Patient Safety beyond the Hospital | Health Policy and Reform

Perspective
Patient Safety beyond the Hospital
NEJM | September 8, 2010 | Topics: Health Care Delivery, Quality of Care

Tejal K. Gandhi, M.D., M.P.H., and Thomas H. Lee, M.D.

The vast majority of health care is delivered in ambulatory settings, yet we are only just beginning to understand the safety risks that exist outside of hospital walls. There are 900 million visits to physicians’ offices in the United States each year, as compared with 35 million hospital discharges,1 and spending on outpatient care is the fastest growing segment of health care spending.2 Yet most patient-safety research and safety-improvement work have been done in inpatient settings; indeed, a search of the Patient Safety Network Web site of the Agency for Healthcare Research and Quality shows that since 2005 only about 10% of patient-safety studies have been performed in outpatient settings.

Experience to date indicates that safety issues in the ambulatory setting differ from those in the inpatient setting in obvious and not-so-obvious ways. There are differences in the types of errors (treatment errors predominate in inpatient settings, whereas diagnostic errors do in outpatient settings), the provider–patient relationship (e.g., adherence is more critical in outpatient settings), organizational structure (ambulatory practices tend to lack the infrastructure and expertise to address quality and safety improvement), and regulatory and legislative requirements (e.g., there are staffing ratios and accreditation requirements for hospitals that do not exist for private practices).3 In addition, the signal-to-noise ratio is much lower in outpatient settings: in ambulatory care, a physician may see 100 patients with chest pain before seeing one with an actual myocardial infarction.

The outpatient setting also presents greater challenges for information transfer. Particularly in the case of patients with complex medical needs, the responsibility for care is often shared by multiple providers at many institutions. These clinicians may never meet, and they often use different medical-record systems. Such care has long, fragile feedback loops. In the hospital, if a patient has an adverse drug event, clinicians become aware of it very quickly; in the outpatient setting, a complication or missed diagnosis may not be identified for months, if ever.

Medication safety is probably the most studied topic in outpatient safety research. One study showed that adverse drug events occurred in 25% of primary care patients and that 11% of these events were preventable.4 Solutions such as electronic prescribing have been advocated, and now national incentives exist for their adoption. But we still have much to learn about the effects of e-prescribing systems on errors and about how these systems can be optimized. For example, we need a better understanding of how to support clinically important decisions without causing unnecessary disruptions in clinicians’ workflow.

We also need better strategies for maintaining accurate medication lists in electronic medical records. Many integrated delivery systems, including ours, are currently struggling with knotty issues such as who is responsible for maintaining the accuracy of the medication list — for example, the deletion of medications that the patient is no longer taking, regardless of which provider prescribed them. If a specialist is the only physician in an organization who sees the patient, does that specialist have the obligation to enter all the patient’s medications and dosages in the medical record? How can we best reconcile differences in medication lists at each ambulatory visit? Addressing these issues usually means more work for clinicians who already consider themselves overworked.

Practices are also now observing that missed or delayed diagnoses are the most common problem leading to malpractice claims in the outpatient setting. As a starting point, our integrated system and others are working on practical strategies for tracking and following up on test results to ensure that communication of these results to providers and patients is 100% reliable. Tests revealing findings that are clinically significant but not critical (e.g., pulmonary nodules or colonic polyps) warrant particular attention. These nonemergency findings must be communicated appropriately to a responsible provider, who must acknowledge their receipt, and systems must be put in place to ensure that any requisite follow-up testing actually gets done.

