|
National Summit on Medical Errors and Patient Safety Research
Panel 4: Reporting Issues and Learning ApproachesTestimony of Marie Dotseth, Minnesota Department of Health (Continued) Select for Part 1 Disclaimer and Copyright Statements What is the effect of tort laws on reporting in all the various current reporting systems? What laws should states put in place to maximize reporting? Providers claim that mandatory reporting will never be truly effective and might be counterproductive, due to the fear of malpractice lawsuits. Physicians fear (rightly)that data collected by the state may be subject to "sunshine laws" and therefore will be discoverable by plaintiffs' counsel. This fear is understandable, based on the current culture of ‘finger-pointing’ when mistakes are made. There needs to be a fundamental shift away from placing blame with the individual and towards fixing fundamental flaws in the system. An example of positive systems change effected by the reporting of adverse events is clearly presented in Joint Commission data. The removal of concentrated KCl (potassium chloride) solution from cabinets on the patients' floor in hospitals has reduced the number of deaths due to the administration of the medication to zero over a period of two years. Culture rarely changes readily on its own—it often needs the threat of some form of incentive, either positive or negative, to encourage true change. A good example of the power of culture in medicine is illustrated through the following example. A hospital's intensive care director found through studies that wearing lab coats was unnecessary and took up precious time. However, no matter how many times he asked, the physicians still wore their coats. It turns out that the different specialties wore different color coats and therefore, wearing the coat was an important mechanism for defining oneself in a group of one's peers and was embedded in culture. If the culture of medicine changes and shifts to more positive systems approaches for dealing with error inherent in a complex system, then "fear of discovery" should no longer be a factor. A blame-free, systems approach to reporting medical error might actually help instigate this change. Research is needed to determine how a reporting system could best be made compatible with the current culture in health care. In addition to this, research that could identify mechanisms through which change could be effected within the medical community, and be viewed in a positive light, would help when establishing a successful reporting system. The Institute of Medicine report encourages Congress to pass peer review protections for this data. The Institute of Medicine is not suggesting that any collected information, linked to identifying provider information, be released to the public. The Institute does suggest that in some cases (a particularly egregious event, or a history of events) the information should be given to the public. The report also suggests that a serious effort must be made to protect information from legal discovery if the reporting system is to succeed. However, the interest generated by the data collected by any mandatory reporting system will likely be substantial, and as a result, it will be critical to ensure strong and appropriate protections for this information. The recent interest surrounding the legislation calling for the release of the information in the National Practitioner's Database points to the level of interest such data will generate with the press and the general public. If providers, insurers or health care institutions fear litigation based on information provided about errors, or even "near misses", they will certainly have a strong incentive to under-report as well as a disincentive to report events internally which often lead to quality improvement. Kansas recently opened up its peer review process to the public after it was determined that a particularly egregious event at a hospital was covered up through the peer review system. This would be a very useful example to study to determine the effect of less comprehensive protections on reporting levels and consequent quality improvements. It is crucial that any system of mandatory reporting address the issue of privacy, because without accurate and complete reporting, the system will be ineffective. Research on reporting systems in different legal climates is necessary to determine the effect of the type and level of data protection and the impact on the level of reporting. Research on the protections of the information collected through the sentinel events reporting system administered by the Joint Commission would provide good information on the effects of extremely strong protections and absolutely no public disclosure on reporting levels (in a voluntary setting). It would be useful to determine whether or not the Joint Commission's regulatory role outweighs its careful attempts at protecting privacy and maintaining a voluntary system. This type of information could contribute to solving the problem of what kind of entity should implement and maintain a mandatory reporting system. Possibly the most important issue is whether reporting systems actually use the data they gather to effect measurable quality improvements at the health care delivery level. What reporting system characteristics best insure quality improvement? After the release of the Institute of Medicine report, President Clinton asked the Quality Interagency Coordination Task Force (QuIC) to review the report's recommendations and make their own as well. This report, "Doing What Counts for Patient Safety: Federal Actions to Reduce Medical Errors and Their Impact," released in February of 2000, suggests that an appropriate framework for a reporting system would include dual systems as well as purposes and presents an excellent framework for a reporting system with the goal of achieving quality improvement.19 The purposes would include learning and accountability and the system would have both a voluntary and mandatory part. Data collected for learning purposes should be tied to incentives to ensure reporting in the mandatory setting (or encourage it in the voluntary setting). An incentive could include valuable feedback to the provider from the reporting system about opportunities for quality improvement. Another critical aspect to the viability of a mandatory reporting system, where the data is used for learning, is appropriate protections from discovery. A valuable aspect of both a mandatory or voluntary reporting system which collects data for the purpose of learning is the inclusion of information on near misses which might have lead to major harm to a patient. In terms of accountability, a mandatory system could include an audit of the safety and error analysis process. This system could be based upon other existing systems for high interest/high profile events. If accountability is a goal, however, a voluntary system alone is not adequate, thus explaining the dual system requirement. The QuIC report also recommended that the Health Care Financing Agency(HCFA) conduct pilot studies with its own mandatory error reporting