Testimony by John N. Morley, M.D., Medical Director, Office of Health Systems Management Regarding Patient Safety in New York State, Conducted by the Senate Committee on Health and the Assembly Committee on Health on October 19, 2009

Mr. Chairman and Members of the legislature, I would like to thank you for this opportunity to address you on an issue that has been the major focus of my career for the last several years. I am very much looking forward to our discussion this morning.

I would like to begin by introducing myself. I have been with the Department of Health since Aug. 1, 2005. Prior to that I was the Medical Director for Albany Medical Center Hospital from July 2001 through July 2005. My clinical background and training includes residencies in Anesthesiology and Internal Medicine and Fellowship training in Infectious Diseases, Pulmonary and Critical Care Medicine. I was an Associate Professor of Anesthesia, Internal Medicine and Surgery at the Albany Medical College.

I became acquainted with NYPORTS and involved in Patient Safety when I was in clinical practice and the Associate Medical Director at Albany Medical Center in the late 1990s. By the time the Institute of Medicine issued its report, "To Err is Human," I was deeply involved in quality and safety in my own institution.

I would like to provide you with some background on NYPORTS. It was created under a different name in the late 1980s. It was developed in response to an awareness that many adverse events in hospitals were becoming known to the Department of Health through reports in the press. In addition, a single case was publicized involving a medication administration error. This error was egregious, causing significant harm to the patient but without any requirement for notification of regulatory agencies. The legislation that created the original Adverse Event Reporting system was unique in the United States at that time. A short list of reportable events included:

  • Death or impairments of bodily functions in circumstances other than those related to the natural course of illness or disease;
  • Fires that disrupt services or cause harm to patients or staff;
  • Equipment malfunction during treatment or diagnosis of a patient which did or could have adversely affected a patient or hospital personnel;
  • Poisoning
  • Strikes
  • Disasters or other emergencies external to the hospital environment which affect hospital operations; and
  • Termination of any services vital to the continued safe operation of the hospital

According to the National Academy for State Health Policy, there are approximately 26 states (and the District of Columbia) that have a reporting system for adverse events. All but one state are mandatory reporting systems. The Joint Commission, the primary deeming authority for the federal Centers for Medicare and Medicaid Services, also has a reporting system, but it is voluntary. The federal government has identified the collection of information on adverse events through designated Patient Safety Organizations (PSOs) as a major tool for safety and quality improvement.

NYPORTS began as a paper reporting system and has gone through several iterations over the last 20 years. In 1995, the first meeting of the incident reporting workgroup was held. Lucien Leape, M.D., a surgeon on the faculty of Harvard Medical School and a leader of the patient safety movement, set the tone and charge for the group and its system development. This group became the NYPORTS Statewide Council, which helped to formulate reporting policy. This is a non-statutory council that is available on an ad hoc basis to advise the Department. In 1998, the Department announced the first Web-based reporting system. The goal was simple – to have a streamlined reporting system that would allow hospitals to not only identify and study their own adverse events but to obtain feedback and to allow comparison with other hospitals in their region and across the state.

In 1999, the IOM report "To Err is Human" caught the attention of the nation and affirmed the goals and efforts of the Department to make health care systems safer. NYPORTS was attracting national attention as a model for adverse event reporting systems.

Currently, NYPORTS has 31 codes identifying 31 reportable adverse events. Approximately one-third of these events require the hospital to convene a team of physicians, nurses, pharmacists and anyone involved in the care of a patient to discuss and understand the causes of the adverse event. This is referred to as a Root Cause Analysis (RCA). The RCA process requires the team to develop corrective actions that address the systems issues within the root causes identified. The hospital must implement these corrective actions to prevent recurrence of a similar event. While the collection of adverse events is seen as a critical first step, the events collected are only of any value when the event is studied to understand what went wrong and/or what led to the adverse event. Without this analysis there can be no change and no improvement in safety.

