In medicine, we are trained to think in terms of risk, probability, and consequence. We learn early that small errors can cascade into catastrophic outcomes. Yet two powerful and deeply human forces routinely undermine our best intentions: the Dunning–Kruger effect and the normalisation of deviance, a term coined by sociologist Diane Vaughan.
From my perspective as a medical professional, these two phenomena represent not abstract psychological concepts, but daily clinical hazards. When they intersect, they create a quiet but dangerous pathway to patient harm.
The Dunning–Kruger effect describes a cognitive bias in which individuals with limited competence overestimate their ability because they lack the insight to recognise their deficiencies. In medicine, this is particularly concerning because clinical environments often reward decisiveness and confidence. Hesitation can be misinterpreted as incompetence, and certainty can feel reassuring to colleagues and patients alike.
Early in training, a medical student or junior doctor may master a small piece of knowledge and feel an inflated sense of understanding. A handful of successful procedures or correct diagnoses can create a powerful illusion of competence. This is sometimes described as the “peak of Mount Stupid” phase, where confidence rises faster than skill.
In healthcare, this overconfidence can translate into real harm. A clinician who believes they “already know this” may:
The most dangerous aspect is not ignorance itself; it is ignorance combined with unawareness. When clinicians do not realise what they do not know, they are less likely to seek help. In high-risk settings such as emergency medicine, anaesthesia, or surgery, this can mean delayed recognition of complications or inappropriate management decisions.
Ironically, as competence develops, confidence often dips. Experienced clinicians understand complexity. They appreciate atypical presentations and diagnostic uncertainty. This “valley of despair” is uncomfortable but protective, it fosters humility, double-checking, and collaborative decision-making. The danger lies in clinicians who never pass through that valley because they never recognise their limits.
If the Dunning–Kruger effect is an internal cognitive distortion, normalisation of deviance is a cultural one.
Normalisation of deviance occurs when departures from established standards gradually become accepted as routine because they do not immediately lead to adverse outcomes. In healthcare, this process is insidious and often invisible.
Consider hand hygiene. A clinician skips handwashing or performs a cursory procedure once because they are in a hurry and the patient appears low-risk. No infection occurs. The shortcut feels harmless. Over time, perhaps a 20 year career, the omission becomes habitual. The deviation from protocol is no longer experienced as a deviation at all.
The same pattern can appear in:
Each shortcut is reinforced by success. “Nothing bad happened” becomes misinterpreted as “it’s safe”. In reality, we are often protected by layers of redundancy and luck. When those layers align unfavourably, catastrophe can emerge suddenly and dramatically.
Healthcare is particularly vulnerable to normalisation of deviance because of constant production pressure. We are asked to see more patients, reduce waiting times, minimise costs, and maintain throughput. Efficiency quietly competes with safety. When clinicians feel squeezed for time, protocols begin to look like obstacles rather than safeguards.
Over time, the gap between “work as imagined” (policies, guidelines, accreditation standards) and “work as done” (what actually happens on the ward at 3 a.m.) widens. This gap is where risk accumulates.
Individually, overconfidence and normalised deviance are dangerous. Together, they are combustible.
Imagine a junior clinician who overestimates their skill in airway management. They have successfully intubated several patients and feel comfortable. In a busy emergency department, pre-intubation checklists are sometimes abbreviated. Monitoring alarms are occasionally silenced to reduce noise. Equipment checks are assumed rather than verified.
The clinician feels confident. The culture tolerates shortcuts. The patient has difficult anatomy. The oxygen saturation falls rapidly. Backup equipment was not fully prepared. What was previously a series of minor deviations may now converge into a life-threatening event.
This is how harm occurs—not through dramatic recklessness, but through accumulated normalisation layered on top of inflated confidence.
Another example is medication administration. A resident may feel comfortable adjusting complex medication doses independently. Pharmacy verification is delayed due to workload. A nurse trusts the physician’s confidence. No one challenges the decision. A dosage error slips through multiple layers of defence because each layer assumed the other was correct.
In such cases, hierarchy amplifies the problem. Junior staff may hesitate to question a confident physician. If that physician is unknowingly operating at the peak of overconfidence, the absence of challenge becomes part of the failure chain.
In healthcare, the margin for error is thin. The consequences of normalised deviance and overconfidence might include:
These events are rarely caused by a single reckless act. Instead, they reflect systemic drift combined with human cognitive bias.
Explore our upcoming accredited courses and discover why so many healthcare professionals recommend Healthcare Skills Training International to their colleagues.
backE-mail:
Tel: +44 (0) 141 648 8400
Healthcare Skills Training International Ltd
West of Scotland Science Park
Kelvin Campus
Glasgow G20 0SP
