Clinical systems · Human factors · Performance under pressure
When Adaptation Becomes the Safety Strategy
When routine care depends on people quietly compensating for weak design, resilience can start to hide risk.
How clinical systems depend on adaptation
Clinical systems can look coherent on paper, with mapped processes, defined escalation pathways and equipment expected to be available and ready to use. During a shift, those arrangements have to accommodate interruptions, missing information, competing priorities and the needs of the patient in front of the team.
Clinicians adapt by moving equipment, changing sequencing and finding ways to keep care moving when the expected process does not fit. Some adjustments respond to circumstances that could not have been fully anticipated; others compensate for a problem the team already knows well, such as an awkward layout, an unreliable connection or information that is never where it needs to be.
Adaptation is part of clinical expertise because no process can anticipate every patient or circumstance. The concern is when a known design weakness requires the same extra effort on successive shifts. The service then depends on people knowing how to work around it, although that dependence may be absent from the process it has documented.
A successful outcome can conceal how much extra work was needed to achieve it.
The gap between the process and the work
A policy describes an intended process, but it cannot capture every condition in which that process will be used. Human factors offers a useful distinction between work-as-imagined and work-as-done: how a task is expected to happen, and what people actually do to complete it in the circumstances they encounter.
A team departing from the written process may have found a better sequence, recognised that a patient’s needs require a different approach, or compensated for missing equipment. Those situations require different responses. Restoring compliance without understanding the reason could remove a useful improvement, while writing every workaround into the procedure could make an avoidable burden permanent.
Incident review provides only part of this picture. Examining routine work with the team can show what had to be changed, which steps depended on a particular person's knowledge, and what was left for someone else to finish. Without that detail, a completed task may be taken as reassurance about the process, even when the team had to compensate for it.
The cost of “making it work”
Repeated workarounds draw on time and attention that the formal process may not account for. Someone remembers which equipment needs an extra check, who can resolve a missing referral, or where information is kept when it cannot be found in the expected place. An experienced team may make this look straightforward, while someone new has to discover the unwritten steps as well as learn the stated process.
New staff learn the workaround through corridor conversations, and the extra step becomes part of the job. Over time, a service can develop an unofficial way of operating alongside its documented one. The written process remains the basis for training and audit, while local knowledge fills the gaps in practice.
When adaptation becomes routine, variability becomes embedded.
Different teams can solve the same recurring problem in different ways, without those arrangements ever being compared or agreed. What happens next may depend on who is on shift and which workaround they have learned. A process that appears dependable with a familiar team can become harder to use when staff rotate or workload rises.
This matters because teams may be inheriting different solutions to the same unresolved weakness. Local knowledge should inform the design, including adaptations that help clinicians respond to a patient’s needs. Where the normal route is unreliable, however, teaching more people the workaround leaves the reason for it untouched.
Digital systems: where the extra work goes
Consider information recorded on paper because a tablet will not sync, then entered into the clinical system later. Paper may be a necessary fallback, but someone still has to transfer the information, check its accuracy and make sure it reaches the people who need it. That work belongs in the design of the fallback, even if it happens after the immediate episode of care.
A planned fallback makes those responsibilities explicit, whereas an informal response to a recurring fault may leave them assumed. An audit of completed records might show that the information was eventually entered, but miss the period when it was unavailable to the next team and the work needed to put it there.
Evaluating the system therefore means following the information from capture through to its use by the next team, including what happens when a connection fails. Improving a screen may help one part of that work while leaving another clinician reconciling two records or chasing information that should already be available.
What redesign actually looks like
Consider a handover in which the agreed structure is clear, but the receiving clinician is interrupted and has to enter information into a record organised in a different order. More training in the spoken structure may help recall, but it will leave the competing task and the mismatch with the record in place.
Watching both sides of the handover can establish where the difficulty occurs and whether it comes from interruption, sequencing, duplication or access to information. The change to test should follow from what that examination shows:
- Agree which competing tasks can wait during the initial handover, while keeping a clear route for urgent concerns.
- Test a record layout that follows the spoken handover, so staff can document without reorganising information as they listen.
- Establish why information is entered twice before deciding whether reliable transfer can replace one of those entries.
- Make key risks easy for the receiving clinician to find and confirm.
The test is whether the change reduces the identified difficulty without creating another one elsewhere. A shorter handover may look more efficient while leaving the receiving clinician to reconstruct the history afterwards. Evaluation needs to follow the work far enough to see that consequence, including under pressure and with staff who were not involved in developing the change.
Standardising the sequence may reduce unnecessary variation, provided there is room for clinical judgement when urgency changes what the team needs to do. The design should make those points of discretion clear so that consistency supports the handover rather than becoming another obstacle.
What the organisation chooses to rely on
The team using a workaround may have little authority over the conditions that make it necessary. They can change their sequence or make another phone call, but may not be able to change the equipment specification, the record system or an agreement between services. Asking them to solve the problem locally again leaves those decisions untouched.
An arrangement may serve a legitimate operational purpose while still creating work that could be reduced or avoided. The review needs to establish which requirements must be preserved and which parts of the process can change, including whether the workaround offers an improvement worth retaining.
Where an immediate fix is not available, someone needs to agree how the temporary arrangement will work, who will review it and what would make it unacceptable to continue. Without that ownership, the next team inherits the same problem with no clear route to a decision.
The review should then return to the work to establish whether the extra step has gone, whether information is available when needed and whether a less familiar team can use the process without being taught a separate, unofficial version. Those are practical ways to examine whether the redesign has addressed the problem that prompted it.
Better design doesn’t remove pressure. It reduces unnecessary friction.
Further reading: NHS England’s discussion of learning from everyday work provides background on observing work as done and involving frontline teams in improvement.
Financial disclosures. Any relevant commercial interests or conflicts of interest are stated at the top of the post to which they apply. Where none are stated, none apply to that post.