with disclosure helps to temper organizations’ fears of the "risk" posed by reporting systems should be explored. In light of the fact that "[al]most all ethicists agree that the patient has an absolute right to know what happened and whether what has happened is the result of an error....[I]t would be improper to argue against public reporting on the grounds that it would give information to the patient" (Jencks, 2000). To the extent that disclosure is increasingly seen as an obligation, concerns about protecting information may seem less potent.

Conclusion

Ultimately, a combination of approaches will be necessary to address broad issues of systems' accountability for health care error. The JCAHO Patient Safety Goals are an important step. Time will tell how effective this effort is, particularly as it develops further. Competing approaches on reporting of errors will continue to bear more analysis and study. The bottom line for all efforts to enforce systems accountability for health care error should be their effectiveness in preventing error. This will require continued research on the outcomes of current and developing approaches in order to determine how well they contribute to the goal of creating a safe health care system.

Authors

David Keepnews, PhD, JD, RN, FAAN
E-mail: keepnews@u.washington.edu

David Keepnews is an Assistant Professor in the Department of Biobehavioral Nursing and Health Systems, School of Nursing at the University of Washington in Seattle, where he is also an Adjunct Assistant Professor in the School of Law and the Department of Health Services, School of Public Health & Community Medicine. He teaches courses on health systems, management, policy, and reimbursement. Dr. Keepnews’ research focuses on health policy and related areas, including health care workforce issues, payment and reimbursement, and outcomes. Previously, Dr. Keepnews has served on the professional staff of the American Nurses Association, as an Assistant Regional Counsel for the U.S. Department of Health and Human Services, as Regulatory Policy Specialist for the California Nurses Association, and as a staff nurse in psychiatric and substance abuse settings.

Pamela H. Mitchell, PhD, RN, FAAN
E-mail: pmitch@u.washington.edu

Pamela H. Mitchell is Associate Dean for Research, Professor of Biobehavioral Nursing and Health Systems, and holds the Elizabeth S. Soule Distinguished Professorship of Health Promotion at the University of Washington School of Nursing. She is also Adjunct Professor of Health Services, School of Public Health and Community Medicine, at the University of Washington. Dr. Mitchell has been engaged in advanced nursing practice, clinical and systems research, and nursing education for over 30 years. Her practice and research are in neuroscience and critical care nursing, with expansion into studying care delivery systems as they impact clinical outcomes. She is founding Director of the Center for Health Sciences Interprofessional Education and Research. This center is developing and disseminating a national curriculum for Faculty Leadership in Interprofessional Education to Promote Patient Safety, funded by the Health Resources Service Administration (HRSA). She also serves on the Patient Safety Steering Committee for the portfolio of patient safety research and dissemination funded by HRSA and the Agency for Healthcare Research and Quality (AHRQ).