In December 2008, the U.S. Department of Health & Human Services, Office of Inspector General (OIG) issued three reports regarding its research with respect to "adverse events" at hospitals. The reports were required by the Tax Relief and Health Care Act of 2006 and will be submitted to Congress in fulfillment of the OIG's legislated mandate to study the incidence of, and payment policies associated with, adverse events affecting Medicare beneficiaries. The reports are of particular interest because of recent efforts by the Centers for Medicare & Medicaid Services (CMS) to deny Medicare payments for certain hospital-acquired conditions and "never events," and will likely guide the development of future legislation, or regulation, to improve the safety of the American health care system.

The first report, "Adverse Events in Hospitals: Overview of Key Issues" (Key Issues Report), based on OIG research, including interviews of 85 stakeholders in patient safety (such as researchers, clinicians, officials from federal and state agencies, oversight entities, and patient safety groups), identifies important issues related to measuring, reporting, and responding to adverse events at hospitals. The second report, "Adverse Events in Hospitals: State Reporting Systems" (State Report), inventoried current state-sponsored programs to require reporting of adverse events. The third report, "Adverse Events in Hospitals: Case Study of Incidence Among Medicare Beneficiaries in Two Selected Counties" (County Incidence Report), attempts to quantify the incidence rate of adverse events in two unidentified counties.

The Key Issues Report

The Key Issues Report identifies seven issues regarding adverse events in hospitals that are deemed most critical in responding to the problems posed by adverse events. The report contains a succinct overview and descriptive analysis of each issue, combining evidence, analysis, and opinion culled from both a review of published works and structured interviews with stakeholders. This research took place between March 2007 and November 2008. The seven key issues are summarized below.

1. Estimates of the incidence of adverse events in hospitals vary widely and measurement is difficult. The OIG's research revealed a wide disparity in the projected incidence of adverse events at hospitals nationwide. Published estimates range from 2.9 percent incidence rate of adverse events as reported in "To Err is Human: Building a Safer Health System," to other studies that place incidence rate of adverse events at close to 20 percent. Stakeholders' beliefs reflected a similar discrepancy, with most estimating the incidence rate at around three percent; however, a number of the stakeholders believe the incidence rate of adverse events is slightly higher. The Key Issues Report includes descriptions of available methods for determining incidence such as the screening of administrative data, electronic medical records surveillance, review of written medical records, patient surveys or interviews, and observation of patient care. The OIG did not find any method optimal and describes the advantages and disadvantages of each.

2. Nonpayment policies for adverse events are gaining in prominence and are viewed as a powerful incentive to reduce incidence, but raise potential drawbacks. Nonpayment policies for adverse events have recently been enacted by CMS and private payers for more than 50 percent of the hospitals that responded to a 2007 survey. Most stakeholders believe the financial pressure resulting from these policies will encourage hospitals to follow recommended guidelines more closely while improving patient safety and reducing adverse events. Stakeholders were concerned, however, about the potential negative effects of the policies such as limiting patients' access to care, increasing hospitals' costs while decreasing revenue, and incentivizing increased testing, which may lead to inappropriate treatment.

3. Hospitals rely on staff and managers to report adverse events internally, but barriers can inhibit reporting. The OIG's interviews revealed that most hospitals rely on staff or department managers to report adverse events to their hospital's quality improvement or risk management department. While most of the stakeholders believed routine reporting of adverse events is an important component of improving patient safety, they expressed concern that such reporting is frequently inhibited. The Key Issues Report identifies several reasons for this inhibition, including lack of time, the belief that such reports will not lead to improvements, an assumption that other staff will report the event, the difficulty in distinguishing adverse events from the harm caused by the underlying disease, and fear of retribution.

4. Hospitals report adverse events to various oversight entities, although stakeholders suspect substantial under-reporting. Currently, many oversight agencies receive adverse event reports from hospitals, including state reporting systems and private patient safety organizations. The Key Issues Report identifies concerns that prevent the collection of accurate and representative data about the incidence of adverse events. Many agencies, for example, request only a subset of adverse events, and the lack of a comprehensive reporting scheme may lead to duplicative reporting. In addition, reporting often imposes a significant administrative burden on hospitals. Hospitals sometimes must comply with multiple requests, some of which require detailed information or the submission of additional materials such as corrective action plans, staff surveys, or medical records. Many hospital administrators find these burdens outweigh any benefit from reporting. The stakeholders advocated adopting a new two-tiered reporting system that includes mandatory reporting for serious adverse events and voluntary reporting for less serious adverse events and near misses.

