Setting Performance Standards and Expectations for Patient Safety
The development and availability of standards for patient safety can serve several purposes. They can either establish minimum levels of performance or can establish consistency or uniformity across multiple individuals and organizations. Another purpose for standards is that they set expectations. The process of developing standards can set expectations for the organizations and health professionals affected by the standards. The publication and dissemination of standards additionally helps to set expectations for consumers and purchasers.
Standards can be developed and used in public regulatory processes, such as licensure for health professionals and licensure for health care organizations, such as hospitals or health plans. Standards can also be developed through private voluntary processes, such as professional certification or organizational accreditation.
Although there are many kinds of standards in health care, especially those promulgated by licensing agencies and accrediting organizations, few standards focus explicitly on issues of patient safety. Furthermore, the current lack of safety standards does not allow consumers and purchasers to reinforce the need for safe systems from the providers and organizations with whom they have contact. All existing regulatory and voluntary standard-setting organizations can increase their attention to patient safety and should consistently reinforce its importance.
Expectations for the performance of health professionals and organizations are also shaped by professional groups, purchasers and consumers, and society in general. Professional groups and leaders play a particularly important role in establishing norms and facilitating improvements in performance through educational, convening and advocacy activities. Large public and private group purchasers and purchasing coalitions also have the opportunity to shape expectations through marketplace decisions.
This chapter describes how performance standards and expectations can foster improvements in patient safety. Although this report has described the importance of a systems approach for reducing errors in health care, licensing and accreditation of individual practitioners and organizations can also play a role in reinforcing the importance of patient safety. The primary focus is on how existing models of oversight can be strengthened to include a focus on patient safety. In this report, the committee did not undertake an evaluation of the effectiveness of public and private oversight systems to affect quality of care. The committee recognizes, however, that as the organizational arrangements through which health care is delivered change, an evaluation may be appropriate since the existing models of oversight may no longer be adequate.
In the health care industry, standards and expectations about performance are applicable to health care organizations, health professionals, and drugs and devices. The committee believes there are numerous opportunities to strengthen the focus of the existing processes on patient safety issues.
RECOMMENDATION 7.1 Performance standards and expectations for heath care organizations should focus greater attention on patient safety.
•Regulators and accreditors should require health care organizations to implement meaningful patient safety programs with defined executive responsibility.
•Public and private purchasers should provide incentives to health care organizations to demonstrate continuous improvement in patient safety.
Changes within health care organizations will have the most direct impact on making care delivery processes safer for patients. Regulators and
accreditors have a role in encouraging and supporting actions within health care organizations by holding them accountable for ensuring a safe environment for patients.
Health care organizations ought to be developing patient safety programs within their own organizations (see Chapter 8). After a reasonable period of time for health care organizations to set up such programs, regulators and accreditors should require patient safety programs as a minimum standard. The marketplace, through purchaser and consumer demands, also exerts influence on health care organizations. Public and private purchasers have three tools that can be employed today to demand better attention to safety by health care organizations. First, purchasers can consider safety issues in their contracting decisions. Second, purchasers can reinforce the importance of patient safety by providing relevant information to their employees or beneficiaries. There is increasing attention in providing information to aid in the selection of health coverage. Information about safety can be part of that process. Finally, purchasers can communicate concerns about patient safety to accrediting bodies to support stronger oversight for patient safety.
RECOMMENDATION 7.2 Performance standards and expectations for health professionals should focus greater attention on patient safety.
•Health professional licensing bodies should
(1) implement periodic reexaminations and relicensing of doctors, nurses, and other key providers, based on both competence and knowledge of safety practices; and
(2) work with certifying and credentialing organizations to develop more effective methods to identify unsafe providers and take action.
•Professional societies should make a visible commitment to patient safety by establishing a permanent committee dedicated to safety improvement. This committee should
(1) develop a curriculum on patient safety and encourage its adoption into training and certification requirements;
(2) disseminate information on patient safety to members at special sessions at annual conferences, journal articles and editorials, newsletters, publications and websites on a regular basis;
(3) recog nize patient safety considerations in practice guidelines and in standards related to the introduction and diffusion of new technologies, therapies, and drugs;
(4) work with the Center for Patient Safety to develop community-based, collaborative initiatives for error reporting and analysis and implementation of patient safety improvements; and
(5) collaborate with other professional societies and disciplines in a national summit on the professional's role in patient safety.
For most health professionals, current methods of licensing and credentialing assess knowledge, but do not assess performance skills after initial licensure. Although the state grants initial licensure, responsibility for documenting continued competence is dispersed. Competence may be considered when a licensing board reacts to a complaint. It may be evaluated when an individual applies to a health care organization for privileges or network contracting or employment. Professional certification is the current process for evaluating clinical knowledge after licensure and some programs are now starting to consider assessment of clinical skills in addition to clinical knowledge. Given the rapid pace of change in health care and the constant development of new technologies and information, existing licensing and accreditation processes should be strengthened to ensure that all health care professionals are assessed periodically on both skills and knowledge for practice.
More effective methods for identifying unsafe providers and better coordination between the organizations involved are also needed. The time between discovery of a problem, investigation, and action can currently last several years, depending on the issue and procedures for appeal or other processes. Efforts should be made to make this time as short as possible, while ensuring that practitioners have available the due process procedures to which they are entitled. States should also be more active in notifying other states when a practitioner's license is rescinded. Although unsafe practitioners are believed to be few in number and efforts to identify such individuals are not likely to improve overall quality or safety problems throughout the industry, such efforts are important to a comprehensive safety program.