The current interest in reducing preventable hospital readmissions has focused attention on the flow of information during transition of a patient’s care from the hospital to other settings. The concept of “no dropped balls” has an inherent appeal, but it requires teamwork among clinicians in these various settings. Among the data that should always reach outpatient clinicians are postdischarge medication changes, follow-up plans and appointments, pending tests, and the details of advanced care planning — yet we know that such information is not always transmitted. For instance, we have found that about 41% of hospitalized patients have pending test results at the time of discharge, and often neither the discharging physician nor the primary care provider is informed about these test results.5 Clear lines of responsibility need to be created to determine who will follow up on which results. In addition, the discharging hospitals need to implement high-quality discharge summaries that are transmitted in a reliable way, while outpatient physicians’ offices need to ensure patient access to timely postdischarge visits during which clinicians can review the discharge materials, reconcile medications, discuss symptoms, and perform appropriate follow-up so that the rate of readmission can potentially be reduced.

These considerations are just the tip of the iceberg when it comes to the problem of ambulatory care safety. Other areas, such as the safety of outpatient procedures, must be explored further. An increasing number of complex procedures are now being performed as day surgery or in nonhospital locations. Safe practices and reporting of serious adverse events should be implemented in these settings with the same rigor as they are in inpatient settings.

Outpatient settings may also offer special opportunities to improve safety — for instance, by permitting the implementation of innovative approaches to engaging patients in their own care. Through the use of portals that patients can use to access their electronic medical records, patients can help monitor their care, become aware of overdue tests or abnormal results, and update their medication lists. In some delivery systems, patients can view their own test results (including abnormal ones) online without having to wait for the physician to “release” them. Many clinicians have already had a patient notice an important abnormality that they had overlooked in a laboratory result; that experience will probably become more common as patients gain broader access to their medical records.

In the pay-for-performance era, robust safety-related metrics may help to improve the safety of ambulatory care. Potential examples of such metrics include the percentage of medication lists that accurately reflect what drugs the patient is actually taking or the percentage of prostate-specific–antigen tests or colon-polyp findings for which appropriate follow-up has been performed. Further work needs to be done to define appropriate measures and create better systems for capturing these kinds of data.

Perhaps the greatest immediate challenge in addressing these safety issues is that of creating a culture of safety in the outpatient setting, which is so often fragmented and disorganized and lacking in clear leadership. In contrast, many hospitals have made large investments in patient-safety teams, safety “walk rounds,” safety-reporting systems, root-cause analysis, and culture surveys. Ambulatory practices need some version of these investments — but few have made them yet.

The way to begin this work is to collect data on outpatient-safety risks through improved reporting of events and near-misses from front-line clinicians. The training of practice leaders in principles of root-cause analysis, quality improvement, and process redesign is also essential so that when they receive reports of safety concerns, they are able to analyze them and enact change. Provider organizations that recognize these challenges and respond to them will be ahead of the game as the spotlight is increasingly focused on care delivered not just by hospitals but by truly accountable care organizations in all settings of care.

Disclosure forms provided by the authors are available with the full text of this article at NEJM.org.
Source Information

From the Division of General Medicine, Brigham and Women’s Hospital (T.K.G., T.H.L.); and Partners Healthcare System (T.K.G., T.H.L.) — both in Boston.

References

1. National Center for Health Statistics. FastStats A to Z. (Accessed August 19, 2010, at http://www.cdc.gov/nchs/fastats.)

2. Accounting for the cost of U.S. healthcare: a new look at why Americans spend more. McKinsey & Company Global Institute. December 2008. (Accessed August 19, 2010, at http://www.mckinsey.com/mgi/publications/US_healthcare/Executive_Summary.asp.)

3. Wachter RM. Is ambulatory patient safety just like hospital safety, only without the “stat”? Ann Intern Med 2006;145:547-549
Web of Science | Medline

4. Gandhi TK, Weingart SN, Borus J, et al. Adverse drug events in ambulatory care. N Engl J Med 2003;348:1556-1564
Full Text | Web of Science | Medline

5. Roy CL, Poon EG, Karson AS, et al. Patient safety concerns arising from test results that return after hospital discharge. Ann Intern Med 2005;143:121-128
Web of Science | Medline

Patient Safety beyond the Hospital | Health Policy and Reform