A new approach to the handling
of medical error at the HSC

As noted in Chapter Four, the HSC was formally committed to the concept of quality assurance. Whether or not it was able to achieve its commitment is a matter that was a significant part of the investigative review undertaken into these 12 deaths.

It is clear from the findings set out in this chapter, that the current reliance on line management, incident reports, and the Standards Committee structure at the Health Sciences Centre was just not sufficient to prevent the events that led to the deaths of these children. While the safeguards that the hospital did deploy must be left in place, they must also be augmented. The HSC must develop an institutional culture in which information about safety hazards is actively sought, messengers are trained to gather and transmit such information, and responsibility for dealing with that information is shared by all. This will require new approaches to quality assurance, risk management and team performance.

Dr. Jan Davies¹ prepared a detailed report on these topics for this Inquest. Her report outlines how modern health-care institutions are incorporating new approaches to quality assurance, risk management and team performance into their operations. The evidence presented to this Inquest indicates that the HSC must adopt and adapt these approaches. The following sections outline a number of key elements in a new approach to the handling of medical error at the HSC.

The benefits of these approaches include:

• For patients (and their families): an improvement in safety of the care delivered, as measured and demonstrated through the review of complications, such as death or injury;

• For health-care workers: an improvement in morale, in part because of improved patient outcome and in part because of an improvement in the process of the delivery of care. Teams that communicate and work together more effectively will also work more efficiently;

• For the organization: an improvement in efficiency. This can be measured against the use of both physical and fiscal resources.

Quality Assurance

Quality Assurance (QA) differs from the traditional method of establishing a particular standard of care as acceptable, or unacceptable. The traditional method often serves only to point the finger at one individual whose performance may not have been optimal. QA is also unlike quality assessment, which seeks only to determine the quality of a service. QA programs go beyond simply ensuring that quality care is being provided. Rather, they seek to improve quality and stress the need to identify problems, develop and implement solutions and monitor results to ensure that solutions work. They also need to ensure that they do not introduce new problems through a comprehensive method which involves working from the solution back to the original problem. This is sometimes referred to as 'closing the loop'.

The primary goal of any QA program should be to provide both excellent patient care and optimal working conditions for the providers of that care. Although health-care workers may need to continue to work under pressure (e.g., during natural or environmental crises) it should be kept in mind that the results of that care will suffer. Thus, quality assurance should also contribute to the development of optimal workplace conditions.

QA programs emphasize the importance of monitoring systemic performance and should be guided by information about the performance of the team, the organization and medicine in general. Data should include problems from everyday life, results concerning the actions of the team and team members' attitudes, the operation and culture of the organization and new trends in the provision of health care. A quality assurance program will therefore require a quantitative database to be effective. Measurement of outcome, however, particularly mortality, may not be sensitive enough to show that there is systemic dysfunction. For this reason, emphasis must be placed on evaluating even those processes of care that might have led, but did not lead, to a complication.

Use should be made of ongoing, episodic and focused measurements and evaluations. In the period under consideration, funding cutbacks led to a decrease in computerized record keeping of the Department of Anaesthesia's data collection program. This should not be allowed to happen again. When quality assurance processes are eliminated or suspended because of financial considerations, there cannot be any assurance that health care may not also be compromised. That is what the compilation and evaluation of such data are intended to provide, and when the data are lacking, that assurance is lacking. Data collection must ensure that the strengths and weaknesses of the organization's practices and training programs are measured and compared with published results from other institutions. Verbal debriefing and written reports can give individuals and teams the performance feedback needed to work towards improvement. Trends in performance must be measured over time to detect areas that require more or less emphasis. Both short and long-term performance data must also be collected to serve as criteria for selection of new team members.

Risk management

It is not possible to eliminate risk from pediatric cardiac surgery. The best team operating with the best equipment in the best organization will not always achieve a successful outcome. Since risk cannot be eliminated, it must be managed through risk management, the cornerstone of which is the assessment and evaluation of risks.

The integration of quality assurance and risk management

Quality assurance programs can be linked with risk management programs. These latter programs attempt to manage and eliminate risk by establishing appropriate defences against errors at each of the three components that QA programs analyse.

The three quality assurance and risk management components interact in the following manner:

Structure: The structure of an organization comprises its personnel, equipment and environment, and administration. Risk management at this level involves applying preventive measures, such as constantly evaluating training and planning, personnel, equipment and environment, to minimize the chance of an error being made. Thus, risk prevention involves the prior evaluation of risk of an error being made. Personnel should be evaluated to ensure optimal numbers, proper credentials and staff well-being. Equipment should be evaluated to ensure that what is needed is present when required, is working, has been serviced regularly and is present in adequate numbers. There should be a plan for both phased and emergency replacement of equipment. The environment should be conducive to the activities that must be carried out. At the administrative level, policies and procedures should be considered and protocols developed to deal with critical events. At the regulatory level, an evaluation should be made as to whether or not the government and other regulatory authorities are supportive of appropriate health care.