The New York State Department of Health has long realized and appreciated our primary function – to assure the care provided to the citizens of New York is as safe and effective as possible. Toward that goal, many tools have been used in different areas of health care. NYPORTS is only one of the many tools the Department uses for data collection and for improvement of safety. In addition, we have the cardiac database that is used to study and improve both cardiac surgery and cardiac interventions. The Trauma registry, Health Care Associated Infection reporting system, Office Based Surgery adverse events reporting system and the Stroke Center Designation program are all used to understand and improve patient safety in New York. The various offices within the Department, including laboratory and epidemiology, collect information from hospitals through more than 30 different reporting systems. Some are clearly directly associated with quality improvement and safety, and others track conditions and indirectly impact outcomes. In addition to hospital and nursing home reporting, we receive approximately 2,000 complaints from the public each year, many of which lead to investigations either onsite or review of medical records.

Department of Health's Use of NYPORTS as a Tool for Patient Safety

The Department receives approximately 12,000 NYPORTS reports annually. A report is periodically issued providing aggregate data - outcomes and events for New York state hospitals as well as trends over time. While we have not done as much analysis of events as we would have liked, we have reviewed and continue to review specific issues such as wrong side or surgery, medication errors and maternal deaths.

When it was identified in 2005 that wrong-sided surgery was an ongoing issue, we convened a panel of clinical experts to review the cases. The panel, which included a representative of The Joint Commission, Dr. Richard Croteau, developed a protocol which addressed each step in the process for a patient's surgery – beginning with the surgical consultation and evaluation in the physician's office, informed consent, scheduling, admission, assessment prior to being moved into the operating room and a final "time out" immediately prior to the surgery. As a result, we created and defined a standard of care with the New York State Surgical and Interventional Procedure Protocol (NYSSIPP). Shortly after we published NYSSIPP, the Joint Commission asked us to participate in its "wrong side surgery summit." The modifications in the Joint Commission's universal protocol as a result of that summit were very similar to the New York protocol.

NYPORTS is a reporting system. Information comes into the Department and is reviewed. Most often the events are collected and nothing further happens with the information. It is expected that the hospital has reviewed the event and taken appropriate action. Often, however, the event can trigger an investigation to gather additional information. The investigation is conducted by the Department's Division of Certification and Surveillance. Of those onsite investigations, select cases are referred to another area of the Department, such as Epidemiology (Bureau of Communicable Disease) or the Office of Professional Medical Conduct. These referrals are relatively uncommon but have caused confusion on the part of hospitals. Is NYPORTS a "safety tool" for the collection of information to make process and system improvements or is it a regulatory tool to be used to say "gotcha" and levy fines? The emotional response to a bad outcome has created a great deal of pressure to fully investigate, cite and punish wrongdoers, regardless of whether at-risk or reckless behavior is present.

While the answer is not popular or desirable, it is clear that this tool has been used for both and in fact I believe it should be used for both – but I also believe clarification of the parameters for referral and refinement of the process is necessary. We have and will continue to identify those cases in which it was clear that reckless behavior played a significant role in the event. These cases require an unambiguous response from the regulatory agency. Most events, however, are not the result of reckless behavior but are the result of human error, to which each and every one of us is susceptible. We have a superb judicial system that while not perfect, does distinguish between the man who picks up a gun and starts firing into a crowd and the man who opens the door that hits the table that causes the gun to drop on the floor and fire. It is only when we know what happened that we can respond appropriately. We must always remind ourselves that the goal is improve and optimize patient outcomes, not to eliminate errors. Humans will always make errors. The goal is to prevent complications and obtain the best possible outcome for the patient. When we learn how errors occur we can create processes that acknowledge human errors and will catch them before they reach the patient.

