Skip Navigation Archive: U.S. Department of Health and Human Services U.S. Department of Health and Human Services
Archive: Agency for Healthcare Research Quality
Archival print banner

This information is for reference purposes only. It was current when produced and may now be outdated. Archive material is no longer maintained, and some links may not work. Persons with disabilities having difficulty accessing this information should contact us at: Let us know the nature of the problem, the Web address of what you want, and your contact information.

Please go to for current information.

Section 6

An Agenda for Research in Ambulatory Patient Safety

Synthesis of a multidisciplinary conference to develop an agenda for research in ambulatory patient safety.

Expansion of Governmental and Regulatory Attention to Ambulatory Care Safety, Including Expansion of Mandatory Reporting Requirements

As reflected by the Florida outpatient surgery cases cited in the section above, media attention to unexpected adverse outcomes often triggers regulatory action. In the current iteration of the national patient safety discussion, this factor has played out to date primarily in discussion of increased mandatory reporting of adverse events, the role of State regulators in investigating such incidents, and the importance of interagency coordination among regulators (American College of Surgeons, October 2000).

Interestingly, the Massachusetts regulatory regime seems to be ahead of the curve in exploring and implementing regulatory responses that might support a systems approach to advancing patient safety. For example, the State's Board of Registration in Medicine has implemented a mandatory adverse event reporting system that de-identifies the individual professionals involved in an incident and is separated by a firewall from the same agency's physician disciplinary function. This approach reflects an understanding that mandatory reporting schemes are not likely to be fully complied with if discipline is a potential action by the agency mandating reports. Yet, it underscores the reluctance of regulators to rely on industry self-regulation in addressing internal quality or safety problems. This blended approach protects individuals from blame or punishment, yet provides an avenue for greater investigation or monitoring by the State of provider risk management initiatives (see also NASHP 2001, Barach and Kelly 2001).

Massachusetts also developed in 2000 a legislative proposal to establish a central focus on patient safety in the executive branch of government, with significant powers to coordinate the activities of all agencies with roles in ensuring patient safety, funding research and setting priorities. Proposals for patient safety centers to coordinate governmental activities also have been introduced in other States (NASHP 2000). In the Federal arena, Secretary of Health and Human Services Tommy G. Thompson announced in April 2001 the establishment of a new interagency Patient Safety Task Force to coordinate a joint effort among several department agencies to improve existing systems to develop an integrated reporting/data collection effort on patient safety. The Federal agencies leading this effort include the Agency for Healthcare Research and Quality (AHRQ), the Centers for Disease Control and Prevention (CDC), the Food and Drug Administration (FDA), and the Centers for Medicare and Medicaid Services (CMS, formerly the Health Care Financing Administration). An earlier interagency coordination effort, the Quality Interagency Coordination Task Force (QuIC), established in 1998, continues to meet with the goal of ensuring that all Federal agencies involved in purchasing, providing, studying, or regulating health care services are working in a coordinated way toward the common goal of improving the quality and safety of care (QuIC 2000).

Assuming that media attention of ambulatory care incidents will continue and probably increase as the parameters of safety challenges in ambulatory care are better understood, it is likely that governmental and regulatory attention in that sector also will increase. As noted elsewhere in this paper, in October 2001, AHRQ announced the award of 94 research and pilot project grants focused on patient safety, several of which addressed issues in ambulatory care. The innovative approaches being developed at the Federal level reflect a policy-setting and regulatory community that is moving towards alignment with principles of a systems approach to safety and accountability. It is not unfounded to speculate that the leadership profile in Massachusetts, among the States, was expedited by several of the external environmental factors discussed in this paper, including media scrutiny following the Betsy Lehman event and the coming together of an active coalition of diverse stakeholders.

