Durgical
EN
Open the app

Human factors engineering: designing for real people

  • Concept

Tired, busy people make predictable mistakes. Good design makes the right action easy and the wrong one hard or impossible.

Updated

What it is

Human factors engineering studies how people interact with tools, tasks and environments, and designs them around human limits: memory, attention, fatigue and stress. In health care it shapes devices, medication labels, the electronic record, alarms and workflows. Its central lesson is that fixes built into the design (forcing functions, automation) prevent more errors than fixes that rely on people remembering (training, policies).

Key facts

  • Workplace design: lighting, noise, interruptions and layout affect error rates; a quiet zone for preparing medicines reduces interruptions.
  • Device design: a pump whose screen hides the units or uses tiny decimal points invites programming errors.
  • Automation helps, but it creates new errors: over-trust in the device and skill loss when it fails.
  • Telemetry and monitor alarms: most alarms in a monitored unit do not need action; tuning limits, delays and electrode care cut them.
  • Situational awareness is lost through task overload, fatigue and fixation on one problem; briefings and shared displays help keep it.
  • Usability testing with real users in realistic scenarios finds design traps before they reach patients.

Worked example

  1. A made-up 30-bed telemetry unit records 600 monitor alarms a day; an audit finds only 60 (10 %) needed a clinical action.
  2. With 6 nurses, each hears about 600 ÷ 6 = 100 alarms a day, of which 90 are noise: staff begin to silence alarms without looking.
  3. Changes: patient-specific limits, a short delay for self-correcting alarms, daily electrode change and stopping telemetry when no longer indicated.
  4. After 2 months: 240 alarms a day, 57 of them actionable. Actionable share = 57 ÷ 240 ≈ 24 %, up from 10 %.
  5. Each nurse now hears 240 ÷ 6 = 40 alarms a day, 60 % fewer than before.
  6. Balancing check: actionable events found fell only from 60 to 57 a day, and no missed arrests were found in the review.
  7. Reading: cutting noise made each alarm more likely to mean something, which is what restores a fast response.

The hierarchy of interventions, strongest first

LevelHow it worksExampleDepends on memory?
Forcing functionThe wrong action cannot be doneEnteral feeding connector that does not fit an IV lineNo
AutomationThe machine does or checks the stepBarcode scan of wristband and medicine before giving itLittle
Standardisation and checklistsOne way of doing it, with promptsOne concentration of a high-alert infusion; surgical checklistPartly
Training, policies, warningsPeople are told what to doA memo or a poster about double-checkingEntirely

Markers

  • Forcing function
  • Alarm fatigue
  • Situational awareness
  • Usability

In practice

  • When an error seems ‘careless’, ask what in the design made it easy; report the design problem, not only the person.
  • Never disable or widen alarm limits without a reason and a plan; adjust them to the patient and document it.
  • Protect medicine preparation from interruptions; use a do-not-disturb signal or space if your unit has one.

Common traps

  • Training is the best fix for a recurring error: training sits at the bottom of the hierarchy; a forcing function or automation works even when people are tired
  • More alarms mean more safety: non-actionable alarms cause alarm fatigue, and real alarms are then missed or answered late
  • Automation removes human error: it moves error to new places, such as over-trust, workarounds and loss of skill when the system fails
  • A checklist is a forcing function: a checklist still depends on someone using it; a forcing function makes the wrong action impossible

Test yourself

Think of the answer, then tap the question to see it.

Oral liquid medicine is twice given into an IV line. Which fix is strongest: retraining, a warning label or oral syringes that cannot connect to IV ports?

Oral syringes that cannot connect: a forcing function. Labels and training depend on attention at the moment of the error.

A nurse silences a monitor alarm without looking and misses a real arrhythmia. Beyond the individual, what should the review examine?

Alarm burden and design: how many alarms are non-actionable, whether limits are tailored and whether telemetry is still indicated. Alarm fatigue is a system problem.

Why can a barcode medication system create new errors even though it is a strong barrier?

People may trust it blindly or work around it (scanning a spare wristband, overriding alerts) when it slows them down; design and workflow must fit the work.

During a long resuscitation, the leader focuses on the airway and misses that no one has checked the rhythm for 6 minutes. What has been lost and how can a team keep it?

Situational awareness. A leader who stays hands-off, regular out-loud summaries and a timekeeper help the team keep the whole picture.

Add to my cards (4)

Elsewhere in DurgiPedia

Connections

← All of epidemiology

Educational content on epidemiological and statistical methods; the examples use made-up numbers to teach the calculation. Not a clinical or legal guide: current Colombian rules are in the official sources.

Sources: Patient safety curriculum guide: multi-professional edition (World Health Organization, 2011); Effects of workload, work complexity, and repeated alerts on alert fatigue in a clinical decision support system (BMC Medical Informatics and Decision Making (PubMed Central), 2017); Patient safety (World Health Organization, 2023).

Spotted a mistake?We are always improving for the community. Tell us what is wrong and our team will review it.Report a mistake

Representative model, for educational purposes only.