Intended for healthcare professionals


Patients' safety

BMJ 2005; 330 doi: (Published 10 March 2005) Cite this as: BMJ 2005;330:553
  1. Daniel Stryer, director,
  2. Carolyn Clancy (cclancy{at}, director
  1. Center for Quality Improvement and Patient Safety, US Agency for Healthcare Research and Quality, Department of Health and Human Services, 540 Gaither Road, Rockville, MD 20850
  2. Center for Quality Improvement and Patient Safety, US Agency for Healthcare Research and Quality, Department of Health and Human Services, 540 Gaither Road, Rockville, MD 20850

Progress is elusive because culture in health care has not changed

Since 2000, when “To Err Is Human” stimulated action to eliminate errors and mitigate the resultant harm in the United States1 and “An Organisation with a Memory” initiated similar efforts in the United Kingdom,2 healthcare systems worldwide have devoted considerable attention to the safety of patients. Yet despite attempts to reduce adverse events through multilevel interventions and information technology, widespread change in the culture of health care remains elusive.

The numbers of affected patients are astounding. In the United Kingdom, adverse events with resultant harm were estimated to occur in some 10% of hospital admissions, equating to more than 850 000 events annually. In the United States, extrapolations based on medical record reviews imply that 44 000-98 000 lives were lost because of medical errors each year. Although some posit that these numbers were inflated, ongoing work indicates that these estimates may be conservative. For example, Davis et al found that 12.9% of hospital admissions in New Zealand were associated with an adverse event.3 Zhan and Miller examined the incidence of 18 hospital based diagnoses suggestive of errors and found that injuries may add 2.4 million extra days in hospital and more than 32 000 deaths per year in the United States.4

Numerous systems for error identification have been developed, the most promising of which combines analysis of automated data, medical record review, and active event reporting.5 The Australian Patient Safety Foundation and the United Kingdom's National Patient Safety Agency have implemented systems for event surveillance. In the United States, legislation is pending that would facilitate sharing reports and analysing errors without fear of increased litigation. This legislation would complement the actions taken by states. Currently, 21 states have mandatory reporting systems; but under-reporting remains a serious challenge.6

Although implementation of reporting systems will shed additional light on threats to safety, unresolved dilemmas remain regarding the appropriate balance between breadth and depth of reports, confidentiality, and the public's right to know, and whether to focus on actionable processes (that is, medical errors) or harms irrespective of antecedents.7 The overarching challenge, ripe for transnational collaboration, is identifying the most effective strategies for translating information on errors and near misses into safety for patients.

Numerous resources can help organisations in implementing evidence based safety interventions. The United States Agency for Healthcare Research and Quality produced an evidence report that reviewed 79 patient safety practices, as well as the first online journal on patients' safety, WebM&M ( The journal highlights five cases of errors or near misses each month, along with the strategies deployed to prevent recurrence and commentary from national experts.8 Some successful practices in reducing errors have been highlighted by the John M Eisenberg patient safety award of the Joint Commission on Accreditation of Healthcare Organisations. The National Patient Safety Agency has produced a toolkit that includes learning modules to help in analysing root causes.9

The implementation of information technology also offers great promise. Recently, the Institute of Medicine called for the continued development of a national health information infrastructure, providing real time access to complete information on patients and decision support.10 In 2004 the Agency for Healthcare Research and Quality plans on spending nearly $60m to support research and implementation of information technology that improves patient safety. Moreover, the United States Department of Health and Human Services and other purchasers are collaborating to identify incentives to accelerate the use of health information technology. These efforts, coupled with the transformation of the NHS through implementation of a national information technology infrastructure, will accelerate the impact on safety.

Despite these steps, we have not seen substantial progress in one critical area—culture—that has the greatest potential to produce sustainable improvements in safety. Both the National Patient Safety Agency and the National Quality Forum, a US public and private consensus development organisation, list a safety culture among its priorities for a safer healthcare system. Culture encompasses commitment to open communication about errors to encourage reporting and analysis. Prerequisite to such openness is the recognition that errors usually result from an imperfect system and that dealing with them through individual blame only discourages reporting. Ensuring openness also entails alignment between the legal and medical staffs; but the medical profession has not embraced the concept that medical error is a systems issue. One study found that 55% of doctors and the public blamed individual health professionals for serious medical errors.11 Similarly, open communication around errors is still not the norm. Liam Donaldson, England's chief medical officer, recently termed the safety culture “weak.”12 Although the concept of culture may be somewhat nebulous, methods exist to assess where an organisation stands.

From the first day of medical school, we are taught: Primum non nocere— above all, do no harm. In recent years, healthcare professionals have awoken to the harm our patients are experiencing despite our best intentions. Through the work of many worldwide, our understanding of the epidemic has increased tremendously and is beginning to pay dividends. More substantial and sustainable improvements, however, will occur only when healthcare organisations truly commit to safety through open communication that does not blame individuals but identifies and addresses flaws in systems.


  • Competing interests None declared.


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