skip to main content

Confidence interval interpretation — MRCPsych Paper A MCQ

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

HardEpidemiology and research methodsConfidence interval interpretationMRCPsych Paper A

In a placebo-controlled superiority trial of a new antipsychotic, positive values favour the new treatment. The adjusted mean between-group difference in symptom reduction is 3 points, with a 95% confidence interval from −2 to 8 points. A difference of 5 points was prespecified as the minimum clinically important difference. Which is the best interpretation?

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

Reveal the answer and explanation

Correct answer: A — The estimate is imprecise and compatible with both no effect and clinically important benefit

Explanation lettering: C = shown as A · E = shown as B · D = shown as C · B = shown as D · A = shown as E

C is correct. The confidence interval includes zero, so the result is not statistically significant at the conventional two-sided 5% level. However, its upper limit exceeds the prespecified 5-point threshold for clinically important benefit, while its lower limit permits slight worsening. The result is therefore imprecise and inconclusive, rather than proof that the treatment is ineffective. A is incorrect because this interval concerns the population mean treatment effect, not individual patient responses. B is incorrect because equivalence requires prespecified equivalence margins and a confidence interval contained within them. D mistakes interval symmetry for evidence of unbiased estimation. E is incorrect because a 95% confidence interval for a mean difference that crosses zero corresponds to p greater than 0.05.

Reference: Gewandter JS et al. Interpretation of CIs in clinical trials with non-significant results: systematic review and recommendations. BMJ Open. 2017;7:e017288. https://bmjopen.bmj.com/content/7/7/e017288