Research — Why clinical errors happen
Errors are rarely failures of knowledge.
International research points to the same conclusion: serious clinical errors arise from flawed clinical reasoning operating inside systems that fail to detect and correct it.
01 — Two failures, together
Reasoning and systems fail at the same time.
A landmark analysis of 100 diagnostic errors in internal medicine — 33 of them fatal — found cognitive and system-related factors in most cases, usually together. The most common cognitive failure was premature closure: ceasing to consider alternatives once an initial diagnosis was reached.1
Deficits in medical knowledge were uncommon. The problem is less what clinicians know than how a diagnosis is formed, tested and revised — and whether the system around them catches it when it goes wrong.
Contributing factors in diagnostic error
Share of cases analysed (n = 93). Categories overlap — most cases involved both.
Source: Graber, Franklin & Gordon, Archives of Internal Medicine, 2005.1
02 — Where it goes wrong
Most errors begin at the bedside.
In 190 diagnostic errors in primary care, the breakdown occurred most often in the clinical encounter — taking the history, examining the patient, and deciding which tests to order. 86.8% of the errors had the potential for moderate to severe harm.2
Malpractice data point the same way: in more than 4,700 diagnosis-related claims, 58% involved failures in the initial diagnostic assessment.3 In hospitals, a study of 2,428 adults who died or were transferred to intensive care found a diagnostic error in 23% — most arising in patient assessment or in ordering and interpreting tests — and an error contributed to 6.6% of deaths.4
Where diagnostic errors originate
Share of 190 primary care errors involving each breakdown. Errors may involve more than one.
Darker bars: breakdowns within the clinical encounter (78.9% of errors overall).
Source: Singh et al., JAMA Internal Medicine, 2013.2
03 — The cognitive traps
How a reasonable mind reaches the wrong conclusion.
Cognitive errors are not carelessness. They are predictable patterns in how experienced clinicians reason under time pressure and uncertainty — which is why they recur, and why they can be identified.1,16
Premature closure
Accepting a diagnosis before it has been verified, and ceasing to consider reasonable alternatives once an initial explanation has been reached.
Tested in CASE: Interpretation · Decision Integrity
Anchoring
Fixing on early information — a referral label, a first impression, a prior diagnosis — and failing to adjust as the clinical picture develops.
Tested in CASE: Interpretation · Decision Integrity
Failure to revise
New observations, results or deterioration that should prompt reconsideration are explained away to fit the working diagnosis, and the response is delayed.
Tested in CASE: Detection · Response Latency
04 — When systems don't catch it
Preventable deaths follow recognisable failures.
A case-record review of adult deaths in English acute hospitals judged 5.2% more likely than not to have been preventable — an estimated 11,859 deaths a year. The leading problems in care were failures to monitor the patient, diagnostic error, and inadequate drug or fluid management.5
In the United States, The Joint Commission reviewed 1,441 sentinel events in 2022, 20% of which resulted in death. The leading causes were failures of communication, teamwork, and consistently following policy.6
Problems in care behind preventable deaths
Share of preventable deaths in which each problem was identified.
Source: Hogan et al., BMJ Quality & Safety, 2012.5
05 — The scale
A burden measured in hundreds of thousands. Much of it preventable.
Estimated each year in the United States from diagnostic error. Vascular events, infections and cancers account for 75% of serious harms.7
In a 2018 sample of 2,809 hospital admissions; 22.7% of the events were preventable and 32.3% were serious.8
Pooled international estimate (6%); 12% of that harm is severe or fatal. Highest in intensive care and surgery.9
Sources: Newman-Toker et al., BMJ Qual Saf 2023; Bates et al., NEJM 2023; Panagioti et al., BMJ 2019.
06 — Australia: when harm reaches the regulator
Most concerns end without regulatory action.
Research by Marie Bismark, David Studdert and colleagues at the University of Melbourne — the largest studies of notifications to Australia's national health practitioner boards — found that the large majority are closed without restrictive action, and that concerns raised by patients and families are the least likely to lead to it.10,11
Regulators assess the conduct and performance of individual practitioners. They are not designed to reconstruct a clinical episode or to examine the system in which care was delivered.
What happens to complaints
Outcomes of complaints under the national scheme, 2012–2013 cohort.
In NSW's co-regulatory system, 70% of complaints were closed with no further action.
Source: Walton et al., Australian Health Review, 2020.12
Notifications from employers were 6.9 times more likely to lead to restrictive action than those from patients or relatives; from other practitioners, 2.9 times.10
Notifications that resulted in suspension or cancellation of registration, in a cohort of 8,307 notifications.10
Of 13,939 finalised notifications ended in restrictive action; 3.7% attracted immediate action while under investigation.11
Rise in average time to complete an investigation between 2018 and 2023; one in four investigations had been open for more than two years.13
Notifications received by Ahpra in 2024/25, up 19% on the previous year. 1.4% of closed notifications resulted in loss of registration or disqualification.14
Of notifications that ended in restrictive action had no interim restriction imposed while the matter was investigated.11
07 — The missing data
The absence of evidence is itself evidence.