Finally, professional societies and groups should become active leaders in encouraging and demanding improvements in patient safety. Setting standards, convening and communicating with members about safety, incorporating attention to patient safety into training programs, and collaborating across disciplines are all mechanisms that will contribute to creating a cul-
ture of safety. As patient advocates, health care professionals owe their patients nothing less.
RECOMMENDATION 7.3 The Food and Drug Administration (FDA) should increase attention to the safe use of drugs in both preand postmarketing processes through the following actions:
• develop and enforce standards for the design of drug packaging and labeling that will maximize safety in use;
• require pharmaceutical companies to test (using FDA-approved methods) proposed drug names to identify and remedy potential sound-alike and look-alike confusion with existing drug names; and
• work with physicians, pharmacists, consumers and others to establish appropriate responses to problems identified through postmarketing surveillance, especially for concerns that are perceived to require immediate response to protect the safety of patients.
FDA's role is to regulate manufacturers for the safety of their drugs and devices; however, even approved drugs can present safety problems when used in practice. Drugs may be prone to error in use due to soundalike or look-alike names, unclear labeling, or poorly designed packaging. FDA standards for packaging and labeling of drugs should consider the safety of the products in actual use. Manufacturers should also be required to use proven methods for detecting drug names that sound or look similar. If necessary, Congress should take appropriate action to provide additional enabling authority or clarification of existing authority for FDA to implement this action. Since not all safety problems can be predicted or avoided before a drug is marketed, FDA should also conduct intensive and extensive monitoring to identify problems early and respond quickly when serious threats are discovered in the actual use of approved drugs.
Current Approaches for Setting Standards in Health Care
Generically, standards can be used to define a process or outcome of care. The Institute of Medicine defines a quality standard as a minimum level of acceptable performance or results or excellent levels of performance or results or the range of acceptable performance or results.1 Other definitions for standards have been enacted through legislation, such as the Occupational Safety and Health Act of 1970, which defines a safety and health standard as one that requires conditions, or the adoption or use of one or
more practices, means, methods, operations or processes, reasonably necessary or appropriate to provide safe or healthful employment and places of employment.2 A variety of standards have also been defined through private organizations, such as the American Society for Testing and Materials (see Appendix B). The committee does not recommend one definition or type of standard over another, but recognizes that standards can be quite varied and that as standards specific to safety are developed, they could take multiple forms and focus.
In health care, standards are set through both public, regulatory initiatives and private, voluntary initiatives. Standards can apply to health care organizations, health professionals, and drugs and medical devices. For health care organizations (e.g., health plans, hospitals, ambulatory care facilities), standards are set through licensure and accreditation and, to some extent, requirements imposed by large purchasers, such as Medicare and Fortune 500 companies. For health care professionals, standards are set through state licensure, board certification, and accrediting and credentialing programs. For drugs and devices, the FDA plays a critical role in standard setting.
In general, current standards in health care do not provide adequate focus on patient safety. Organizational licensure and accreditation focus on the review of core processes such as credentialing, quality improvement, and risk management, but lack a specific focus on patient safety issues. Professional licensure concentrates on qualifications at initial licensure, with no requirements to demonstrate safe and competent clinical skills during one's career. Standards for drugs and medical devices concentrate on safe design and production, with less attention to their safe use. Current standards in health care leave serious gaps in ensuring patient safety.
Performance Standards and Expectations for Health Care Organizations
Standards and expectations for health care organizations may be established through oversight processes, primarily licensing and accreditation requirements. Additionally, large public and private purchasers may also impose demands on health care organizations. Each is discussed in this section.
Licensing and Accreditation
There is a great deal of variation in state licensure requirements for health care organizations. Responsibility for licensure rests at the state level,
with each state setting its own standards, measurement, and enforcement. Although standards and measurement can be made more similar, enforcement is always likely to vary to some extent depending on the level of resources devoted by a state to this activity.
In many states, licensure and accreditation are intertwined. For hospital licensure, 44 states accept the Joint Commission on Accreditation of Healthcare Organization's evaluation, in whole or in part, as a condition for licensure (Margaret VanAmringe, JCAHO, personal communication, February 23, 1999). Some states may additionally require compliance with other standards related to building safety or medical care issues that are tracked in that particular state. The remaining states do not link hospital licensure and accreditation. Although the overwhelming tendency to use JCAHO increases the consistency of standards nationally, differences in application also contribute to the variation in ensuring patient safety. For licensure of health maintenance organizations (HMOs), some states rely on private accrediting bodies, primarily the National Committee for Quality Assurance (NCQA), to conduct reviews of health plans. It should also be noted that other health facilities, such as some ambulatory care centers or physicians' offices, may not be licensed at all and are generally not subject to traditional methods of oversight. One of the few mechanisms in place today that more broadly examines care in the ambulatory setting is managed care organizations.
Three private-sector agencies play a role in organizational accreditation: JCAHO, NCQA, and the American Accreditation Healthcare Commission/URAC. Each effort, to some degree, encompasses aspects of standard setting and performance measurement.
