Intended for healthcare professionals


Delivering safe health care

BMJ 2001; 323 doi: (Published 15 September 2001) Cite this as: BMJ 2001;323:585

Safety is a patient's right and the obligation of all health professionals

  1. Paul Barach, editor, Quality in Health Care (pbarach{at},
  2. Fiona Moss, editor in chief, Quality in Health Care (fmoss{at}
  1. Center for Patient Safety, Department of Anesthesia and Critical Care, University of Chicago, Chicago, IL 60637, USA

    See advertisement in clinical research edition (facing p 610), general practice edition (facing p 623), and other editions (facing p 583)

    One fundamental guarantee that we cannot give our patients is that faults and errors in the healthcare system won't harm them. Of course, health care is by its nature risky. Not everyone undergoing surgery for an aortic aneurysm survives. Many interventions carry risks. But these risks are mostly small and usually quantifiable. Ideally, patients understand the possible risks and benefits before choosing to undergo a procedure. For some patients these are difficult decisions. Though doctors may discuss risks of treatment, they do not speak about risks of harm from the system—or even about such harm when it occurs.

    Recent studies in the United States, Australia, and the United Kingdom and reports from the US Institute of Medicine and the UK Department of Health have drawn attention to the chronic “unsafeness” of health systems worldwide.1-7 This attention is not new. What is new is that preventable, iatrogenic injuries are being quantified and openly discussed. For example, adverse drug reactions have become a national issue in the United States—studies show that adverse drug events occurred in 6.5% of hospitalisations.8 By calling for solutions, these reports have highlighted the tensions between accountability and improvement, needs of individual patients and benefit to society, and production goals and safety.

    Most causes—and solutions—lie in the systems of care and how we work. Healthcare professionals, however, focus energy on individual patients, tackling difficulties in the system as they appear—often as separate problems and not in parallel. Individual care is of course crucial. But unless attention is given to the system our patients are at risk from a faulty service. For example, inadequate handovers can mean that vital information is lost between different care givers and services. Is it that the word “system” is anathema to many doctors? Just getting health professionals to work harder or exhorting them to be safer will not help; the system of care must be redesigned. We must instil a chronic sense of unease—a constant awareness of risk in every action.9 Such attention to risk enables crews of aircraft carriers to launch and land several planes every day on decks the size of two football fields with virtually no adverse events. All hands know that one oversight can lead to disaster.10

    Theories of quality improvement in complex systems have helped the understanding of safety in health care. Safety is the aim, and improving skills and techniques is the method to get there. Much is known about how to build safer systems and reduce risk, but little of this knowledge is embedded in health care—and until it is, the sustained changes in behaviour of individuals and organisations that are needed for safer care are unlikely. Punishment will not help.

    The knowledge, skills, and attitudes needed for safe practice are not normally acquired in medical school. The disciplines in which risk management and quality improvement are important are wide ranging and cut across professional, clinical, and organisational boundaries. Some of these disciplines—cognitive psychology, ethics, bioengineering, mathematics, statistics, information science, ethics, and law—will be familiar. Others—change management, team work, organisational behaviour, systems theory, disaster analysis, and human factors—may not be. Not all these disciplines need be given their own space in the curriculum, but each should support the development of understanding about safety. How long, though, should we wait before all medical schools and training programmes include safety of patients as a central objective?

    Doctors have mostly avoided the question of how safety can become central to their work. Employing an expert will not reduce harm. A general call to embrace safety may influence a few people but will not change systems. Care will be safer when we learn to work as teams and understand the team as a microsystem—a small, focused, organised unit with a set of patients, technologies, and practitioners.11 Some important changes that health professionals can make may be very low tech and seem trivial. How would methicillin resistant Staphylococcus aureus survive if all doctors always washed their hands after examining a patient? We know this would make a difference. The difficulty lies in implementing what we know.

    Improving safety of patients should be one of the highest priorities of healthcare leaders. Perhaps things are changing. In the United Kingdom the National Patient Safety Agency has just been set up, and in the United States President Bush has increased the budget of the Agency for Healthcare Research and Quality by $100m to promote research on safety of patients.12

    Easy access to research on improving safety may help doctors and other health professionals make care safer. Quality in Health Care, a journal of the BMJ Publishing Group, has included papers on safety in the past. From March 2002 the journal will become Quality and Safety in Health Care. It will continue to publish papers on quality improvement but will include more papers on safe care and safe practice. We invite readers to send us these. Changing attitudes and practices will be hard work. Patients are being placed at unnecessary risk and many are harmed; they expect that we will offer safer care.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.