Expansion of Accreditation and Certification Activity

The same influences prompting increased regulatory activity also are driving expanded accreditation requirements imposed on ambulatory care facilities and the physicians and physician extenders practicing in them. Conference participants discussed several recently implemented accreditation initiatives, developed by a variety of accreditation bodies working in this field. These initiatives, highlighted in earlier sections, are being fueled not only by public concern over the lack of rigorous accreditation and oversight in ambulatory care, but by the development of new standards by professional societies such as the American College of Surgeons (American College of Surgeons 2000). As new evidence about ambulatory care accumulates, and new guidelines based on that evidence are developed based on such evidence, increased accreditation, certification and recertification requirements are likely to expand accordingly (American College of Surgeons, October 2000).

Expansion of Tort Liability

Because the standard of care applied by the courts in professional liability litigation is defined by community practice, it evolves with community practice. As new evidence-based practices are produced by researchers and adopted into practice, they will be incorporated into the standard of care. The emphasis in the healthcare community on expediting applied research and evidence-based practice probably translates into a more rapid ratcheting up of the standards applied by the courts. Conversely, practice in accordance with evidence-based practice research or guidelines can be expected to strengthen the ability to successfully defend a claim generated by an adverse outcome.

Other factors impacting the standard of care applied by the courts include what might be called administrative or management "best practices" and a clearly documented shift in the doctrine of informed consent. Examples of administrative best practices include new duties to proactively follow up with patients who miss appointments or do not return telephone calls. The shift in informed consent doctrine, now adopted in most States, is away from a standard determined by what a reasonable physician would tell the patient and toward a standard defined by what a reasonable patient would want to know (AMA/SSMLP 1990). Both of these shifts create new duties (and new opportunities for failure to satisfy these duties) that may impact ambulatory care settings to a greater degree than acute care settings, where noncompliant patients can be more effectively controlled and communication/understanding gaps between provider and patient more easily addressed.

New standards implemented by the JCAHO in July 2001, requiring disclosure of untoward outcomes to patients also may play a role in modifying liability exposure (JCAHO 2001). While many commentators argue that such disclosure has always been a requirement mandated by professional ethics, efforts to manage liability risk in practice often have led to attitudes or practice that discouraged disclosure (Wu et al. 1997). Advocates of disclosure contend that it will not generate new litigation and may discourage lawsuits brought by those who are angry or primarily interested in fully understanding what led to an unexpected adverse outcome (Wu 1999, Kraman and Hamm 1999). The research on these questions is not definitive at this time.

Expansion of State and Federal Legislative Activity

In the wake of the IOM's 1999 Report, To Err is Human, numerous responses to the patient safety problem were introduced in Federal and State legislatures. State legislative efforts to establish patient safety offices or centers in government already have been addressed in an earlier section of this paper. Other bills citing patient safety improvement rationales addressed such issues as prevention of retaliation for whistle-blowing, physician-profiling programs, nursing shortage concerns, among others (NASHP 2000).

In the Federal arena, the most notable objective of numerous patient safety bills introduced in the 106th Congress (see AppendixThree) was the establishment of Federal confidentiality protections for adverse event reports issued by healthcare organizations to government or to accreditation bodies for quality or safety improvement purposes (Parsons 2000). If enacted these bills would expand to varying degrees State law protections for peer review proceedings, shielding the information generated for such proceedings from use by disciplinary authorities or in professional liability litigation. Modeled somewhat after efforts in the aviation safety community to create a legal infrastructure for an effective national reporting system, the patient safety bills enjoyed significant bipartisan support when initially introduced in 1999 and 2000. However, they have not advanced significantly in the Federal legislative process in the 107th Congress, despite high hopes of the patient safety community for continued bipartisan support, particularly in the United States Senate. Any momentum the bills may have had seemed to stall in the summer of 2001. Most of the Federal bills introduced in 1999 and 2000 have not been reintroduced as yet in the 107th Congress.