JCAHO accredits more than 18,000 health care organizations, including hospitals, health plans, home care agencies, and others.3 Its longeststanding accreditation program applies to hospitals. JCAHO accredits hospitals for three-year periods based on compliance with its standards in the areas of patient rights and patient care: organizational performance; leadership; information management; and nursing and medical staff structures. Approximately 85 percent of hospitals are accredited byJCAHO. Both Joint Commission-accredited hospitals and those accredited by the American Osteopathic Association are deemed to meet Medicare conditions of participation. JCAHO is incorporating performance information into the accreditation process through its Oryx system, in which hospitals will collect clinical data on six measures and submit performance data on these measures. This
system was introduced in 1997 and is required by the Joint Commission for a hospital to be accredited. Eventually hospitals will have to demonstrate specific Oryx performance to maintain their accreditation status.
NCQA accredits health plans for periods of one, two, or three years. The accreditation process covers areas related to quality improvement, credentialing, members' rights and responsibilities, preventive health services, utilization management, and medical records. Approximately 14 states incorporate accreditation into their licensure requirement for health plans; another six states require that health plans have external reviews, most of which are done by NCQA (Steve Lamb, NCQA, personal communication, March 2, 1999). A number of states also require that health plans serving public employees and/or Medicaid enrollees be accredited. NCQA's performance dataset, the Health Plan Employer Data and Information Set (HEDIS), looks at indicators of effectiveness of care, access or availability, satisfaction, health plan stability, use of services, and costs. Beginning in July 1999, accreditation criteria began to incorporate HEDIS measures, initially being used only if they increase a health plan's overall score.4 Accreditation status will also change with the top 20 percent of health plans earning the status of ''excellent."
URAC was established in 1990 and offers nine different accreditation programs for managed care organizations, such as health plan accreditation, health network accreditation, health utilization management accreditation, and network practitioner credentialing.5 Individual managed care organizations can seek accreditation under different sets of programs depending on the range of services they offer. URAC accreditation focuses on preferred provider organization (PPO) and point-of-service (POS) plans. Approximately 22 states have incorporated Commission/URAC accreditation into their regulatory structures.
Purchaser Requirements and Demands
Both private and public purchasers have the ability to encourage health care organizations and providers to pursue continuous improvements in patient safety. Large group purchasers, such as Fortune 500 companies or the Health Care Financing Administration, and purchasing coalitions that provide insurance to large numbers of people are well positioned to exert considerable leverage in the marketplace.
Private Group Purchasers
There are numerous examples of large private employers that incorporate quality issues into their decision-making process when selecting health plans and providers to offer to employees.6 Xerox Corporation ranks health plans according to various quality indicators, including accreditation status, satisfaction ratings, and quality indicators. ARCO evaluates health plans based on 50 different quality and access criteria, and ties the employer contribution to the premium level of the highest-ranking plan. In a survey of 33 large purchasers in four states, 45 percent reported using HEDIS data (i.e., NCQA's Healthplan Employer Data and Information Set quality indicators), 55 percent reported using accreditation data, and 53 percent reported using consumer satisfaction survey data to choose a health plan.7
Although some large employers have incorporated quality considerations into their purchasing decisions, this is not the norm. A 1997 survey of 325 U.S. companies found that most employers consider provider network characteristics, but only a fraction consider quantifiable measures of access, quality or outcomes.8 Another survey found that nearly two-thirds of midsize and large employers are unfamiliar with NCQA accreditation, the most widely used accreditation program for health plans.9
Clearly, there is much opportunity for large employers to place greater emphasis on quality, and specifically patient safety, issues when making decisions to contract with a specific health plan and in the design of payment and financial incentive systems to reward demonstrated quality and safety improvements.
Health Care Financing Administration
As a major national purchaser of health care services, HCFA sets standards through payment policies and conditions of participation for the organizations with which it contracts. HCFA provides health insurance for 74 million people through Medicare, and in partnership with the states, Medicaid, and Child Health Insurance programs.10 It also performs a number of quality-focused activities, including regulation of laboratory testing, surveys and certification, development of coverage policies, and quality improvement initiatives.
The peer review organizations (PROs) monitor the utilization and quality of care of Medicare beneficiaries through a state-based network.11 They have three functions. First, they conduct cooperative quality improvement projects in partnership with other quality-focused organizations. Among the
current projects are programs on diabetes, end-stage renal disease, influenza campaign, and quality improvement systems for managed care. Second, PROs conduct mandatory case review in response to beneficiary complaints, as well as educational and outreach activities. Third, they oversee program integrity by ensuring that Medicare pays only for medically necessary services. Patient safety has not been identified as a priority to date, however, HCFA is giving serious consideration to making patient safety a higher priority.12
Medicare and Medicaid survey and certification activities are aimed at ensuring that providers and suppliers for these programs meet health, safety, and program standards.13 They deal with issues related to the effective and efficient delivery of care to beneficiaries, ensuring their safety while in health care facilities and improving their quality of care. HCFA relies on state health agencies as the principal agents to perform certification activities through their licensure activities. As already noted, state health departments, in turn, often rely on JCAHO as part of licensing a hospital.
Standards for Health Professionals
Performance standards and expectations for health professionals may be defined through regulatory and other oversight processes, such as licensing, accreditation, and certification. Standards and expectations may also be shaped by professional societies and other groups that voluntarily promulgate guidelines or protocols and sponsor educational and convening activities.