system. HCFA will work with a chosen state to further study the effects of requiring reporting of a specified list of carefully defined events, with information being made public on a hospital-by-hospital basis. More pilot studies such as this one should be undertaken in an attempt to clearly quantify (in terms of level of reporting) the effects of reporting, in a public context, on initiating quality improvement. A major issue with any reporting system is the analysis and use of the collected data to effect measurable quality and safety improvements in health care systems. The recent National Academy for State Health Policy report states that out of fifteen states with mandatory reporting systems, the collected data is used in the following way: ten states aggregate it to identify trends, while nine impose sanctions, and nine assure corrective actions. Importantly, only two use it to develop quality improvement projects, while four collect the data but do not do ongoing analysis.20 Several organizations opposed to mandatory reporting have raised concerns that the data collected will not be used in a productive manner, such as to make systems quality changes and provide feedback to providers, but instead will likely only be used to punish individuals. Concerns have been voiced specifically from the American Hospital Association that the data will be used only to count mistakes and not actually be translated into action at the healthcare level. Ensuring appropriate and beneficial uses for the information collected from any reporting system will be critical to the success of any reporting system. Research on standard ways to both analyze data and communicate trends is necessary to ensure the success of reporting systems. Furthermore, concrete processes need to be identified to transform feedback about unsafe trends and patterns in health care delivery into quality improvement at the care delivery level. Research into appropriate sanctions as well as incentives for quality improvement should be undertaken prior to the establishment of any new mandatory reporting systems at the state level. A major potential use of the reporting data to pursue quality improvement is the identification of significant trends for certain conditions or procedures which would be appropriate for special initiatives by medical providers and public health agencies to address as statewide provider education and community-based awareness activities. In the case of Minnesota, with the potential reporting system being administered out of the state public health agency, there are opportunities for analyzing the trends in medical errors in coordination with traditional epidemiologic surveillance systems to identify public health priorities for prevention interventions which have the greatest impact for improving health status for the population. If there are trends from the medical error reporting data which occur across many provider organizations and appear to have a major impact at a regional or statewide level, then these data can be used for developing special initiatives which would be synergistic with public health initiatives. Current examples which have been well-established public health priorities are diabetes care management, early diagnosis of certain types of cancer, prompt treatment at the early signs of stroke and appropriate use of antibiotic medications. These initiatives may involve statewide provider education, community awareness campaigns for the public and more focused epidemiologic surveillance activities. By establishing a reasonable set of priorities for special initiatives where the leadership could be provided by the state public health agency, the use of medical error reporting could be made more effective by promoting the quality of health care in coordination with addressing public health priorities. Several examples of quality improvement through feedback from reporting systems can be cited. The NYPORTS system published a newsletter entitled "NYPORTS: News and Alerts" identifying major trends and issues discovered through reporting in response to a request by providers for feedback on quality improvement. This newsletter points out that if the NYPORTS reporting system is to be an effective quality improvement tool, providers must include information regarding process improvements and lessons learned. This newsletter represents a concrete product from a mandatory reporting systems attempting to provide feedback to providers and initiate quality improvement. Other products should be modeled after this one for future use in reporting systems as a method of providing timely feedback to providers. A second example of quality improvement resulting from a reporting system is seen in the NASA's Aviation Safety Reporting System (ASRS). This reporting system was established at NASA, a well respected and highly esteemed research institution, because pilots were reticent about reporting to the regulatory Federal Aviation Administration which controls licensure and can impose sanctions. This system receives 36,000 reports per year, and has instituted several systems improvements as a result of data collected by the reporting system.21 The ASRS is an excellent example of a non-punitive system that functions well due to its simplicity and ease of use, as well as confidentiality. Research into the applicability of such a system to the health care industry should be undertaken. The Institute of Medicine report is a very critical first step in identifying appropriate actions at the state level around the issue of patient safety. The overriding concern with mandatory reporting systems is lack of proven effectiveness. Policymakers, feeling pressure from the public to take action, may establish reporting systems that do nothing more than provide highly generalizable, unidentifiable trends that are largely useless. The easy default action has become setting up a mandatory reporting system at the state level. States are not even certain that they are the appropriate site for such a system or that quality improvement should be the primary goal of state or federal government involvement; perhaps an accrediting organization, such as JCAHO or NCQA, or the private sector would be more desirable or effective at administering a reporting system. Before states take steps towards implementing costly mandatory reporting systems, best practices should be determined through research. Appropriate roles of the affected parties should be defined and expectations about output and feedback should be explicit and clear. 19. The QuIC. "Doing What Counts for Patient Safety: Federal Actions to Reduce Medical Errors and Their Impact." Feb. 2000; page 51. 20. See note 6. 21. Linda Connell. "Aviation Incident Reporting: Valuable Information for Safety" Presentation at the Dallas Symposium on National Patient Safety, June 29, 2000. Current as of September 2000 Internet Citation: Testimony of Marie Dotseth, Panel 4: Reporting Issues and Learning Approaches. Written Statement. National Summit on Medical Errors and Patient Safety Research. September 2000. http://www.quic.gov/summit/wdotseth1.htm Return to Testimony Directory |