Safety and adverse events in hospitals can be compared to the transportation industry or more specifically – motor vehicles – another area in which great advances have been made in safety but we have not eliminated errors. When the accident occurs, the outcome is reviewed. Was it a scratch, a fender-bender or a collision with death and a great deal of damage? Was alcohol or other substances are involved? Was someone shaving or applying makeup while looking in the mirror or reading the newspaper (i.e., not paying attention)? Was speeding a relevant issue? It could be at 5 mph above the speed limit on a rainy day or 50 mph above the limit. Sometimes a bolt of lighting strikes a tree that falls on a truck. Sometimes people make mistakes because we are human, sometimes we demonstrate at-risk behavior with relatively minor actions, and sometimes we are reckless.

When addressing or responding to any type of transportation event it is critical to understand "what went wrong." It is much easier to understand what went wrong when information can be gathered without fear of punishment. That is not to suggest that there is no role for punishment, but rather that we must be clear on the criteria and consistent in our expectations.

NYPORTS education for providers has emphasized culture change in health care facilities, specifically the concept of a just culture. A just culture recognizes that individuals should not be held accountable for systems failures over which they have no control; however, it does not tolerate conscious disregard of standards, policies and procedures that promote risky or reckless behavior affecting the health of patients. In removing the "blame" from adverse event reporting, organizations such as the Federal Aviation Administration (FAA) and the Veterans Administration (VA) have succeeded in collecting the information necessary for understanding the underlying causes of adverse events and in implementing systems-based responses. This approach has allowed these organizations to achieve significant improvement in their safety records.

The airline industry learned a long time ago that firing the professional involved in an event without changing systems results in eliminating the only person who now has the experience to know how (and why) not to make the mistake a second time. Addressing systems issues brings issues out into the open for discussion, study and design of systems intended to prevent recurrence of adverse events.

But even after data is gathered, analyzed and policy and protocols created, there is still much to be done to bring about a safer environment. The Veterans Administration has an internationally acclaimed Patient Safety Center. In their Spring 2000 publication "Ambulatory Outreach," Dr. James Bagian and Dr. John Gosbee point out: "Without facility culture change, no policy, procedure, rule or regulation will make caregivers comply with a systems approach to patient safety."

Completeness of Reporting

I would like to respond to an issue that has been raised in the press and in the New York City Comptroller's report released last spring. The Department has received criticism for its monitoring of the completeness of reporting to NYPORTS. Dr. Charles Billings, the architect of the NASA Aviation Safety Reporting System, states that "… in the final analysis, all reporting is voluntary." Dr. James Bagian, a physician, an astronaut and the Chief Patient Safety Officer of the Veterans Administration, has made the same statement to Congress in testimony. Both these physicians are acutely aware of the complexity of the clinical condition and the requirement for interpretation of both the condition and the event definitions.

Underreporting of events occurs for many reasons. It is critical to acknowledge that underreporting can be the result of poor system design for collecting adverse events in the facility. Large complex hospitals have a great deal going on, and every person has a large list of responsibilities. The first concern of every provider is the care of the patient. A great deal of effort has gone into improving documentation of the care over the last decade and now we are asking them to document twice – once in the chart and a second time for their incident reporting system. Once the patient is cared for and the extensive documentation completed, assuming someone remembers that a particular event falls into the category for required reporting, a decision is to be made as to whether a particular case meets the definition of a reportable event.

Then there are the very real and significant disincentives including shame, liability and concerns about retaliation – both personal and institutional. Add to this the challenge an institution must face when it is very aggressive about reporting every possible case, resulting in either the Department of Health or the public interpreting the large number of events as bad care. The institution with a lower number of events reported may be perceived as a very good performer or a very poor reporter.

As I indicated earlier, if our goal is to improve safety and outcomes, this will not happen by merely collecting checks in a database as gunslingers collected notches on a gun. It is much more complex than that. In addition, we do have what is likely considered around the world as the gold standard in accuracy for such a large database. Our Cardiac database is verified in two different ways. Resources are dedicated to compare results with the Statewide Planning and Research Cooperative System (SPARCS). It takes close to a full year for the information in SPARCS to be available for comparison of cases. In addition, trained clinician reviewers are dedicated to review of cardiac medical records to verify the accuracy of the information we receive. This is an extremely costly and resource-intensive process. If we believe that complete reporting is necessary for improvement to take place, we need to be prepared for significantly increased costs, significant additional delays in generating reports and for what potential – improved data or improvement in safety and outcomes.