The lack of rapid progress in moving legislative proposals forward underscores the importance of efforts to develop shared understandings among stakeholders and, where possible, consensus. Without a shared understanding of objectives and the basics of safety science approaches to managing apparent and latent risk, reporting may continue to be scanty and opportunities lost to improve patient safety. Should there be mandatory reporting, voluntary reporting, or both? By whom and to what? With what kinds of protection for confidentiality and privacy, with what privilege from discovery and protection from liability? What will be required to enable learning improvement of patient safety? What is the potential for consensus around the position that information developed in connection with reporting systems should be privileged for purposes of Federal and State judicial proceedings and civil matters, and for purposes of Federal and State administrative proceedings, including with respect to subpoenaed materials, testimony, and other forms of legal discovery? The ability to effectively address these policy questions relies on achieving some shared belief that preventing harm is a goal more fundamental than punishing those who make treatment errors. That degree of consensus among stakeholder leaders continues to be a challenge impeding systems-based approaches to patient safety in the legislative arena.

Expansion of Public and Private Payor Activities and a Business Case for Safety

The organization and delivery of health care in the United States is inextricably related to and driven by its financing. Conference participants emphasized that the financial incentives provided by both public and private purchasers of health care services will help determine rates of change and improvement in patient safety. This would appear to be particularly true in the absence of rapid legislative reform addressing the perceived obstacles to patient safety improvement in our legal system and regulatory infrastructure. In contrast to State and Federal legislatures, purchasers have moved forward rapidly to use their leverage to initiate patient safety improvement. In November 2000, the Leapfrog Group consortium of large public and private payors rolled out a series of specific interventions targeted to achieve quick gains in safety and reliability in acute care settings (Leapfrog Group 2000, Milstein et al. 2000). Leapfrog spokespersons have characterized these interventions as a first wave. Additional targets are being developed that may apply more directly to ambulatory care settings. The Washington Business Group on Health (WBGH) has identified medication error in prescribing, dosing, dispensing and use across both acute and ambulatory delivery settings as a high priority for their members (WBGH 2000).

For ambulatory care, it is apparent that there are presently few financial and economic incentives aimed at increased reporting, sharing of lessons learned, expanded disclosure to patients experiencing untoward outcomes or other improvements recommended by the IOM. The majority of ambulatory physician practices have little incentive to add either capital or labor costs to their practices in order to improve patient care. This has been a continuing problem with the dissemination and adoption of evidence-based quality improvement technologies and techniques in ambulatory care over the last decade.

However, there are indications that this situation is changing. Employers are negotiating contracts with HMOs through direct contracting that include incentives for measuring and improving quality of care. It is clear that these incentives must be present and significant if they are to induce real changes in practice and provider behavior as they relate to patient safety and quality of care.

Conference participants repeatedly cited the need to develop and test ways to make a "business case" for improving patient safety, also expressed as the need to find ways for those who invest time and resources to improve to also receive the (financial and other) benefits of the improvements. As noted above, this can only be accomplished if changes in reimbursement can be designed to induce changes in patient safety related practices. This includes measuring and documenting improvements in the quality of patient care and the reduction of medical error and funding to pay for structural, technological and personnel related changes in ambulatory practice and patterns of patient care (Bates, Cohen et al. 2001; Institute of Medicine 2001, Chapter 8).

There is some published evidence, mostly from studies showing cost reductions in specific areas of care, that improving patient safety and quality can reduce costs of health care (e.g., Classen et al. 1997, Clemmer et al. 1999). Many believe that potential cost savings will prove to be substantial, perhaps as much as 30 or 40 percent, based on unpublished and anecdotal reports, and experiences in other industries. If so, resources saved could be used to improve access to care for underserved populations, to fund additional medical research, to provide new effective therapies, and to reduce costs to patients, employers, and the public. However, it is not realistic to expect effective changes in practice patterns if major changes in payment for care are not made.

