The phrase “pyjama time” entered the literature as a joke and stayed as a metric. Hours spent in the record after the last patient has left is now a reasonable proxy for how badly a system fits the work it was bought to support.
Where the minutes actually go
Audit logs are unsentimental. In most primary care deployments the single largest category is not free-text authoring; it is navigation — opening, searching, scrolling and switching between contexts to assemble information that already exists in the record.
The second largest is usually reconciliation work created by an interface: duplicate problems, medications arriving from two sources, results needing manual association with an order.
If a clinician has to visit four screens to answer one question, the answer is not the problem. The route is.
Interventions that measurably help
- Collapse the pre-visit picture into one view, including the gap list, so review happens once.
- Default the note to the structure the measure needs, rather than asking for it twice.
- Route results to the person who can act, not only to the person who ordered.
- Kill inbox messages that require no decision — they are a tax levied in attention.
Measuring whether it worked
Track time, not satisfaction
Satisfaction surveys move with morale and leadership changes. Time in the record after hours moves with the thing you changed. Instrument it before the rollout, segment by role, and resist the urge to average across specialties with very different workloads.
Give the time back visibly
If an efficiency gain is immediately absorbed by added panel size, the next change request will be met with justified cynicism. Returning even part of the recovered time is what makes the second round of improvements possible.