Process: Process deals with what was done and how it was done. This is the stage of care where it is possible to detect, mitigate or recover from errors. Risk management strategies for the Process component focus on identifying high-risk activities and their potential for bad outcomes, and attempting to minimize complications by preventing the development of further problems. For example, it might be determined that even though a surgical team and equipment were functioning at their peak, the severity of a patient's condition might require that the patient be transferred to another, larger facility.

Outcome: In assessing the Outcome component, efforts are made to decrease the consequences of an error. QA programs in this area focus on what is measured and monitored. For example, a QA program would ask if near-misses are monitored and investigated or if only those with complications are assessed. The Outcome component of risk management focuses on trying to limit loss or damage after any complication and on formulating methods to prevent future problems. To this end, surgical programs need effective and timely methods to recognize, manage and investigate adverse outcomes, as well as adverse processes.

One of the most effective approaches to risk management is the development of a Critical Incident Review Policy.

Critical incident review policy

A critical incident review policy should provide immediate, multidisciplinary identification, management and review of critical incidents and accidents. To avoid impeding communication and facilitate fact-finding, the policy should be implemented non-judgmentally. Reviews might include any of the involved medical, nursing, or other staff. Consideration must also be given not only to the actions of the individuals involved, but to faults in the system in which those people work and which contributed to the problem.

At the HSC, a critical incident review policy for pediatric cardiac surgery might encompass those working in the cardiology clinic, radiology suite, pre-operative (pre-assessment) clinic and nursing unit, operating room, intensive care unit and post-operative nursing unit. In such a structure, it would be appropriate to appoint one or several team members to the position of risk manager.

A review would be triggered by the occurrence of either a 'critical incident' (one in which process of care was not as planned or anticipated, even though outcome of care was acceptable) or a complication or 'accident' (one with true adverse outcome, such as death or serious bodily harm). In addition, a review could be activated by a department head or senior administrator.

The goal would be to have an initial review conducted within 48 hours and a report completed within 72 hours. Before such a review was conducted, priority would be given to the care of the persons involved. All of the equipment should be isolated, the facts documented and the family informed. The hospital should have the facilities to care for and debrief family and staff members.

The review should consist of:

• a survey of the scene;
• a review of all documentation;
• the establishment of a timeline of events;
• a review of all other pertinent information; and
• a review of all pertinent policies and guidelines.

Within 48 hours, there should be a meeting of a review committee. The purpose of the meeting should be to corroborate facts, identify contributing factors and develop recommendations. Either the department head or risk manager of the major department involved should chair the meeting, which should include those involved in the event. Minutes should be kept but not necessarily circulated.

The department risk manager should prepare the report. The final form of the recommendations developed by the review committee should be submitted, with a plan for their implementation and monitoring to the hospital or health authority risk manager. This person may choose to release the recommendations to the patient's family, but the report need not become part of the patient's hospital record.

The responsibility for ensuring that recommendations are implemented should lie with the hospital and regional health authority or equivalent.

One of the issues that should be considered is that of the need for confidentiality of the proceedings of the review committee, and protection of the information obtained from public disclosure.

At present the medical standards committees of the hospital are protected from disclosure through provisions in provincial legislation that allow members of those committees to refrain from answering questions about the discussions at the committees. If there is to be openness in the discussion of matters at quality assurance and risk management reviews, consideration will have to be given to providing similar protection for those processes. However, the need for protection from disclosure of the discussions from those proceedings will need to be balanced against the right and the need for patients and their families to know what occurred. The detail of how that would be accomplished is beyond the scope of this report. However, it is safe to say that consideration needs to be given to providing for such protection and disclosure.

According to Davies's report, a review policy similar to that discussed above is currently in use by the Calgary Regional Health Authority (CRHA). This policy was developed from one originally proposed and implemented in the Department of Anaesthesia at Foothills Hospital.

There are probably others as well. They can serve as a model for what needs to be put into place at the HSC.

Team performance

Problems in leadership, teamwork, communication and decision-making loom large in this report's account of the history of the Pediatric Cardiac Surgery Program in 1994. The HSC needs to take steps to address these human factors, since problems with human factors play a significant role in contributing to negative surgical outcomes.