5. Public disclosure of adverse events can benefit patients, but also raises legal concerns for patients and providers. Many stakeholders viewed disclosure to patients and the public as an ethical obligation. They believe routine disclosure will provide a substantial benefit for patients by assisting them in making health care decisions and pressure hospitals to improve safety. However, the OIG expressed concern that reporting adverse information outside of a hospital can create legal problems for the hospital, particularly if it results in a waiver of state peer review protections. Moreover, the OIG is concerned that frequent reporting to a number of oversight entities may result in a loss of patient confidentiality.

6. Information to help prevent adverse events is widely available, but some hospitals and clinicians may be slow to adopt or routinely apply recommended practices. Stakeholders believed beneficial information and strategies about how to prevent adverse events is readily accessible, but often underutilized. The OIG found that many hospitals do not routinely practice known patient safety precautions, and many clinicians often do not comply with fundamental patient safety measures as recommended. The report identifies some of the reasons that providers do not apply proven patient safety practices, including the fact that many physicians believe standardized guidelines should not replace their training and judgment, and many hospitals lack the infrastructure to require physicians to implement new strategies. Physicians were also found to view adverse events as isolated incidents, lessening their interest in changing their practice patterns to a more regimented standard of care.

7. Interviews and the literature reveal strategies that may accelerate progress in reducing the incidence of adverse events in hospitals. Stakeholders described the current environment among hospitals and policymakers as "being on the verge of accelerated progress."

The Key Issues Report identified several strategies that can be used to reduce the incidence of adverse events and improve quality of care:

  • Create a national body to lead patient safety efforts
  • Focus on the use by hospitals of recommended practices and evidence-based clinical practice guidelines
  • Establish practical and accurate data collection methods for measuring the incidence of adverse events
  • Expand the use of electronic health records
  • Monitor and analyze the impact of policies to deny hospitals all or a portion of their billings for admissions complicated by adverse events
  • Improve the utility of adverse event reporting to outside entities by streamlining reporting mechanisms and by evaluating the comparability of data across entities

The State Report

The OIG's State Report provides a summary of state requirements for reporting adverse events. The OIG found that 26 states had an adverse event reporting system actively receiving reports from hospitals as of January 1, 2008 and that one state had taken action to develop such a system. These reporting systems were found to be disparate, differing in the list of reportable events, the criteria for determining which events are reportable, the information that must reported about a given event, and the strategies for obtaining reports and protecting confidentiality.

Of the 26 states with adverse reporting systems, 23 states had established their own lists of reportable events, while the other three states used the National Quality Forum's (NQF) list of serious reportable events. The states that established their own lists differed widely in the events included, with some states identifying very specific adverse events and others laying out broad standards.

Criteria for determining whether an event is reportable focus primarily on the level of harm the patient suffered. All states with reporting systems use a patient's death as a criterion for determining whether an event is reportable, and most use other levels of harm such as physical harm, the likelihood of additional medical care, and the occurrence of long-term harm or permanent disability. Only Pennsylvania requires reporting near misses. In addition to harm, some states require hospitals to consider additional criteria such as the length of time the patient experienced a negative outcome, whether the condition persisted at the date of discharge, and whether persons other than the patient also were affected.

The information hospitals must provide in reporting adverse events differs from state to state. All states require identification of the type of adverse event, and most require a summary of the event and the location of the hospital in which it occurred. Nineteen of the 26 states require specific information about the patient, 17 require the results of a root-cause analysis, and 20 require a corrective action plan and/or a risk-reduction strategy.

Staff in the majority of the 26 states with adverse event reporting systems acknowledged that hospitals under-report adverse events. States have developed several strategies to improve reporting, including administrative review of data to identify specific instances of under-reporting, monetary penalties for failing to report, improvement of legal protections to prevent improper disclosure, and the provision of useful information from the reports. To ensure the reliability of the reports, 23 of the 26 states conduct audits or investigations of the event.

Finally, states also differ in the ways they attempt to prevent adverse events. Of the 26 states with adverse event reporting systems, 21 used or plan to use the reports to communicate with hospitals through annual or semiannual reports, state-led patient safety training, patient safety bulletins and alerts, and Web sites.

