skip to main content

Critical appraisal of a pragmatic emergency-department randomised trial — MRCEM SBA MCQ

Instant feedback + full explanation. One question, done properly.

EasyProfessional practice and researchCritical appraisal of a pragmatic emergency-department randomised trialMRCEM SBA

An emergency department is appraising a randomised trial of an early-discharge pathway versus four hours of observation for adults with CT-confirmed uncomplicated ureteric stones whose pain is initially controlled. Some patients allocated to early discharge remain for observation when their pain recurs; some allocated to observation leave early. The primary outcome is unplanned hospital admission within seven days of randomisation, including admission during the initial attendance. Outcome data are available for every participant. The department wants to estimate the effect of offering the early-discharge pathway in routine practice. Which primary analysis best addresses this question?

Educational content. Not a substitute for clinical judgement or local policy.

Reveal the answer and explanation

Correct answer: B — Compare admission rates according to each participant’s randomised pathway.

The decision is whether to *offer* the early-discharge pathway, including what happens when a patient cannot complete it. The primary analysis should therefore compare all participants in their originally randomised groups. This is an intention-to-treat analysis: it retains patients whose recurrent pain leads to observation and those who leave observation early. Complete outcome ascertainment means that no participant needs to be excluded for a missing admission outcome. NICE defines intention-to-treat analysis by initial random allocation despite non-adherence or switching; CONSORT 2025 explains that retaining all randomised participants in their allocated groups preserves the benefits of randomisation. ([nice.org.uk](https://www.nice.org.uk/glossary?letter=i)) A is attractive because it describes the care received, but pathway changes may depend on recurrent pain and thus produce groups with different admission risks. C excludes non-adherent patients for the same post-randomisation reason. D excludes admissions during the initial attendance, although the trial explicitly counts them. E selects patients according to pain recurrence after randomisation, removing a clinically important reason the pathway may fail. These restricted comparisons could address separate descriptive or secondary questions, but not the stated primary question.

Reference: NICE Glossary: Intention-to-treat analysis (Undated; checked 27 September 2026) — https://www.nice.org.uk/glossary?letter=i CONSORT 2025 explanation and elaboration: updated guideline for reporting randomised trials (14 April 2025) — https://www.bmj.com/content/389/bmj-2024-081124