The Quality in Australian Health Care Study, published in 1995, reviewed hospital admissions across New South Wales and South Australia. It found adverse events in 16.6% of admissions, and judged around half of them preventable.15,17
It has not been repeated. Hospital harm has since been re-measured in large-scale studies in England and the United States; Australia has no current national measure of how many patients are harmed, or die, as a result of adverse events in care.
National and large-scale studies of hospital harm
Selected studies cited on this page, by year of publication.
08 — What this means for investigation
Each finding points to a question an investigation must answer.
The CASE framework is built around the failure patterns the research identifies. Each dimension tests for one of them.
| The research finding | The investigative question | CASE dimension |
|---|---|---|
| Failures in history-taking, examination and monitoring | Did the clinical environment recognise the signs that something required attention? | 01 Detection |
| Premature closure and anchoring | Was the available evidence correctly interpreted, and were alternatives considered? | 02 Interpretation |
| Delays in ordering and interpreting tests | Were responses initiated within clinically relevant timeframes? | 03 Response Latency |
| Cognitive factors present in most diagnostic errors | Were decisions reasonable on what was knowable at the time, without hindsight? | 04 Decision Integrity |
| Communication and teamwork failures in sentinel events | Were risks communicated, acknowledged and acted upon through escalation pathways? | 05 Escalation Integrity |
| Most concerns closed without regulatory action | Do the record and subsequent accounts reflect what happened, and were disclosure obligations met? | 06 Governance & Disclosure |
Confidential enquiries
Concerned about a clinical event, adverse outcome or patient safety risk?
CID offers confidential preliminary enquiries and defined-scope assessments to help determine an appropriate course of action.
References
- Graber ML, Franklin N, Gordon R. Diagnostic error in internal medicine. Arch Intern Med 2005;165(13):1493–1499. doi:10.1001/archinte.165.13.1493
- Singh H, Giardina TD, Meyer AND, et al. Types and origins of diagnostic errors in primary care settings. JAMA Intern Med 2013;173(6):418–425. doi:10.1001/jamainternmed.2013.2777
- CRICO Strategies. 2014 Annual Benchmarking Report: Malpractice Risks in the Diagnostic Process. Cambridge, MA: Risk Management Foundation of the Harvard Medical Institutions; 2014.
- Auerbach AD, Lee TM, Hubbard CC, et al. Diagnostic errors in hospitalized adults who died or were transferred to intensive care. JAMA Intern Med 2024;184(2):164–173. doi:10.1001/jamainternmed.2023.7347
- Hogan H, Healey F, Neale G, Thomson R, Vincent C, Black N. Preventable deaths due to problems in care in English acute hospitals: a retrospective case record review study. BMJ Qual Saf 2012;21(9):737–745. doi:10.1136/bmjqs-2011-001159
- The Joint Commission. Sentinel Event Data 2022 Annual Review. Oakbrook Terrace, IL; 2023.
- Newman-Toker DE, Nassery N, Schaffer AC, et al. Burden of serious harms from diagnostic error in the USA. BMJ Qual Saf 2024;33(2):109–120. doi:10.1136/bmjqs-2021-014130
- Bates DW, Levine DM, Salmasian H, et al. The safety of inpatient health care. N Engl J Med 2023;388(2):142–153. doi:10.1056/NEJMsa2206117
- Panagioti M, Khan K, Keers RN, et al. Prevalence, severity, and nature of preventable patient harm across medical care settings: systematic review and meta-analysis. BMJ 2019;366:l4185. doi:10.1136/bmj.l4185
- Spittal MJ, Studdert DM, Paterson R, Bismark MM. Outcomes of notifications to health practitioner boards: a retrospective cohort study. BMC Med 2016;14:198. doi:10.1186/s12916-016-0748-6
- Bradfield OM, Bismark MM, Studdert DM, Spittal MJ. Characteristics and predictors of regulatory immediate action imposed on registered health practitioners in Australia: a retrospective cohort study. Aust Health Rev 2020;44(5):784–790. doi:10.1071/AH19293
- Walton M, Kelly PJ, Chiarella EM, Carney T. Management and outcomes of health practitioner complaints in Australia: a comparison of the national and New South Wales systems. Aust Health Rev 2020;44(2):180–189. doi:10.1071/AH18262
- Australian Government Department of Health, Disability and Ageing. Independent Review of Complexity in the National Registration and Accreditation Scheme: Final Report (led by S. Dawson). Canberra; 2025.
- Australian Health Practitioner Regulation Agency. Annual Report 2024/25: Notifications. Melbourne: Ahpra; 2025.
- Wilson RM, Runciman WB, Gibberd RW, et al. The Quality in Australian Health Care Study. Med J Aust 1995;163(9):458–471.
- Scott IA, Crock C. Diagnostic error: incidence, impacts, causes and preventive strategies. Med J Aust 2020;213(7). doi:10.5694/mja2.50771
- Ranasinghe I, Hossain S, Ali A, et al. SAFety, Effectiveness of care and Resource use among Australian Hospitals (SAFER Hospitals): a protocol for a population-wide cohort study of outcomes of hospital care. BMJ Open 2020;10(8):e035446. doi:10.1136/bmjopen-2019-035446