The business case for improvement must be made around payment and an improved rate of return on investment through savings generated by lowering the cost of rescue and follow-up care to patients injured during the care process. It was suggested that in prepaid, managed care settings, the effects of improvement can be more easily documented than in a fee-for service setting. This demonstration of the cost-effectiveness of patient safety improvement is necessary if sustainable improvements in the system are to be made. Therefore, there need to be changes in the CPT driven RBRVS system that ties specific procedures and practices to patient safety. This system alone, through Medicare and Medicaid, accounts for a large proportion of the expenditures for health care in the United States.

It may be easier to develop a business case for safety in more nearly closed systems such as Kaiser Permanente, or the Harvard Medical Institutions (with the Risk Management Foundation and CRICO), or in a health system that has a very high market share in a community (Intermountain Health Care in Utah, perhaps). In selected situations, it may be possible to assemble a coalition that is able to internalize costs and benefits from purchasers, payers, and providers.

We also need to understand the full range of indirect as well as direct costs of patient injuries due to medical management. We need to understand the costs of changes to improve quality and safety of ambulatory care (and who bears them) and the benefits and to whom they accrue. In some cases, the entity that invested the time and resources to improve patient safety is not the entity that benefits in terms of reduced costs of, for example, avoided treatment and hospitalizations. For example, improving safety of office practice may reduce costs of drugs and hospitalizations for patients, but these savings will accrue not to the practice, but to the insurance company (Institute of Medicine 2001, Chapter 8). Further, it is not clear that entities that provide safer care are recognized or rewarded in any way for better safety (for example, by attracting more patients), though the large employers in the Leapfrog Group indicate the intent to change that. How can we enable more of the benefits to flow to those who bear the costs? Can we make it less expensive or easier to do it the safer, better way? These are the questions that need to be addressed.

Recommendation 11: Model the financial implications of improving safety and quality for various players in the current organization and financing of care, including physicians, medical group practices, hospitals, integrated care systems including staff model group practices that provide prepaid care to populations, traditional health insurers, self insured employers, patients, and the Medicare program. Explore the effects of adverse selection. Develop potential modifications in or alternatives for organization and financing that would align interests to support and reward the various professionals and organizations to improve safety and quality.

Expansion of Technological Tools to Advance Patient Safety

Several sections of this paper discuss the role of information management tools and other innovations in technology that have the potential to improve systems operation and prevent failures that result in patient harm. The standardization and customization challenges in effectively utilizing technological innovation in ambulatory care have already been highlighted. For purposes of discussion in this section, it is important to acknowledge the impact of such innovation on the external environment. The Leapfrog Group and its assemblage of private and public purchasers already are using their leverage to drive implementation of CPOE in hospitals. As evidence accumulates, it is likely that parallel incentives for adoption of CPOE, handheld information technology, etc. will be established in the ambulatory care arena as well. Notably, the October 2001 AHRQ patient safety awards of grant funding will underwrite at least two projects designed to measure the effectiveness of handheld information management tools in ambulatory care settings.

It should also be noted that Leapfrog Group standards or other tools or practices that are presented as evidence-based will mostly likely be assimilated into the legal standard of care. Ambulatory care providers who do not develop strategies to implement such innovation may be disadvantaged in both the market and the courts.

Health Care Professional Education

Conference participants opined that we should expect that education in the basics of safety science will appear across the spectrum of care and be subject to accreditation and licensure requirements (O'Leary 2000). Education and training at all levels should include core concepts on systems effects, human factors, complex problem solving in the field, continuous improvement, and safety culture (Barach and Moss 2001, Institute of Medicine 2001 chapter 9). Education should rely in part on behavioral change principles such as case based learning, small peer discussion, video feedback, simulation, and team training.

The question is how quickly can we ramp up to achieve these educational objectives? Optimally, these efforts need to occur early in health professional education and be designed to break down barriers to collaboration horizontally (across disciplines) and vertically (in the chain of command).

Expansion of Patient Education and Engagement Activities

As noted in other sections of this paper, responsibility for improving patient safety is particularly widely shared in ambulatory care settings, and extends to patients and families of patients as well as clinicians. An effective systems approach then must look at the systems that do or could support patients/families and synergies with the support systems and tools developed for ambulatory care providers.