In my opinion, there are events or codes for which our reporting is almost certainly accurate, though not perfect, and codes we believe where reporting is not at all complete. Events that are most serious and relatively rare are very likely to be reported to us. In fact, we have higher numbers of reports of some events for New York State than The Joint Commission has for all 50 states and the District of Columbia.

Those codes for more common but less serious events can be very difficult to interpret in view of a complex patient, a complex health care system where patients are admitted and discharged from more than one hospital in very short time frames, and definitions that are not always entirely clear. The codes for less serious, more common diagnoses can either be acquired out of the hospital, and may be the indication for admission, or acquired in the hospital as a complication of an underlying disease or condition. It is these events that are likely less commonly reported to NYPORTS as compared to the frequency of their actual occurrence. The bottom line, however, is that we do not have a "gold standard" for comparison of many of these conditions in order to know the actual totals.

The Department and its Bureau for Certification and Surveillance are aware of those organizations with lower reporting rates. We work with them to educate staff on the codes and the definitions for those codes to improve their reporting responsibilities. Periodically, reportable adverse events are identified by other means than by hospital reporting. Chart reviews are done for other purposes and in the review of those cases a reportable event may be identified. More than 2,000 complaints are received annually from patients or families, and an investigation of a complaint can turn up a reportable event. When an event is identified that we believe met criteria for reporting, it is brought to the attention of the institution. This will frequently result in the issuance of a Statement of Deficiency and perhaps even a fine. The Department has issued 1,272 NYPORTS-related citations from 2005 to the present.

National Influence and Comparative Experience

In summary I would like to say that NYPORTS was created as a system for collection of adverse events. That system has evolved over the last 20 years as health care has evolved, and it is now one of many tools we are using to improve patient safety. It was the first statewide adverse event reporting system and later the first Internet-based reporting system. Rhode Island, Tennessee and several other states have adopted similar reporting systems. NYPORTS was intended to provide information to the Department and to provide feedback on a state level to hospitals for their use to prevent further such events. A great deal of time and work has gone into collecting this information and many changes – improvements – have been made, but it is clear that we can and must do much more. I would like to see this reporting system be clearly identified as a tool for patient safety, as so many other states and the federal government have identified the importance of such a tool for this purpose. I also believe we can provide greater clarity for the industry as to what information is to be utilized purely for process and systems improvements and what information is to be referred to other areas of the Department for evaluation and response. This has been a major objective of the federal patient safety organizations, and the Agency for Healthcare Research and Quality (http://www.pso.ahrq.gov/ ) has spent a great deal of effort defining the firewall that must be maintained between the regulatory role and function of government and the non-punitive side that must understand the processes and the errors before we can respond. Once again, the transportation industry has shown us what can be accomplished. If you would like to know more about their reporting system you can find a great deal of information on the Web (http://asrs.arc.nasa.gov/). In this system NASA receives and reviews reported information and issues safety alerts and newsletters to the airline industry. The Federal Aviation Administration will take regulatory action when events involve risky behavior.

We will continue to work with national health care experts and the our health care industry to ensure that New York has the best patient safety systems, strengthening the confidence of patients and stakeholders alike.

I am well aware, as are you, of the level of public frustration in the area of patient safety. We have not accomplished nearly enough nor nearly as much as we hoped in the 10 years since the IOM report "To Err Is Human." But we have made measurable progress and we are building a foundation on which a great deal more can and will be accomplished. This is a very large boat to be turned, but we are overcoming inertial forces and change is taking place. Cardiac care has better outcomes, trauma care is improved, transplant surgical outcomes are improved, and health care associated infections are dropping. But there is far more yet to be accomplished, and I am completely confident we have only seen a small fraction of the improvements to be made in the next 10 years.