Somewhere in the weeks before an ARCP, appraisal or portfolio review, most doctors end up doing the same thing: reconstructing months of work from rotas, theatre lists, clinic letters and memory. The scramble is close to universal, and it is not a discipline problem. Logging competes with clinical work and loses, week after week, until a deadline forces the issue.
So does it matter when you log? The evidence says yes, and the mechanism is blunter than most portfolio advice admits:
- Work you do not record close to the event has a measurable chance of never being recorded at all.
- What does get recorded late is thinner and less accurate than it feels from the inside.
A quarter of cases never made the log
The cleanest test of how much slips through comes from surgical case logs, because entries can be checked against theatre records. At a University of Maryland trauma centre, researchers compared every orthopaedic case over four months against the trainees' mandatory ACGME logbooks. Trainees failed to log 24% of their cases, 465 of 1,925. Among the cases they did log, 46% were missing procedure codes that the attending surgeons had recorded for the same operations (Okike et al., Journal of Surgical Education, 2018). These were doctors whose accreditation depended on the log.
A 2021 evaluation from Sierra Leone's surgical task-sharing training programme found the same pattern from the other direction. Cross-checking 278 electronic logbook entries from 14 users against hospital records, only 67.3% of logbook entries matched the records, and an estimated 50.7% of the procedures in the hospital records had never been entered in the logbook at all (Sung et al., BMC Medical Education, 2021).
Neither study looked at UK portfolios. One is US orthopaedics, the other a small early evaluation in a low-income surgical training setting, and applying them to ePortfolio evidence is an analogy, not a proof. As of mid-2026, nobody has run the equivalent study on a UK training portfolio. But the shared finding is hard to dismiss: doctors with strong incentives and a mandatory system still failed to capture a large share of their own work, and the missing share was invisible to them.
The analogy carries an uncomfortable implication. The days most worth evidencing, the arrest you led, the difficult family discussion, the complex list, are exactly the days with the least slack for logging. Left to end-of-rotation memory, your strongest cases are the ones most likely to vanish.
Why the scramble happens, precisely
When internal medicine trainees across north west England were surveyed about portfolio reflection in 2024, the two barriers that dominated were having no time to do it properly and seeing no benefit in doing it (Lodge et al., Cureus, 2024). Most trainees were writing reflections anyway. Compliance was high; perceived value was low. That combination describes a task done under duress at the last responsible moment.
The pre-deadline backfill has never been studied directly, but anyone who has sat an ARCP knows it. If career-invested doctors with high compliance still miss a quarter of their cases, the fix is not trying harder in May. The fix is moving capture to the point where the information still exists.
The standard your clinical records already meet
There is already a professional norm for recording work close to the event; it just is not usually applied to portfolios. Good medical practice 2024 requires formal records of your work, including patient records, to be clear, accurate, contemporaneous and legible, and the GMC defines contemporaneous as made at the same time as the events you are recording, or as soon as possible afterwards (GMC, 2024). Defence organisations give matching advice for clinical records; the MDU's record-keeping guidance sets no fixed time window, only the expectation that you record while the detail is reliable.
Your portfolio is an educational record, not part of the medical record, a distinction UK reflective practice guidance draws explicitly, and no regulator requires portfolio entries to be contemporaneous. Carrying the norm across is a borrowed discipline, not a rule. It is worth borrowing for the same reason it exists in the clinical setting: a record made close to the event is more complete, more accurate and more defensible than one reconstructed later. A portfolio entry is, at its core, a record you will one day ask a panel to trust.
How close to the event is close enough
Neither the GMC definition nor the recall evidence yields a precise cut-off, so treat this as a working rule rather than a regulation.
A same-day rule that survives a real rota
- Capture a skeleton the same shift: what happened, your role, one line on why it mattered. Seconds, not minutes, and no patient identifiers.
- Expand it into a full entry within the week, while the clinical detail is still yours rather than the notes'.
- If a case deserves reflection, write the honest first version early. Polishing later changes wording; delay changes substance.
- Batch the admin, not the capture. Linking evidence to curriculum outcomes can wait for a quiet hour; the facts of the case cannot.
- Anything still unlogged after a rotation change, log now anyway. A dated late entry beats a missing one.
The barrier is friction, not resolve
The gap between intending to log and logging is mostly friction: finding a terminal, remembering a password, facing a long form at the end of a long shift. Removing that friction is the reason we built Alessia. Capturing a case at the point of care takes seconds on your phone, offline if you have no signal, and AI voice-to-notes turns a spoken summary into a transcript and draft notes you review and edit, so the blank page never happens at the bedside end of the process. The habit the evidence supports only works if the tool asks almost nothing of you at the time.
Before the next deadline cycle
Same-day capture is the practice; what you capture and where it goes are their own questions. What belongs in your portfolio, by route and by grade, is covered in our guide to what goes in your portfolio. If you rotate between systems, our guide to moving portfolio evidence between systems covers keeping those entries portable. Both sit in our Portfolio Fundamentals hub, which collects the practices this article takes as its starting point.
Three questions worth answering honestly:
- How many of last month's most significant cases exist anywhere in your portfolio right now?
- What is the longest gap between an event and the entry that records it, and would you trust the detail in that entry?
- If capture took thirty seconds, what would be left of the scramble?
Sources
- GMC, Good medical practice (2024), Domain 3: Colleagues, culture and safety
- GMC, Good medical practice (2024), overview and effective date
- Okike K, Berger PZ, Schoonover C, O'Toole RV. Do orthopaedic resident and fellow case logs accurately reflect surgical case volume? Journal of Surgical Education, 2018;75(4):1052-1057
- Sung S, Horthe H, Svendsen OV, van Duinen AJ, Salvesen O, Vandi A, Bolkan HA. Early evaluation of the transition from an analog to an electronic surgical logbook system in Sierra Leone. BMC Medical Education, 2021
- Lodge J, Neptune C, Tun NN. Factors affecting engagement with portfolio-based reflection among internal medicine trainees: a survey-based study. Cureus, 2024
- MDU, Effective record-keeping
- GMC, Reflective practice guidance hub (The reflective practitioner)