Patients are beginning to interact with other stakeholders in the patient safety community, to climb the learning curve with respect to the basics of safety science, and to develop new pathways for engagement. Patient and family use of information provided by employers while selecting providers is a core component of the Leapfrog Group initiative. Leapfrog Group members also have access to training materials for employees designed to assist them in building skills needed to be a more effective partner in the treatment process (Leapfrog 2000). Patients are active members of patient safety coalitions in Massachusetts and elsewhere, and participate regularly on the National Patient Safety Foundation's listserv. In June 2001, the consumer group Public Citizen joined with American Medical Student Association to petition the Occupational Safety and Health Administration to limit resident work hours, citing as a reason the likelihood of increased human errors as a function of fatigue (Preidt 2001). In May 2001, the JCAHO issued its first Sentinel Event Alert prompted by consumer reports of brain injuries to neonates as the result of failure to diagnose or treat post-discharge neonatal hyperbilirubinemia (JCAHO Sentinel Event Alert 2001).

This last example underscores that reporting and data gathering systems designed to optimally surface information about emerging risk in ambulatory care should not ignore the role patients and families can play as data contributors. Fears of liability exposure on the part of providers and the failure of our legal and insurance systems to develop mechanisms for systematically compensating victims of systems failure continue to be massive obstacles to effective provider/patient working relationships. However, consumer pressure for more disclosure of information and more effective working relationships may be a potent new argument for reform of the liability system (Hatlie 2001).

Summary and Conclusion

While the interdependent factors in the external environment explored above are not under the control of individuals or organizations in the healthcare sector, they can be influenced by effective leadership and communication strategies. Lessons from other industries underscore the importance of building stakeholder relationships based on the common and fundamental value of minimizing avoidable human injury. Effective strategies to disseminate safety science through diverse stakeholder groups can be built on a platform of shared values. Commenting on the aviation safety community's efforts to advance legislation supporting aviation safety reform, a leader in that movement, Charles Billings, M.D., observed (Billings 1998):

"One reason that an incident reporting system worked in aviation...was that the entire aviation community—essentially all of the stakeholders, including air passengers—were involved in the process from the beginning and became advocates for the reporting system (as well as severe, but constructive critics)."

Effective and trusting stakeholder engagement is arguably more important to the advancement of ambulatory care safety than safety for inpatient care, because the need for coordination of information and care in the ambulatory setting is more complex and there is a wider distribution of responsibility. Engagement of the external stakeholders appears to be a crucial step toward alignment of incentives and support for a systems approach to ensuring the safety of care and minimization of avoidable patient harm in the ambulatory care setting.

Return to Contents


1 This project was supported by grant number R13 HS10106 from the Agency for Healthcare Research and Quality (AHRQ), with additional support from the Centers for Medicare and Medicaid Services (CMS) and with assistance from the Partnership for Patient Safety (p4ps). AHRQ and CMS not only provided funding for this conference, but staff from both AHRQ and CMS also collaborated with MGMA and p4ps in planning the conference. A number of staff from AHRQ and CMS participated in small group discussions (see list of participants in Appendix One), and made valuable contributions as facilitators and discussants.

2 Appendix One to this report is a list of participants.

3 For a description of activities of a number of Federal agencies, see the report of the Quality Interagency Coordination Task Force of the U.S. Government (2000).

4 In that sense, this report is complementary to others that focus principally on particular interventions (e.g., Shojania et al. 2001).

5 Some conference participants would go further, pointing out that deficiencies in knowledge and skills of an individual clinician or other person are in part systems issues. Shared responsibility of health care organizations is recognized through organizations' responsibility for credentialing, privileging, continuing education, and training. But for much of ambulatory care, there are few requirements or mechanisms for credentialing and privileging beyond basic licensing.