Given the variation among state adverse event reporting programs, the OIG concluded that state adverse event reporting systems' data are unsuitable for a national platform. While states have identified several strategies for improvement, hospitals continue to under-report adverse events learned through complaints, audits, or other means. The OIG concludes that "states may find it prudent to consider other means to more effectively ensure reporting by hospitals."

The County Incidence Report

The County Incidence Report is a focused attempt to measure the incidence of adverse events in two selected counties in different states during a one-week period in August 2008. The report omits the location of the counties in the study, but warns that these results are limited to a specific place and time and should not be regarded as nationally representative of the Medicare population. The County Incidence Report also recognizes that the incidence rate is closely tied to the method used to identify when an adverse event has occurred. This makes comparisons across multiple researchers' data difficult to achieve.

In the County Incidence Report study, the OIG reviewed a random sample of 278 hospitalized Medicare beneficiaries selected from all beneficiaries hospitalized in the two counties during the involved time period. Reviewers conducted a two-stage review, beginning with interviews of beneficiaries, reviews of hospitals' bills and medical records, and the analysis of administrative data. If these steps suggested a potential adverse event, the reviewers had physicians conduct a full onsite review of the medical record to determine whether an adverse event had occurred. For the study, adverse events included: (1) events on NQF's list of Serious Reportable Events; (2) events on CMS' list of hospital-acquired conditions; and (3) events that resulted in serious harm to the patient on the National Coordinating Council for Medication Error Reporting and Prevention (NCC MERP) harm scale, which includes prolonged hospital stay, permanent harm, life-sustaining intervention, or death. The study found that 15 percent of the randomly selected patients experienced an adverse event with the requisite level of harm during their hospital stay. Notably, another 15 percent of patients experienced one or more adverse events during their hospital stays that the OIG classified as "temporary harm" such as low-level infections or allergic reaction to medication. Generalized for the two counties, the margin of error was 4.5 percent.

The study revealed data likely to be of interest for future policy decisions. First, while the 15 percent incidence of adverse events is much higher than the 2.9 percent reported in 1999 by "To Err is Human," the findings emphasize that the incidence rate is highly sensitive to the definition of adverse event. The majority of the adverse events detected in the random sample qualified as adverse based on the NCC MERP harm scale, and not based on the NQF or CMS lists. Less than one percent of the patients experienced an event on the NQF list and less than four percent experienced a CMS hospital-acquired condition. Moreover, 28 of the 43 patients experienced the lowest level of harm available to classify an event as an adverse event, defined as a temporary harm that prolonged a hospital stay. The County Incidence Report also emphasized that differing criteria used to define adverse events can complicate the measurement of the incidence rate, as some events are found on both lists with differing detail, and both lists exclude some events that seriously harm the patient.

Second, the majority of adverse events that were reported did not lead to higher Medicare reimbursement. Only one of the 11 events that were on CMS' list of hospital-acquired conditions, and one of the three events that were on the NQF list, actually led to a higher Medicare reimbursement. In addition, the OIG noted that some hospitals in the study did not include diagnosis codes reflecting hospital-acquired conditions in claims for Medicare reimbursement. In fact, seven of the 11 hospital-acquired conditions identified in the sample from CMS' list were not reported in the billing data.

The County Incidence Report substantiates concern about the incidence of adverse events in hospitals. The OIG has funding through 2009 to continue learning about methods for identifying adverse events in hospitals, and is currently expanding its study of the adverse event incidence rate to provide a better national estimate of adverse events affecting Medicare beneficiaries.

Conclusion

Taken together, the Key Issues Report and State Report represent interesting findings that may be used by regulators to guide future developments, and the County Incidence Report suggests that the incidence rate of adverse events may be much higher than previously thought. The reports make clear that reducing the incidence of adverse events is a high priority for the government and support the current payment policy to deny coverage for certain adverse events as a means to prevent them. The reports also highlight the disparity in the patchwork of state-imposed mandatory reporting requirements and advocates in favor of a national mandatory reporting requirement for the most serious adverse events. In fact, CMS notes its intent to seek an amendment to the current Patient Safety and Quality Improvement Act of 2005 to mandate national reporting of adverse events. Providers should expect further analysis by the OIG on this issue as the government continues its quest to drive quality and patient safety through payment changes and public reporting.

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.