Licensing, Certification, and Accreditation
Compared to facility licensure (as discussed in the previous section) there is even greater variation found in professional licensure. There are several reasons for this. First, professional licensure is structured through individual licensing boards for each regulated profession in the state.14 The result is variation both within states and across states. Within states, there is little coordination of management or dissemination of information among different boards.15 Across states, there is variation in what is considered a complaint and in the rate at which disciplinary action is taken. Variation in what is considered a "complaint" influences what is investigated and what can be shared and when. A call to the licensing board may be considered a complaint, or a complaint may be recognized only when there is a formal
charge. It is not clear, therefore, when information can be shared: when something is filed (which may or may not lead to a charge), while it is being investigated, after there is a charge, or only if disciplinary action is taken. Inconsistencies permit unsafe practitioners to move to different jurisdictions before a complaint can be investigated and handled.16
Although not a comprehensive measure of effectiveness, there is wide variation in the rate at which state licensing boards take serious disciplinary actions against physicians, ranging from 0.85 per 1,000 physicians in Louisiana to 15.40 per 1,000 physicians in Alaska, based on data from the Federation of State Medical Boards.17 Across the country, the rate was 3.76 actions per 1,000 physicians in 1998. States that appeared to be doing a better job (more disciplinary actions) tended to have better funding, and more staff, conducted proactive investigations (as opposed to waiting for complaints), used other available data (e.g., Medicare or Medicaid data), had good leadership, were independent from state medical societies and other parts of state government, and had a reasonable statutory framework for conducting their work. Board action can also be quite slow. For example, the Virginia Board of Medicine takes an average of more than two and a half years to resolve a case.18
The National Council of State Boards of Nursing has endorsed a mutual recognition model for interstate nursing practice to encourage reciprocal arrangements between states for licensing and disciplinary action (Carolyn Hutcherson, National Council of State Boards of Nursing, personal communication, June 1, 1999).19 The goal would be to make licensure more like the rules used for a driver's license. That is, licensure is recognized across state lines, but the nurse would still be subject to the rules of a state while in that state (e.g., even if a driver's residence is in Maryland, the driver can still get a speeding ticket in Texas).
Another issue related to professional licensure is that there is no continuing assessment or required demonstration of performance after initial licensure is granted, except for physician assistants and emergency medical technicians.20 In general, the state is involved in initial licensure or followup of complaints; processes for documenting continued competence are voluntary.
For example, physicians may voluntarily seek board certification through one of 24 specialty medical boards that have been approved by the American Board of Medical Specialties (ABMS).21 The specialty boards set professional and educational standards for the evaluation and certification of physician specialists. Initial certification is granted by passing written and
oral examinations. Recertification occurs at seven- to ten-year intervals, although not all boards require recertification. Recertification is granted based on self-assessment, examinations, and credentialing (e.g., unrestricted license, good standing in practice, hospital privileges (Linda Blank, American Board of Internal Medicine, personal communication, May 18, 1999). A minimum number of continuing education credits may also be required. At the present time, there is no assessment of practice skills, although some specialty boards have committed a broader and more timely assessment of competence.22
Another voluntary approach is the American Medical Accreditation Program (AMAP), which is being developed by the American Medical Association. AMAP is a voluntary process, begun in 1998, for the accreditation of individual physicians that is designed to measure and evaluate individual physicians against national standards and peer performance.23 The program will evaluate physicians in five areas: (1) credentials; (2) personal qualifications (including ethical behavior and participation in continuing medical education, peer reviews, and self-assessment of performance); (3) environment of care (including a site review of office operations and medical records); (4) clinical processes (including standardized measures of key patient care processes and comparative feedback to the physician); and (5) patient outcomes (including standardized measures of patient outcomes, perceptions of care, and health status). Although this is a national program, it is being implemented on a state-by-state basis.
A comparable process is found in nursing, which recognizes specialty practice through board certification. One such specialty certifying body is the American Nurses Credentialing Center (ANCC), a subsidiary of the American Nurses Association. Specialty certifying boards set professional and educational standards for the defined specialty and determine a mechanism for establishing continued competency through the recertification process, which occurs every three to five years, depending on the specialty. Although safety is not an explicit focus of certification exams, areas covered may relate to safety, for example, medication errors. Nurses may pursue certification voluntarily, although some states require it for licensure at advanced levels such as nurse practitioner (Ann Carey, R.N., American Nurses Credentialing Center, personal communication, July 20, 1999). Certifying organizations are exploring alternative ways to validate continued competency in addition to continuing education.
Health care organizations are also involved in assessing the continued performance of professionals when hiring nurses or credentialing physicians
for hospital privileges, network membership, or employment. Again, there is little consistency in the standards used and little opportunity for communication across organizations. For example, an unsafe provider may be dismissed from one hospital, with no notification to the licensing board and limited ability for the next hospital to find out the reasons for the dismissal.