It is beyond the scope of this Inquest to provide a detailed description of the improvements in team performance that the HSC should undertake if the Pediatric Cardiac Surgery Program were to be restarted. However, it is clear that these efforts would need to be ongoing, as opposed to a simple start-up initiative.

Ongoing, overall initiatives to improve team performance

The initiatives would have to pay attention to:

• the necessity for team building through a briefing and orientation process that included all players;
• the initiation of pre-operative briefing sessions that focused on the plan for the operation (including the anaesthetic) and plans to address any contingencies, should problems arise with the patient, equipment or personnel;
• the planning and the use of standard operating procedures;
• the development of leadership skills;
• the provision for team development or maintenance;
• emphasizing team communication;
• clarity in the decision-making process;
• a process of conflict resolution;
• post-operative debriefing sessions (for all operations, regardless of outcomes); and
• a stress management component.

Team Training

An important element of team performance is training that is specifically designed to:

• decrease or reduce the probability of errors occurring;
• correct errors before they have an impact; and
• contain or decrease the severity of the consequences of those errors that have been made.

Such training focuses on the development of concepts and behavioural strategies as measures to manage error. Teams should learn to address those day-to-day issues, particularly those relating to communication and decision-making, that arise between different members of the team, such as between doctor and nurse or surgeon and anaesthetist. These issues require an understanding of how teams and their individual members manage errors.

One component of such a training program would involve training in the principles of crisis management, using simulations of critical incidents. The purpose of such crisis training is to:

• provide participants with predetermined responses to critical incidents which can be called upon when needed, and
• instruct participants in the co-ordinated integration of all available resources to maximize safe patient outcomes.

However, health-care workers need to learn how to work together, not only during crises but also, more importantly, under normal circumstances, when nothing goes wrong. Teams that normally work well together will probably make fewer errors and encounter fewer problems. Then, if things do go wrong, the team will have already dealt with any interpersonal and organizational difficulties before the crisis.

Training to manage error must be inaugurated into the earliest training of doctors, nurses and other health-care workers. Without this type of training from the start of professional life, health-care workers will find it difficult to support and enhance the change from doctor-based to true team-based care. To be effective, the value of quality assurance, risk management and team performance must become embedded in the HSC culture. This will require active promotion by senior staff responsible for training and evaluation. These senior staff will therefore need additional training themselves.

Both health-care workers and administrators should establish the details of the program. Effective programs in the operating room will involve surgeons, anaesthetists, operating-room nurses, anaesthetic assistants, perfusionists, anaesthetic (respiratory) therapists, technicians, unit assistants and clerical staff.

Such training cannot be a single lesson, taken once, and assumed to 'fix' the problem. Repetition and reinforcement are vital if the desired outcomes are to be achieved.

Finally, if there is to be a Pediatric Cardiac Surgery Program at the Winnipeg Health Sciences Centre, it must incorporate ongoing policies of quality assurance, risk management and team performance. These must be ongoing programs that stress the need to identify problems, develop and implement solutions and monitor to ensure that solutions work and do not introduce new problems.

Recommendations

It is recommended that: The Province of Manitoba consider legislation that requires hospitals throughout Manitoba to establish appropriate quality assurance and risk-management programs that accord with the principles and suggestions contained in this report, and that legislative protection be granted to the discussions that form part of those processes, provided that the right of patients and their families to full disclosure of what has occurred to them during the course of treatment is not compromised.

It is recommended that: The HSC, in conjunction with the Winnipeg Regional Health Authority, develop a quality assurance and risk management program employing the principles and suggestions contained in this report.

It is recommended that: The HSC exclude doctors who have been involved in a case that is under review by any of the hospital standards committees from participating in the decision-making process relating to such a review.

It is recommended that: The HSC Department of Surgery develop an appropriate database for all surgical procedures, but particularly for pediatric cardiac surgical operations. The database should include information such as cross-clamp times, cardiopulmonary bypass times, total circulatory arrest times, amount of blood products used and such other relevant information as would allow for the proper monitoring of surgical trends within a given program or for a particular surgeon.

It is recommended that: Pediatric cardiac surgical data be collected in a way that makes it possible to compare Winnipeg procedures with those performed in other centres.

It is recommended that: The HSC establish a clear policy on how staff is to report concerns regarding risks for patients. This policy must ensure that there is no personal or professional jeopardy to the person who is making the report. It should be clear to every staff member to whom they are to present such reports.

It is recommended that: The HSC administration ensure that all staff members are made aware of their responsibility to use incident reports and fully chart problems with the process of delivery of care and any complications in the outcome of care.