It is common knowledge that adverse events occur regularly in healthcare settings. While many of these events have no long term consequences or reach public attention, serious failures of modern medicine can be mortifying for patients, their loved ones and their healthcare practitioners.

The risks and adverse events associated with hospital practice in Australia were documented in the Quality in Australian Health Care Study.¹ In this issue of the Journal, Bhasale et al report the incident-monitoring study in Australian general practice² -- the first reported systematic research of adverse events in general practice. Comparable research in general practice in the United Kingdom has focused on significant event auditing to provide clinical material for facilitated case discussions of specific health issues,^3-5 while earlier Australian research considered the educational implications of critical events.^6,7

This long-awaited study provides valuable insights into some key quality issues for Australian general practice. The authors report on 805 critical incidents submitted by anonymous, volunteer general practitioners (GPs). Each adverse event was described by the GP, who also estimated its potential for harm, immediate consequences, mitigating and contributing factors, and preventability.

The study is based on an incident-monitoring method developed for anaesthetics,⁸ but is critical incident monitoring appropriate for general practice? We do not know whether some GPs reported only those incidents which they believed had serious potential outcomes and which they thought might be of particular interest to the researchers. Perhaps individual GPs deliberately avoided reporting cases which might have shown them in a bad light; perhaps the reverse is true. Bhasale et al acknowledge some of the methodological issues raised by their study: the non-random sample of GPs and sole reliance on GP selection and self-reporting of critical incidents. The low volunteer rate (42%) and the bias towards relatively experienced GPs may indicate that many GPs are reluctant to be involved in research of this nature.

A key feature of general practice is the presentation of undifferentiated medical problems, so diagnostic uncertainty is present in many initial consultations. Vigilance is required to ensure that, where possible, such early symptoms and signs are investigated in an appropriate and timely manner, while the risks and costs of unnecessary investigations are avoided. This is often a very difficult balance to achieve in clinical practice.

However, Bhasale et al report errors that arise from basic oversights rather than uncertainty: failure to review the patient's history, inadequate recognition of symptoms and signs, complications arising from drug treatments, poor legibility of prescriptions, and abnormal test results being missed or not acted upon by the practitioner. These problems may relate directly to lapses at the individual practitioner level or at the practice level. Some of these problems may soon be solved by the decision support features of computerised medical record systems and computerised prescribing.

Of particular concern in the study is that 4% of the incidents were reported to result in a patient's death and 17% were reported to result in major harm, either organic or psychological. However, there is no clearly demonstrated causal link between the critical events and the deaths reported. The authors state explicitly that this type of study cannot measure the prevalence of potential or actual harmful incidents in general practice. We emphasise that it would be inappropriate to make any such generalisations from the data provided.

Bhasale et al highlight key areas for quality improvement in Australian healthcare. Many of these have been alluded to in previous commentaries on hospital care,^9,10 and include problems with communication between healthcare professionals and the potential dangers for patients as they move across boundaries in the healthcare system. Failure to provide hospital discharge information to GPs is a longstanding and fundamental flaw in the system. At present, there are no adequate mechanisms for reporting critical incidents or for providing feedback about outcomes, both negative and positive, to all healthcare professionals. It is encouraging to note that integration of general practice into the healthcare sector is currently the focus of considerable research effort in Australia.

Problems of communication with patients, particularly the elderly, those with poor English language skills and those with mental health problems, are also highlighted by Bhasale et al. These groups must be deemed populations at particularly high risk of adverse events. General practitioners need support and awareness-raising strategies to enable them to better address the healthcare needs of these groups and other potentially high risk groups, such as people of Aboriginal or Torres Strait Islander origin and people with multiple chronic health problems.

There are additional areas where quality improvement initiatives could take place. These include enhancement of the existing process for accreditation of general practices, refinement of the Quality Assurance and Continuing Education program of the Royal Australian College of General Practitioners, and development of systems for reporting critical incidents and negative outcomes to all healthcare professionals.

It must be remembered that, even if the best available evidence were applied to every clinical decision made by healthcare workers in this country, critical incidents would still occur. Unfortunately, the current system does not assist individual practitioners to express concerns about errors they witness or make as, in the main, peer support is lacking when adverse events occur in general practice. Clearly, there is a need for processes that allow personal grieving while providing a systematic response to problems.

Michael R Kidd
Professor of General Practice, University of Sydney
and Member of the New South Wales Ministerial Advisory Committee
on Quality in Health Care, Sydney, NSW

Bronwyn M Veale
Senior Research Fellow, Department of Evidence-Based Care and General Practice
Flinders University of South Australia, Adelaide, SA

1. Wilson RM, Runciman WB, Gibberd RW, et al. The Quality in Australian Health Care Study. Med J Aust 1995; 163: 458-471.
2. Bhasale AL, Miller GC, Reid SE, Britt HC. Analysing potential harm in Australian general practice: an incident-monitoring study. Med J Aust 1998; 169: 73-76.
3. Pringle M, Bradley CP, Carmichael CM, et al. Significant event auditing. A study of the feasibility and potential of case-based auditing in primary medical care. Occas Pap R Coll Gen Pract 1995; 70: I-VIII.
4. Bradley CP. Turning anecdotes into data -- the critical incident technique. Fam Pract 1992; 9: 98-103.
5. Robinson LA, Stacy R, Spencer JA, Bhopal RS. Use facilitated case discussions for significant event auditing. BMJ 1995; 311: 315-318.
6. Sim MGB, Kamien M, Diamond MR. From novice to proficient general practitioner: a critical incident study. Aust Fam Physician 1996; 25 Suppl 2: S59-S64.
7. Holmwood C. How do general practice registrars learn from their clinical experience? A critical incident study. Aust Fam Physician 1997; 26 Suppl 1: S36-S40.
8. Williamson JA, Mackay P. Incident reporting. Med J Aust 1991; 155: 340-344.
9. McNeil JJ, Leeder SR. How safe are Australian hospitals? Med J Aust 1995; 163: 472-475.
10. Wilson RM, Harrison BT. Are we committed to improving the safety of health care? Med J Aust 1997; 166: 452-453.

Home | Issues | eMJA shop | My account | Classifieds | More... | Contact | Topics | Search

The Medical Journal of Australia    eMJA