The Pew Health Professions Commission conducted an extensive investigation of licensure and continued competency issues. Its report identifies four places in which assessment of competency can occur: upon entry into practice, for continuing authorization to practice, reentry to practice, and after disciplinary action.24 The report recommended increased state regulation to require health care practitioners to "demonstrate their competence in the knowledge, judgment, technical skills and interpersonal skills relevant to their jobs throughout their career." They note that considerations of competence should include not only the basic and specialized knowledge and skills, but also other skills such as "capacity to admit errors." In their view, the current system that relies on continuing education and disciplinary action after a problem has occurred is insufficient. The trend toward computer-based testing should facilitate greater attention to skill assessment in the future. Physician licensure tests and physician recertification are moving toward interactive, computer-based testing, and nursing is also testing a computerized system for initial licensure.25
The Role of Health Professional Societies and Groups
Professional societies, groups, and associations can play an important role in improving patient safety by contributing to the creation of a culture that encourages the identification and prevention of errors. Few professional societies or groups have demonstrated a visible commitment to reducing errors in health care and improving patient safety. Although it is believed that the commitment exists among their members, there has been little collective action. The exception most often cited is the work that has been done by anesthesiologists to improve safety and outcomes for patients.
Anesthesiology has successfully reduced anesthesia mortality rates from two deaths per 10,000 anesthetics administered to one death per 200,000–300,000 anesthetics administered (see Chapter 2). This success was accomplished through a combination of:
• technological changes (new monitoring equipment, standardization of existing equipment);
• information-based strategies, including the development and adoption of guidelines and standards;
• application of human factors to improve performance, such as the use of simulators for training;
• formation of the Anesthesia Patient Safety Foundation to bring together stakeholders from different disciplines (physicians, nurses, manufacturers) to create a focus for action; and
• having a leader who could serve as a champion for the cause.26
To explore the ways that professional societies could improve patient safety, the Institute of Medicine (IOM) convened a one-day workshop on September 9, 1999 with 14 health professionals representing medicine, nursing, and pharmacy (workshop participants are included in the acknowledgments). These leaders are interested and involved in issues related to patient safety and are active in professional societies, although they did not participate in the workshop as representatives of these societies. Four broad roles were identified that could be employed, individually or in combination, to create a culture of safety. These roles are: (1) defining standards of practice; (2) convening and collaborating among society members and with other groups; (3) encouraging research, training and education opportunities; and (4) advocating for change.
One way that professional societies contribute to standards of practice is through the promulgation and promotion of practice guidelines. A number of professional groups have produced practice guidelines and defined best practices in select areas. Guidelines produced by the American College of Cardiology (ACC) and the American Heart Association Task Force of Practice Guidelines are consistently cited models. They have produced sixteen guidelines ranging from coronary artery bypass graft (CABG) to management of chronic angina.27
Pharmacy has also devoted significant attention to patient safety. The American Society of Health-System Pharmacists (ASHP) has published extensively on safe medication practices. Reduction of medication errors has been an identified priority for a decade and is reflected through publications in professional and scientific journals, educational programming, and advocacy. Included among the standards and guidelines is a widely disseminated list of the top priority actions for preventing adverse drug events in hospitals.
Practice guidelines can also be written through a more interdisciplinary approach, such as the perinatal guidelines published jointly by the American
College of Obstetricians and Gynecologists (ACOG) and the American Academy of Pediatrics. There is now a fourth edition of these guidelines. As recognition has grown that errors are caused by failures in systems, interdisciplinary collaboration may become increasingly necessary for redesigning complex systems of care. Participants at the workshop suggested that professional societies develop guidelines devoted specifically to patient safety and the incorporation of patient safety considerations into other guidelines.
One of the most visible activities of professional groups is their convening function. Through annual conferences and specialty meetings, professional groups can develop and communicate standards, values, and policy statements to membership and key opinion leaders. Meeting conclusions may also be disseminated through their own and other journal publications. There are few examples of specialty meetings or conferences where patient safety has been explicitly included on the agenda. Additionally, there are few interdisciplinary conferences devoted to issues of patient safety. Participants at the workshop proposed a national conference that would bring together all health professions and professionals from other disciplines (e.g., industrial engineering, human factors analysis) and other industries (e.g., airline pilots).
Clinical training and education is a key mechanism for cultural change. Colleges of medicine, nursing, pharmacy, health care administration, and their related associations should build more instruction into their curriculum on patient safety and its relationship to quality improvement. One of the challenges in accomplishing this is the pressure on clinical education programs to incorporate a broadening array of topics. Many believe that initial exposure to patient safety should occur early in undergraduate and graduate training programs, as well as through continuing education. Clinical training programs also need to ensure that teaching opportunities are safe for patients. One workshop participant told of a monitoring device used to alert staff to possible problems with the patient that was turned off because it was seen as interfering with the teaching experience.
The need for more opportunities for interdisciplinary training was also identified. Most care delivered today is done by teams of people, yet training often remains focused on individual responsibilities leaving practitioners inadequately prepared to enter complex settings. Improving patient safety also requires some understanding of systems theory in order to effectively analyze the many contributing factors that influence errors. Again, the "silos" created through training and organization of care impede safety improvements. Instruction in safety improvement requires knowledge about work-
ing in teams, using information and information technology, quality measurement, and communicating with patients about errors. A background in other disciplines is also relevant, such as cognitive psychology, systems theory, and statistics.28 Principles of crew resource management used to train personnel who work together in airline cockpits might also be applicable to health care. Training should also emphasize better communications across disciplines. This is important when the members of a care team are in one physical location, such as a hospital or office setting, but becomes even more important when the care team may not be in one place, such as a team providing home care.
