Health care safety: what needs to be done?

The Quality of Health Care in America project, initiated by the Institute of Medicine (IOM) in June 1998, was charged with developing a 10-year strategy to improve quality. Its first report, To err is human,6 estimated that about 40 000 Americans lost their lives in 1997 due to medical misadventure. Although this is a small proportion of the 33 million hospital admissions, it is equal to the number of deaths from breast cancer or road trauma. The report classified errors into overuse, underuse or misuse of interventions, suggesting that economic rewards and sanctions would address the first two types of error, and that legislation and regulation, as well as promulgation of best practice, would be most effective for the third group.

The project proposed:

• a national centre for patient safety;

• standardised and mandatory reporting of adverse events and voluntary reporting of episodes in which patients narrowly escape serious misadventure;

• legislation to extend peer review protection to patient safety and quality improvement;

• safety performance standards for organisational and health professional licensing;

• stronger enforcement of standards for the safer use of drugs; and

• a requirement that health care organisations make continuous improvement in patient safety one of their key objectives.

The project’s final report in 2001, entitled Crossing the quality chasm,7 argued for the importance of clear and explicit expectations for performance in health care, offered 10 rules to guide physician–patient relationships, emphasised the value of incentives in achieving improved safety and quality, and endorsed the wider use of accumulated research evidence in clinical practice.

The IOM report has enlisted a wide variety of interested players and changed the way many health care professionals and managers think and talk about medical errors and injury. It has also accelerated changes in practice needed to make health care safer. The latter include 30 evidence-based safe practices, some of which the Joint Commission on Accreditation of Healthcare Organizations now requires of hospitals.8,9

US health care safety advocates Leape and Berwick recently outlined five main barriers to progress in safety.8 One barrier is medical culture’s commitment to individual professional autonomy — the behaviour changes needed to create a culture of safety are often perceived as challenging professional autonomy. Another is the fear that public reporting of hospital mortality data would undermine public trust, while fear of malpractice liability inhibits discussion or admission of errors. A combination of factors, including the complexity of modern health care, professional fragmentation, hierarchical authority structures, and diffuse accountability, together form a barrier to developing the sense of common purpose that safe health care requires. In addition, there is lack of agreement as to the appropriate measures of patient safety. Finally, a major barrier is that current financial arrangements may actually reward less safe care, as patients are charged for additional services arising from treatment errors.

Leape and Berwick assert that improving patient safety will require the use of electronic medical records, acceleration in the pace of adoption of safe practices, team training for health professionals, and open disclosure of iatrogenic injuries to patients. Patient safety will need to be made as important a corporate objective as financial health, and financial incentives for safe care, along with a process of national goal setting for safety, will need to be put in place.

Other commentators, however, claim that despite convincing evidence and recommendations from expert panels, the “quality problem” has never made it onto the US national agenda.10

The UK National Health Service National Patient Safety Agency11 collects and analyses information on patient safety, sets national goals and tracks progress. In 2003, the Agency published a monograph Seven steps to patient safety. These include building a culture of safety by promoting reporting, managing risks and identifying and assessing processes that could go wrong, communicating with patients and the public to share safety lessons, and implementing solutions to prevent harm.12 The effectiveness of this strategy has not been assessed.

What is happening in Australia? This year has seen a “Review of future governance arrangements for safety and quality in health care”, chaired by the New Zealand Health and Disability Commissioner, Rob Paterson. While acknowledging a long list of achievements by the Australian Council for Safety and Quality in Health Care, a smaller (nine member) Australian Commission on Safety and Quality in Health Care was proposed to replace it.13

The new body would be accountable to the Australian Health Ministers and their Advisory Council. It would establish a national strategic framework in which the many quality and safety initiatives in the states and territories might operate coherently. One criticism made in the review of the previous council was its lack of connection to the many safety and quality programs throughout Australia. Among its 16 recommendations, the review committee urged that one of the functions of the new Commission should be recommending national data sets for safety and quality, and nationally agreed standards for safety and quality improvement.

The concerns expressed by Wilson and Van Der Weyden have been reinforced by others14,15 — despite all the reports and programs on quality and safety improvement in the US, UK and Australia, all these countries lack evidence of measurable progress in patient safety. So what should we do?

First and foremost, we must create and put in place a system that measures quality and safety: the QAHCS should be repeated, say every 5 years. Perhaps the new Commission will have the grunt and resources to do this. It is hard to conceive of a strategic framework in which the outcomes cannot be measured. Analysis of trends every 5 years would allow us to set new goals for safety and quality. The federal health minister should have available safety reports from across the country — the Paterson review accepts this notion and adds that the same should be true for state and territory ministers.13 Without coordinates, mapping is a nonsense.

Second, our health service financing should be re-engineered to provide incentives for safer care. Differential Medicare rebates could be offered to practitioners who demonstrate superior safety in their practice. National and state awards could be given for demonstrably safer care and for systems that improve the safety of care. The Australian Health Care Agreements16 should contain financial incentives to the states to improve the infrastructure for medical and surgical care, so that system errors are reduced and safety is increased. Models and checklists for safe care could be systematically developed and updated by the learned colleges and specialty groups, and provided to facilitate these processes.

Third, the National Patient Safety Education Framework of the Australian Council for Safety and Quality in Health Care, which was designed for the benefit of all health care professionals, should be brought to fruition.17 The safety principles and tools in the curriculum could be used in all undergraduate educational programs for health workers, and as in-service training for all front-line health workers and managers.

These educational programs would place a high premium on developing skills in team formation and teamwork, and on the principles of the managerial technique of continuous improvement. By learning from every error and examining the processes of clinical practice, health service managers may be encouraged to develop their organisations to facilitate reporting, measuring, learning and sharing ideas about how to improve safety. This must be linked to financial support sufficient to change the health care environment to one in which error and injury are less likely to occur.

We offer these “good ideas” — monitoring, education, financial incentives and improved management of hospitals — that we think deserve strong support. Ultimately, the concern highlighted by Wilson and Van Der Weyden will only be answered when we have new and ongoing statistics about hospital safety in this country. That is why monitoring comes first.