Ethics Ambiguity: When Two Answers Look Right, and Why Your Intuition Is the Problem

If you have spent any time in the clinical years, you have felt the specific, blood-boiling frustration of an ethics question where two options seem perfectly defensible. You sit there, comparing them, trying to find the logical "gotcha," only to find the explanation tells you one is "more correct" than the other based on a subtle nuance of the GMC (or board) guidance. It feels like a trap. It feels like bad medicine.

Let’s cut the fluff: ethics is not about finding the "right" answer. It is about identifying the specific professional judgement the examiner is testing. If you are struggling with this, it is likely because you are treating ethics like a science problem rather than a procedural one. Here is how I handle the ambiguity, based on my time surviving clinical rotations and the hundreds of questions I have logged in my "questions that fooled me" notebook.

The Question Bank Baseline: Why UWorld and Amboss Are Just the Beginning

Most of us spend $200-400 for access to curated, physician-written practice question banks like UWorld or Amboss. These are the gold standard for a reason. They provide a baseline of clinical knowledge that is peer-reviewed and vetted. They force you into retrieval practice, which we know—through cognitive science—is infinitely more effective than re-reading your notes for the tenth time.

However, these banks are inherently generic. They have to be. They are written to accommodate the widest possible audience. When you hit an ethics question in these banks, you are testing your ability to interpret the examiner's logic, not the reality of the hospital ward. The problem arises when you rely only on these banks to develop your professional judgement.

The Quality Variance Table

Source Type Quality Indicator Use Case Curated Banks (UWorld/Amboss) High (Expert reviewed) Standardising your knowledge base LLM-based quiz generators (e.g. Quizgecko) Variable (Requires curation) Testing specific, niche guideline recall Peer-created flashcards (Anki) Personalised (Context-heavy) Long-term retention of "fooled me" points

Moving Beyond Generic Banks: The AI Quiz Pipeline

Because board exams reward the ability to navigate specific guidelines, using a LLM-based quiz generation pipeline has become a core part of my workflow. If I am struggling with capacity assessments or end-of-life care, I do not just grind more generic questions. I take the specific guideline summary from my hospital or the national board, paste it into an AI tool, and ask it to generate questions based strictly on that text.

Tools like Quizgecko are useful here, but you must be wary. AI-generated questions can be "fluffy" or factually hallucinatory if you aren't careful. The trick is to treat the AI as a drafter, not an author. Uploading your notes or pasting specific guideline summaries creates a closed loop where the AI can't invent "expert" logic—it has to use the source material you provided.

How to Spot Low-Value Questions

If you are using an LLM or a less reputable question bank, watch for these red flags:

The "All of the Above" Trap: If the distractors are purely structural and not based on clinical reasoning, discard the question. Ambiguity without a Reference: A high-quality ethics question points to a specific principle (e.g., Beneficence vs. Autonomy). If the "correct" answer changes based on a subjective opinion rather than a documented framework, it is low-value. Lack of Nuance: If a question ignores the context of a patient's history or local legal constraints, it is not helping you build clinical judgement.

The "Two Defensible Answers" Strategy

When you encounter a question where both A and B look correct, stop looking for the "better" answer. Start looking for the "primary" concern. In ethics, this usually follows a hierarchy:

Legal/Statutory Requirement: Does the law demand this? (Usually beats clinical preference). Patient Safety: Is there an immediate risk to life? Autonomy/Consent: Is the patient capable? Professional Guidance: What do the GMC/Board guidelines explicitly state as the "first step"?

If you are stuck between two options, write down why each one is tempting. Usually, you will find that Hop over to this website one answer addresses a secondary outcome, while the other addresses the immediate, legally protected process. Professional judgement is knowing which step to take first in the cascade.

Integrating with Anki for Long-Term Synthesis

Retrieval practice is the only way to make this stick, but it is not enough to just see the answer. When I get a tricky ethics question wrong, I don't just add the answer to Anki. I add the logic of the distractors.

My Anki cards for ethics usually look like this:

Front: "Patient requests assisted dying. Why is option X (providing palliative sedation) preferred over option Y (ignoring the request) in this specific guideline?" Back: "Option X addresses the underlying distress (clinical duty) while remaining within legal boundaries, whereas Option Y fails to address the patient's immediate suffering, which is a breach of the duty of care."

By focusing on the "why," you stop guessing and start applying frameworks.

Final Thoughts: Trust Your Process, Distrust the Hype

Don't be fooled by the marketing claims that you can "boost your score fast" with some magical new study tool. Clinical reasoning is a slow-burn process. It is the result of thousands of questions answered, hundreds of mistakes logged, and a disciplined approach to checking your answers against actual evidence-based guidelines.

Keep your "questions that fooled me" list. Keep timing your sessions so you know if you are rushing through ethics (a common cause of misreading the question). And above all, recognise that when you find two right answers, you aren't fighting a bad question—you're fighting your own tendency to prioritise one ethical principle over another without checking the board's rubric first.

Now, set your timer for 45 minutes, dive back into your bank, and stop looking for the "magic" shortcut. It’s all in the practice.

Edit

Pub: 10 Apr 2026 21:56 UTC

Views: 6