Few professional groups have sufficient resources to devote to research support, although many have established research and education foundations. The need for greater collaboration in developing regional databases was noted. A key advantage of establishing these at the regional level is the ability to obtain a sufficient number of cases for meaningful analysis. The number of cases of any particular event in a single hospital or clinical setting is usually too small to be able to generalize across cases and identify a way to make system improvements. Regional data systems can increase numbers to improve analytic power and can facilitate collaboration to understand the extent and nature of errors in health care. Professional societies and groups could participate in efforts to coordinate a research agenda and the development of databases to provide information on the extent and nature of errors in health care.
Professional groups can also serve as advocates for change. Professional groups have been able to call attention to a health risk and create awareness. For example, pediatricians have been active in promoting increased immunization rates, the American Heart Association has promoted diet and exercise to prevent heart disease, and the American Medical Association (AMA) has been an outspoken opponent against smoking. Professional groups have not been as visible in advocating for patient safety and communicating such concerns to the general public and policy makers. A notable exception has been the formation of the National Patient Safety Foundation (NPSF) by the AMA in 1997 (see Chapter 4). The NSPF has taken a visible role in advocating for improvements in patient safety and communicating with a broad array of audiences. Professional societies can play a role not only in informing their members about patient safety, but also in calling attention to the issue among the general public.
Implementation of activities to increase the role of health professionals in patient safety must occur at multiple levels. Although some professional
groups influence and communicate with just their own members, other groups have the potential to influence many audiences. For example, the American Board of Medical Specialties has the potential to influence 24 professional medical societies. The Accreditation Council for Graduate Medical Education and the American Association of Colleges of Nursing have the potential to influence numerous training programs. The Association of American Medical Colleges can influence multiple medical schools and academic medical centers. There are many other similar groups that coordinate across multiple organizations. These "high leverage" groups are critical players in encouraging action among their constituent organizations. They should use their influence to promote greater awareness of patient safety and to consistently reinforce its importance.
Standards for Drugs and Devices
The Food and Drug Administration is a major force in setting standards for medical products and monitoring their safety. FDA regulates prescription and over-the-counter drugs, medical and radiation-emitting devices, and biologics, among other things. This discussion focuses on its activities related to drugs and devices. It should be noted, however, that the FDA regulates manufacturers, not health care organizations or professionals. There are two opportunities for FDA to ensure and enhance patient safety: during its approval process for drugs and devices, and through postmarketing surveillance.
FDA has regulatory authority over the naming, labeling, and packaging of drugs and medical devices. FDA approves a product when it judges that the benefits of using the product outweigh the risks for the intended population and use.29 For drugs, the approval process examines evidence of the effectiveness of the drug and the safety of the drug when used as intended. For devices, FDA looks at the safety and effectiveness of the device compared to devices already on the market or else looks for reasonable assurance of safety and effectiveness.
A major component of postmarketing surveillance is conducted through adverse event reporting.30 Reports may be submitted directly to the FDA or through MedWatch, FDA's reporting program. For medical devices, manufacturers are required to report deaths, serious injures, and malfunctions to FDA. User facilities (hospitals, nursing homes) are required to report deaths to both the manufacturer and FDA, and to report serious injuries to the manufacturer. For suspected adverse events associated with drugs, report-
ing is mandatory for manufacturers and voluntary for physicians, consumers, and others. All reports are entered into the Adverse Event Reporting System (AERS) or another database, which is used to identify problem areas or increased incidence of an event.
FDA receives approximately 235,000 reports annually for adverse drug events and approximately 80,000–85,000 reports on device problems. Despite the extensive testing that FDA requires before drugs and devices are approved, side effects or other problems invariably show up after they have been released and used widely. Not all risks are identified premarketing because study populations in premarketing trials are often too small to detect rare events, studies may not last long enough to detect some events, and study populations may be dissimilar from the general population.31 Some of these initially unknown risks can be serious or even fatal. The problem is likely to continue and possibly worsen in the future because of the number of new drugs being introduced. In 1998 alone, FDA approved 90 new drugs, 30 new molecular entities (drugs that have never been marketed in this country before), 124 new or expanded uses of already approved drugs, 344 generic drugs, 8 over-the-counter drugs, and 9 orphan drugs, or almost two actions every day of the year.32 Approximately 48 percent of the prescription drugs on the market today have become available only since 1990.33 Medications are also the most frequent medical intervention, with an average of 11 prescriptions per person in the United States.34
FDA has three general strategies it pursues for corrective action. The first (and most commonly pursued) is negotiation with the manufacturer to make the desired changes. The extent of cooperation from the manufacturers can vary. In terms of drugs, names are the most difficult to change, particularly once a name has been trademarked by the company (Jerry Phillips, OPDRA, personal communication, May 4, 1999). Second, FDA may take regulatory action against manufacturers to require changes. This could include name changes or withdrawal of a product from the market. The final type of action that FDA can take is communication about risks, including letters to physicians, pharmacists, and other health professionals, postings on the Internet, and publication of clinical and consumer journals. FDA decisions about corrective action are made on a case-by-case basis, by considering the unexpectedness and seriousness of the event, the vulnerability of the population affected, and the preventability of the event.35
Some concerns have been expressed over the responsiveness of FDA to reported problems. Concerns have related to the timeliness and effectiveness of the agency's response or that the response to a given problem may
not be strong enough given its seriousness. For example, five drugs were removed from the market in between September 1997 and September 1998, but almost 20 million people had been exposed to their risks before they were removed.36 Terrenadine was on the market for 12 years, even though researchers earlier identified it as causing deaths; it was removed from the market by the manufacturer only after a substitute was developed.37
There have been calls for better methods for obtaining more information about the harm caused by drugs (e.g., greater use of active surveillance systems that look for indicators of problems rather than waiting for reports to be submitted) or for the establishment of an independent drug safety review board.38 In the fall of 1998, FDA changed the process for follow-up on reported drug problems with the creation of a new Office of Post-Marketing Drug Risk Assessment (OPDRA). Before, incidents were reviewed by a committee, triaged, and sent back to the division that did the original review. This dispersed responsibility for review and follow-up led to variability in response. Now, OPDRA will conduct an analysis of all reported events and develop recommendations that are sent to the manufacturer and the director of the FDA division that conducted the original review. The division director must report to OPDRA in 60 days on the status of the recommendations. OPDRA estimates that approximately half of the causal factors that contribute to adverse events are issues to which it can respond (e.g., labeling problems); the remainder are outside its scope (e.g., bad handwriting) (Jerry Phillips, OPDRA, personal communication, May 4, 1999).
