Studying While On Call: A Realistic System for Ophthalmology Residents

An on-call roster does not need a fictional two-hour evening study plan. It needs a menu of tasks that can survive interruption. The core rule is simple: clinical duties, handover, supervision arrangements and local fatigue policy come first. Revision is for genuine quiet gaps and protected time, never a reason to be unavailable or to work through a task that needs your attention.

This system uses retrieval practice and distributed practice as learning principles, not promises. The evidence does not prescribe one schedule for every resident, prove an exam outcome, or determine when any individual is sufficiently rested. Use your programme’s actual rules and adjust the plan to your real roster.

Build a task menu before the shift

The worst moment to choose study material is an unexpected ten-minute gap. You will search, scroll and select the most comfortable task rather than the one that addresses your current weakness. Build three lists in advance and keep them visible offline or in one simple note.

The time categories are not a productivity target. They are an interface: when a gap appears, match it to a pre-decided task. If the gap disappears, stop without calling the day a failure.

Choose retrieval over passive drift in short gaps

Retrieval practice means attempting to bring information to mind before checking it. The classic education literature supports practice testing and distributed practice as generally useful strategies, while also warning against treating a learning technique as a guaranteed result for every context.¹² On call, that makes a practical difference: a five-minute task can be a specific question, a one-line differential or an image label, not a vague promise to “read retina.”

Make the prompt small enough to complete honestly:

• “Name the three signals that make this DR stem PDR rather than severe NPDR.”
• “From memory, draw the aqueous outflow sequence, then check it.”
• “What is the single distinction between subretinal and sub-RPE fluid?”
• “Why was question 18 wrong: fact gap, wording error, image-recognition error or rushed reading?”

The point is not to maximise a question count. It is to produce a visible answer that can be checked. If you guessed correctly, mark it as a guess; guessed-right material is often a better review item than a clearly known fact.

Use an interruption note, not a fragile concentration ritual

Call work is interruptible. Design for it. At the bottom of every study block, write a one-line restart note before you put the material away:

“Stopped after OCT question 4. Next: explain why this space is subretinal, then review answer key.”

When you return, do the next action rather than rereading the entire page to reconstruct where you were. This protects the small gap from being consumed by restart friction. It also shows whether a task was too large: if every interruption leaves a confusing note, shrink the task until the next action is obvious.

Keep the error log equally light. Use four fields: topic/question, your chosen answer, why it was wrong, and next action. “Review glaucoma” is not an action. “Compare angle-closure mechanism with plateau-iris mechanism using one diagram” is. The log is a queue, not a diary of every question you have done.

Separate an on-call gap from a protected session

On-call gaps are good for recall, answer review and one contained comparison. Protected time is for material that needs continuity: a new difficult chapter, a timed block, a case rehearsal, calculation practice or a full topic map. Mixing the two creates a common trap: attempting heavy work during duty, being interrupted, then interpreting the incomplete session as poor discipline.

Use this decision rule:

The last row matters. Sleep deprivation can affect cognition and performance, but this article cannot set a safe threshold or advise a resident to override their programme’s duty arrangements. If fatigue, workload or supervision is creating a safety concern, use the formal local route rather than trying to solve it with a better flashcard schedule.³

The post-call reset: reschedule rather than compensate

After a difficult call, do not create a debt narrative. Mark the unfinished item as “moved,” not “failed.” Return to it in the next protected session. If you choose to study later and the task is compatible with your actual capacity and local rules, make it a low-load retrieval block rather than a high-stakes mock or new dense topic.

This is a planning inference, not a medical prescription: a roster system that assumes every lost session must be made up immediately tends to become unusable. The repair is to keep a short “next protected session” queue. Limit it to three items. When a fourth item arrives, choose what to defer instead of letting the backlog become an unscannable list.

A weekly pattern that tolerates real duties

Replace “every day must look the same” with minimum and ideal versions.

The “minimum” version keeps the system alive during crowded weeks. The “ideal” version gives a direction when time opens. Neither is a promise of consistency, progress rate or examination result.

Make the material phone-friendly but source-conscious

Use short material for short gaps, but keep a source trail for anything you do not understand. The free ophthalmology MCQs are one verified route for a brief closed-book set; the general MCQ quiz offers a mixed second pass. The ophthalmology glossary is useful when one definition is blocking a question. For longer planning, use the resources hub and the Handwritten Exam Ready Notes page only to inspect current resource information.

Do not turn the number of installed apps, open PDFs or saved tabs into a measure of preparation. Keep one question source, one error queue and one reliable place to verify a term. On a busy roster, fewer moving parts make it easier to resume a task without choosing a new system every week.

Match error type to the next task

An error log is useful only if it changes the next action. Do not schedule the same generic “revision” response for every wrong answer. Label the underlying error, then choose the smallest task that addresses it.

For example, “I got glaucoma wrong” is not a useful tag. “I confuse disc haemorrhage with a field-defect progression clue” produces a real next task. The distinction matters on call because you want a task you can start in one minute, not a large topic name that creates another search session.

Use a weekly source list, not an endless backlog

At the beginning of the week, choose a small source list: one question set, one reference/notes source, one error log and one protected-session task. This is not a claim that four tools are optimal; it is a guardrail against opening a new resource whenever a difficult item appears. Add a new source only if it replaces an existing one or answers a clearly documented gap.

Keep the queue deliberately incomplete. A realistic queue might contain one topic weakness, one image-pattern weakness and one exam-mechanics task. If several shifts are busy, carry the queue forward unchanged rather than doubling it. Distributed-practice research supports returning to material across time, but it does not require “catching up” every skipped interval or using a single universal spacing schedule.¹²

The practical test is whether you can tell, at the start of a short break, exactly what the next action is. If your list contains ten chapters, five apps and three mock papers, it is a wish list rather than an on-call system.

A two-week audit

At the end of two weeks, do not ask whether you followed the ideal calendar. Ask four concrete questions:

1. Which task sizes actually occurred on my roster?
2. Which interruptions repeatedly destroyed a task, and how can I make that task smaller?
3. Which error type is recurring: fact, visual pattern, calculation, vocabulary or rushed reading?
4. Which planned tasks should move to protected time or be removed?

Then change one variable only — for example, replace ten unplanned MCQs with one five-question set plus an interruption note. The aim is a system you can restart, not an increasingly elaborate schedule you abandon after one difficult week.

Sources and scope

The learning literature below supports retrieval/distributed-practice concepts, not a resident-specific study prescription or clinical-safety claim. Duty, handover, fatigue, supervision and rest requirements remain governed by the local institution and appropriate professional/occupational channels.

Footnotes

1. Cepeda NJ, Pashler H, Vul E, Wixted JT, Rohrer D. Distributed practice in verbal recall tasks: a review and quantitative synthesis. Psychological Bulletin. 2006. Live-checked 18 August 2026. ↩ ↩²

2. Dunlosky J, Rawson KA, Marsh EJ, Nathan MJ, Willingham DT. Improving students’ learning with effective learning techniques. Psychological Science in the Public Interest. 2013. Recheck endpoint before publication. ↩ ↩²

3. Alhola P, Polo-Kantola P. Sleep deprivation: impact on cognitive performance. Neuropsychiatric Disease and Treatment. 2007. Live-checked 18 August 2026. ↩