Most ‘error logs’ fail because they become a second notebook. The goal is not record-keeping. The goal is behavioural change: you stop making the same mistakes. The solution is a small, aggressively pruned list that you retest weekly until errors vanish.
The rule: small enough to finish weekly
If you can’t retest your error log in one sitting every week, it is not an error log — it is a guilt document. Keep it small (30 items is ideal) and delete relentlessly once automatic.
The Bottom Line
- Max 30 active items. Anything bigger is too broad or too undisciplined.
- Each item is a 10-second prompt with a clear cue and action.
- Weekly retest + delete what’s now automatic.
1
Step 1 — Capture only high-value errors
Include: thresholds, contraindications, escalation triggers, common discriminators, and repeated misreads. Do not include low-yield trivia.
2
Step 2 — Standardise the format
Format: Cue → Decision → Action. Example: “AF + CHA2DS2-VASc X → anticoagulate → choose agent with contraindications in mind.”
3
Step 3 — Add a ‘why you missed it’ tag
Knowledge / misread / threshold / timing / bias. This helps you design the *next* week’s training rather than repeating the same approach.
4
Step 4 — Retest weekly (single sitting)
Set a 20–30 minute timer. Retest all items closed-book. Any item answered instantly gets deleted. Ambers stay. Reds get rewritten narrower.
5
Step 5 — Monthly reset
Once per month, wipe the list down to only stubborn Reds. This keeps the system lean and prevents ‘collection mode’.
Practice
Test your knowledge
Apply this concept immediately with a high-yield question block from the iatroX Q-Bank.
SourceRoediger & Karpicke (2006): Testing effect (PubMed)
Open Link SourceCepeda et al. (2006): Spacing effect meta-analysis (PubMed)
Open Link