New: our faculty are still reviewing these cards. Spotted an error? Use “Report an issue” on the card in the app, or contact us.

Bias, trials & screening flashcards

15 high-yield MRCP Part 1 cards from the Intellect Medicos Basic sciences & statistics course. Try to answer each question before you open it — then study them properly with spaced repetition, which brings each card back just before you would forget it.

Study these 15 cards Save progress — free account

1. Screen-detected cancer patients seem to survive longer only because the diagnosis is made earlier, not because death is delayed. Which bias is this?

   Show answer

   Lead-time bias.

   Why? Survival is measured from an earlier starting point, so it looks longer even if the date of death is unchanged.

2. A screening programme mainly picks up slow-growing, less aggressive tumours, making survival look better. Which bias is this?

   Show answer

   Length-time bias.

   Why? Slow-growing disease spends longer in the detectable pre-symptomatic phase, so it is over-represented among screen-detected cases.

3. Which bias is especially common in case-control studies because cases remember past exposures differently from controls?

   Show answer

   Recall bias.

   Why? People with a disease search their memory harder for possible causes, inflating reported exposure.

4. Which graph is used to detect publication bias in a meta-analysis, and what appearance suggests bias?

   Show answer

   Funnel plot: asymmetry suggests publication bias.

   Why? Effect size is plotted against study size; small negative studies are missing from one side if they went unpublished.

5. What is intention-to-treat analysis, and why is it used?

   Show answer

   Every participant is analysed in the group they were randomised to, whatever happened afterwards. It preserves randomisation and avoids overestimating benefit.

   Why? Excluding drop-outs or switchers (per-protocol analysis) reintroduces confounding, because those who stop treatment often differ systematically.

6. Phase ______ trials test safety and dose in small numbers of healthy volunteers; phase 4 is post-marketing surveillance for rare and long-term adverse effects.

   Show answer

   1

   Why? Phase 2 is the first use in patients with the disease (efficacy and short-term side effects); phase 3 is the large RCT against standard care that supports licensing.

7. Phase 1 trials test safety and dose in small numbers of healthy volunteers; phase ______ is post-marketing surveillance for rare and long-term adverse effects.

   Show answer

   4

   Why? Phase 2 is the first use in patients with the disease (efficacy and short-term side effects); phase 3 is the large RCT against standard care that supports licensing.

8. Name three of the Wilson-Jungner criteria for a screening programme.

   Show answer

   Any three: important health problem; recognisable latent or early stage; suitable, acceptable test; accepted treatment; natural history understood; facilities available; agreed policy on whom to treat; cost-effective; continuing process.

   Why? These WHO (1968) principles decide whether screening does more good than harm.

9. A case-control study recruits both cases and controls from hospital inpatients, producing a spurious association. Which bias is this?

   Show answer

   Berkson's (admission rate) bias — a form of selection bias.

   Why? Hospital patients differ from the general population (e.g. more comorbidities), so the groups are not representative.

10. Trial participants improve their behaviour simply because they know they are being observed. What is this called?

    Show answer

    The Hawthorne effect.

    Why? It can exaggerate improvements in both arms of a study and in audits.

11. In a diagnostic accuracy study, only patients with a positive index test go on to have the gold-standard test. Which bias results?

    Show answer

    Work-up (verification) bias, which distorts sensitivity and specificity.

    Why? Test-negative patients are rarely verified, so false negatives are missed (sensitivity overestimated) and true negatives are under-counted (specificity underestimated).

12. Which design feature of a trial controls for both known and unknown confounders?

    Show answer

    Randomisation (with adequate allocation concealment).

    Why? A confounder is linked to both exposure and outcome (e.g. smoking in a coffee–heart disease study). Matching, restriction, stratification and multivariable analysis only deal with confounders you have measured.

13. Which observational study design is most efficient for studying the causes of a rare disease?

    Show answer

    A case-control study.

    Why? It starts with people who already have the disease and compares their past exposures with controls. A cohort study suits rare exposures and measures incidence; a cross-sectional study measures prevalence.

14. Which type of evidence sits at the top of the hierarchy of evidence (level 1a)?

    Show answer

    A systematic review or meta-analysis of randomised controlled trials.

    Why? Pooling several RCTs reduces random error and bias compared with any single study.

15. A trial aims to show that a new oral anticoagulant is not unacceptably worse than warfarin, within a pre-set margin. What type of trial is this?

    Show answer

    A non-inferiority trial.

    Why? Superiority trials test whether the new drug is better; equivalence trials test whether the difference lies within a margin in both directions (−delta to +delta); non-inferiority trials test one direction only.