6 According to Bates, roughly half the medication errors they find in ambulatory care involve differences between what patients do and what physicians expect them to do (David Bates, comment in conference discussions, December 1, 1999). Differences between physicians' expectations and patients' actions may be explained by financial barriers, patients not understanding physicians' explanations for their recommendations, and other factors. Differences between inpatient and outpatient care may be in part due to less reliable processes and systems for communication and coordination in outpatient care.

7 As articulated by Donald M. Berwick, M.D., M.P.P., President and CEO of the Institute for Healthcare Improvement, "We are carrying the 19th century clinical office into the 21st century world. It's time to retire it. We are on the threshold of remaking entire care systems, beginning with the clinical office itself—and the courageous among us will get there first. The changes we face have never been bigger...or more exciting."

8 The precise definition may not be particularly important. Some would include long term care facilities, and perhaps any care that is not provided to hospital inpatients. Also, because there are many (risky) links between ambulatory care and hospital care, we need to include interactions among various components of care including with the hospital.

9 Diagnostic endoscopy of the small and large intestines is the most frequently performed procedure (Hall and Lawrence 1998).

10 The most frequent ambulatory surgery procedures were for cataracts (2.5 million), malignant neoplasms (800,000), rheumatoid disease (600,000) and benign neoplasms (600,000) (Hall and Lawrence 1998).

11 A fourth of the events resulted in death or permanent disability in the Australian study, and two thirds were judged preventable.

12 Such a large study of what is happening in ambulatory health care would yield information on what the occurrences/adverse events/injuries and errors in non-hospital settings, to whom they are occurring, in which settings, associated with which procedures, associated with which caregivers or characteristics of caregivers, etc. That knowledge would be very useful to focus research and to set priorities. Some conference participants suggested that without such data, funding for research in ambulatory patient safety would be given lower priority than research in other areas where the magnitude of the problem can be better documented.

13 Rothschild and Leape (2000) note that the shift from inpatient to outpatient care induced by managed care has made it more difficult to identify and document adverse events.

14 Where integrated electronic medical records are in place, integrating information across care sites and providers, information would presumably be more complete and amenable to research.

15 See the Methods section above for an explanation of the evidence that each of the 11 recommendations represent consensus of the conference participants.

16 Both risks and checks and balances to manage risk may differ by size and type of ambulatory facility, by size of physician group practices, close association with a hospital or integrated system, and whether there is increased regulation, as for ambulatory surgicenters and office based surgery, in some States.

17 One technique is to identify observable events that imply a high likelihood of an adverse event having happened: administration of vitamin K, Benadryl, or Narcan; transfers from ambulatory facilities to an Emergency Department or hospital (Classen et al. 1992, Jha et al. 1998). Other events that may be associated with increased risk include missed or canceled appointments, refills of prescriptions more frequently than normal, and multiple new prescriptions.

18 Case studies by liability insurers combine investigation from a clinical and systems point of view with a legal point of view. Was the patient injured? What were the factors that caused and contributed to the injury? Were mistakes made that contributed to the injury? Is there evidence of substandard practice and therefore potential malpractice liability? In the experience of the St. Paul Companies, as the length of time between the claim and the event leading to that claim lengthens, it becomes more difficult to get information about the events leading to the incident identified in the claim, as well as related information from other sources (personal communication from Dorothy Berry, conference participant, The St. Paul Companies, November 2000).

19 A description of each funded project is available at

20 See earlier section Administrative Data. References for the statement are Classen et al. 1991, 1992, O'Neill et al. 1993, Jha et al. 1998, Bates et al. 1998, Malpass et al. 1999, Thomas and Brennan 2000, Gandhi et al. 2000a, 2000c, Honigman et al. 2001a, 2001b.

Return to Contents
Proceed to Next Section

Page last reviewed December 2001
Internet Citation: Section 6: An Agenda for Research in Ambulatory Patient Safety. December 2001. Agency for Healthcare Research and Quality, Rockville, MD.


The information on this page is archived and provided for reference purposes only.


AHRQ Advancing Excellence in Health Care