With regard to medical devices, in recent years, FDA has increased its requirements and guidance to manufacturers on designing devices to take into account human factors principles and user testing. Attention to human factors could improve simplicity of use, standardization of controls, and default to a safe setting during failure (e.g., loss of power). For example, intravenous infusion pumps vary markedly in their mode of operation and types of controls. Because they are expensive, hospitals do not replace old pumps when new ones become available, which results in different models being used. The lack of standardization among the models increases the likelihood of error when the pump is set up. Controls on defibrillators can also vary in position, appearance, and function on different machines, leading to errors when they are used rapidly in emergency situations. Although the increased attention to human factors principles does not affect devices already on the market, over time it is expected that manufacturers will become more accustomed to using human factors in the design of medical devices.
With the passage of the Safe Medical Device Act of 1990, FDA was
granted the authority to require manufacturers of medical devices to establish and follow procedures for ensuring that device design addressed the intended use of the device and its users.39 Final rules for this act became effective in June 1997. FDA has continued to emphasize to manufacturers the importance of human factors and is expected to issue a manual of engineering and design guidelines for manufacturers in 1999.
In terms of drugs, the use of human factors principles could reduce confusion of medications that occur because of brand names that look alike or sound alike, labels that are hard to read, and look-alike packaging. Wrong doses also occur frequently because of factors such as the lack of standardized terms in the display of contents. For example, contents displayed by concentration (e.g., 10 mg/mL) rather than total amount (e.g., 100 mg) can result in an overdose. There may also be inconsistent placement of warnings on a label or inconsistent use of abbreviations. Most recently, more than 100 errors have been reported in the use of Celebrex (prescribed for arthritis) and its confusion with Cerebyx (an antiseizure medication) and Celexa (an antidepressant).40 FDA does not have guidance for using human factors principles in the packaging, labeling, or naming of drugs as exists relative to medical devices.
The main sources of standards for health care organizations and professionals today are through licensing and accreditation processes. However, medical errors and patient safety are not an explicit focus of licensing and accreditation. Although licensing and accreditation standards do speak to the characteristics of quality improvement programs, and patient safety and error reduction may be part of these programs, many licensed and fully accredited organizations have yet to implement the most rudimentary systems and processes to ensure patient safety. Furthermore, the extent of variation in licensure within and across states suggests that there is no reliable assurance of safety to patients, even for those facilities and professionals covered under current rules.
Although current standard-setting authorities in health care are not devoting adequate attention to patient safety issues, the committee considered and rejected the option of recommending the creation of yet another regulatory authority. The recommendations contained in this chapter direct the existing regulatory structures to increase attention to patient safety issues. Licensing agencies and accrediting organizations have to hold health care
organizations accountable for creating and maintaining safe environments. Professional licensing bodies should consider continuing qualifications over a lifetime of practice, not just at initial licensure. Standards for approving drugs and devices must consider safety for patients in actual use and real-life settings, not just safe production.
The actions of professional groups and group purchasers in setting standards and expectations are also critical. Professional groups shape professional behavior by developing practice guidelines and identifying best practices and through educational, convening and advocacy activities. All could be enhanced by a sharper focus on patient safety issues. Group purchasers have the ability to consider safety issues in their contracting decisions, and to reinforce the importance of safety by providing relevant information to employees and beneficiaries.
1. Institute of Medicine, Clinical Practice Guidelines, Directions for a New Program, eds. Marilyn J. Field and Kathleen N. Lohr, Washington, D.C.: National Academy Press, 1990.
2. ''All About OSHA," OSHA 2056 1995 (revised), Occupational Safety and Health Administration, Department of Labor, www.osha.gov.
3. "Joint Commission Accreditation," http://www.jcaho.org/whoweare_frm.html.
4. "NCQA Accreditation," Business and Health, 17(1):21 (January, 1999).
5. "About URAC," http://www.urac.org.
6. General Accounting Office. Private Health Insurance: Continued Erosion of Coverage Linked to Cost Pressures. GAO/HEHS 97–122. Washington, DC: July 1997b.
7. Hibbard, Judith H.; Jewett, Jacquelyn J.; Legnini, Mark W., et al. Choosing a Health Plan: Do Large Employers Use the Data. Health Affairs. 16(6):172–180, 1997.
