Use Error

Distinction from task analysis

Both methods are necessary, but they work at different levels. The task analysis looks at the interaction between a user and the product and breaks it down into action steps. Workflow analysis looks one level higher: at the process in which this interaction is only one episode among several, such as the course of a surgical procedure, medication administration on a ward, a dialysis session, care in the emergency department or the daily routine of use in the home environment.

The practical consequence: A product can be excellently usable on its own and still lead to errors in the real workflow, because it demands attention at the wrong moment, because it has to fit into a device landscape it was not designed for, or because the person operating it is not the same person who prepared it. Such problems are structurally invisible at the level of task analysis.

What is examined

A workflow analysis systematically captures the dimensions that reach beyond the individual operating procedure:

• Actors and division of roles: who prepares, who operates, who checks, who documents. It is rarely the same person.
• Handover points: shift changes, patient transfers, changes of responsibility. Handovers lose contextual knowledge that would be needed for safe continued operation.
• Parallel devices and information sources: other medical devices, the patient record, monitoring, paper documentation. This is where risks of confusion and misassignment arise.
• Interruptions: phone calls, alarms from other devices, questions from colleagues. Actions that are interrupted and later resumed are a known driver of use errors, because a step is easily skipped on resumption.
• Time constraints: how much time is actually available for a step and what happens when it is not enough.
• Environmental conditions: lighting, noise level, space, hygiene requirements; captured as use environment factors.

Data collection methods

The core of workflow analysis is on-site observation. Job shadowing and accompanying staff over a complete shift or an entire treatment cycle yield insights that cannot be obtained by asking, simply because well-practiced routines have become self-evident to those involved and thus impossible to put into words.

Other methods have proven useful as complements: interviews with representatives of all roles involved, not only the main user group; the analysis of standard clinical procedures and hygiene requirements as a target-versus-actual comparison; and the visualization of the workflow as a process diagram with separate swim lanes for each role, which makes handovers and parallelism immediately visible. Event and time logs help to quantify the frequency and duration of interruptions instead of estimating them.

Typical findings

The most frequent and at the same time most valuable finding of a workflow analysis is deviations between the prescribed and the lived workflow. Where users regularly bypass, shorten or replace a documented step with their own solution, this is usually not a discipline problem but an indication that the intended workflow collides with real conditions. Such workarounds are design information, and evaluating them is also the point at which the distinction from abnormal use has to be drawn carefully.

Integration problems that are not visible on the individual device also regularly come to light: alarm cascades from several devices that together create a stressful situation; risks of confusion between similar connectors or containers from different manufacturers; or states that can no longer be clearly recognized after an interruption. Without a workflow analysis, such findings appear only in the field, through post-market surveillance and incident reports, and thus at the most expensive point in time. Workflow findings often form the basis for identifying hazard-related use scenarios and critical tasks within the use-related risk analysis in the first place. Risks arising from handovers, role changes or parallel devices are otherwise easily overlooked at the level of the individual operating action.

Effect on specification and test realism

The results of a workflow analysis take effect in two places. First, in the use specification: user groups, context of use and environmental conditions are justified empirically instead of assumed. Second, in the study design: task sequence, realistic distractions, team constellation and the initial state of the device can only be designed realistically if the real workflow is known.

A summative test in which participants work undisturbed and in an ideal sequence, while real use is shaped by interruptions and parallel tasks, examines a use behavior that does not exist in this form. Workflow analysis is therefore one of the most effective measures against a summative evaluation that is formally correct but substantively unreliable.

Regulatory relevance

IEC 62366-1 requires the use specification to describe the intended user groups and the use environment and to derive use scenarios from them. Workflow analysis is the methodological way to base this information on observation instead of assumptions; IEC 62366-2 gives further guidance on context analysis in this regard.

The EU MDR (Medical Device Regulation, Regulation (EU) 2017/745) requires in Annex I the reduction of risks arising from ergonomic features and the intended use environment; workflow analysis provides the empirical basis for describing this use environment realistically instead of assuming it. The FDA Human Factors Guidance expects that the use environment is appropriately represented in human factors validation testing, including the factors that realistically make use more difficult. Here, too, workflow analysis is the foundation: without it, it cannot be justified which environmental conditions must be recreated in the test setup and which may be neglected.

Frequently asked questions (FAQ)

Sources

• IEC 62366-1:2015+AMD1:2020, Medical devices, Part 1: Application of usability engineering to medical devices
• IEC/TR 62366-2:2016, Medical devices, Part 2: Guidance on the application of usability engineering to medical devices
• Regulation (EU) 2017/745 on medical devices (MDR)
• FDA Guidance: Applying Human Factors and Usability Engineering to Medical Devices

Related terms

• Context of use
• Emergency use
• Use scenario
• Task analysis
• Use environment factors
• Use specification
• Contextual interview

← Back to the wiki overview