8. Washington Business Group on Health and Watson Wyatt Worldwide. Getting What You Pay for—Purchasing Value in Health Care. Bethesda, MD: Watson Wyatt Worldwide, 1997.
9. KPMG Peat Marwick. Health Benefits in 1997. Montvale, NJ: 1997.
11. "Quality of Care Information, HCFA Contractors, Peer Review Organizations," http://www.hcfa.gov/qlty-56.htm, last updated July 26, 1999.
12. Timothy Cuerdon, Office of Clinical Standards and Quality, Health Care Financing Administration, Testimony to the IOM Subcommittee on Creating an External Environment for Quality, June 15, 1999.
13. "Medicare and Medicaid Survey and Certification," http://www.hcfa.gov/medicaid/scindex.htm
14. O'Neil, Edward H., and the Pew Health Professions Commission, San Francisco, CA: Pew Health Professions Commission, December 1998.
15. Finocchio, Leonard J.; Dower, Catherine M.; Blick, Noelle T., et al., Strengthen-
ing Consumer Protection: Priorities for Health Care Workforce Regulation, San Francisco, CA; Pew Health Professions Commission, October 1998.
16. Maintaining State-Based Licensure and Discipline: A Blueprint for Uniform and Effective Regulation of the Medical Profession, Federation of State Medical Boards of the United States, Inc., 1998. http://www.fsmb.org.
17. Sidney M. Wolfe, "Public Citizen's Health Research Group Ranking of State Medical Boards' Serious Disciplinary Actions in 1998," April, 1999, http://www.citizen.org. Serious disciplinary actions include revocations, surrenders, suspensions and probations/restrictions on licensure.
18. Timberg, Craig, "Virginia's Physician Discipline Board too Slow to Act, Audit Finds," Washington Post, June 15, 1999, p. A20.
19. Finocchio, Dower, Blick, et al., 1998.
20. Finocchio, Dower, Blick, et al., 1998.
21. "What Is ABMS?" http://www.abms.org/purpose.html.
22. Prager, Linda O. Upping the Certification Ante. American Medical News. 42(21): 1, 1999. Also, American Board of Internal Medicine, Continuing Professional Development. September, 1999. See also: Norcini, John J., "Computer-Based Testing Will Soon Be a Reality," Perspective, American Board of Internal Medicine, Summer, 1999, p. 3.
23. "A Definition," http://www.ama-assn.org/med-sci/amapsite/about/define.htm.
24. Finocchio, Dower, Blick, et al., 1998.
25. Computerized Clinical Simulation Testing (CST), National Council of State Boards of Nursing, Inc., 1999. http://www.ncsbn.org.
26. Pierce, Ellison C. The 34th Rovenstine Lecture, 40 Years Behind the Mask: Safety Revisited. Anesthesiology. 87(4):965–975, 1996.
28. President's Commission on Consumer Protection and Quality in Health Care. Quality First: Better Health Care for Ali Americans. Final Report to the President of the United States, March 1997.
29. Food and Drug Administration, "Managing the Risks From Medical Product Use, Creating a Risk Management Framework," Executive Summary, Report to the FDA Commissioner from the Task Force on Risk Management, May 1999.
30. Additional strategies include field investigations, epidemiological studies and other focused studies.
31. Food and Drug Administration, "Managing the Risks From Medical Product Use, Creating a Risk Management Framework," Report to the FDA Commissioner from the Task Force on Risk Management, U.S. Department of Health and Human Services, May 1999.
32. Food and Drug Administration, "Improving Public Health Through Human Drugs," CDER 1998 Report to the Nation, Center for Drug Evaluation and Research, USDHHS.
33. Shatin, Deborah; Gardner, Jacqueline; Stergachis, Andy; letter. JAMA. 281(4): 319–320, 1999.
34. Friedman, Michael A.; Woodcock, Janet; Lumpkin, Murray M., et al. The Safety of Newly Approved Medicines, Do Recent Market Removals Mean There Is a Problem? JAMA. 281(18):1728–1934, 1999.
35. Food and Drug Administration. Managing the Risks From Medical Product Use, Creating a Risk Management Framework. Report to the FDA Commissioner from the Task Force on Risk Management, U.S. Department of Health and Human Services, May 1999.
36. Friedman, Woodcock, Lumpkin, et al, 1999. Also Wood, Alastair J.J. The Safety of New Medicines. JAMA. 281(18):1753–1754, 1999.
37. Moore, Thomas J; Psaty, Bruce M; Furberg, Curt D. Time to Act on Drug Safety. JAMA, 279(19):1571–1573, 1998.
38. Moore, Psaty, and Furberg, 1998. See also Wood, Alastair J.J. and Woosely, Raymond. Making Medicine Safer, The Need for an Independent Drug Safety Review Board. N Engl J Med. 339(25):1851–1853, 1998.
39. Weinger, Matthew; Pantiskas, Carl; Wiklund, Michael, et al. Incorporating Human Factors Into the Design of Medical Devices. JAMA. 280(17):1484, 1998.
40. Look Alike/Sound Mike Drug Names, Ambiguous or Look-Alike Labeling and Packaging. ISMP Quarterly Action Agenda: April–June, 1999, ISMP Medication Safety Alert, July 14, 1999. Institute for Safe Medication Practices, Pennsylvania.