A practical introduction to medical statistics

M R C O G PA R T 2 · S O S P O C K E T C A R D · K N O W L E D G E A R E A 2 – T E A C H I N G A N D R E S E A R C H

A Practical Introduction to Medical Statistics Scally AJ. A practical introduction to medical statistics. TOG 2014;16:121–8. doi:10.1111/tog.12081

P<0.05 ±1.96×SE 80% CONVENTIONAL SIGNIFICANCE THRESHOLD 95% CI = EFFECT SIZE ± THIS TYPICAL TARGET POWER (1−Β)

DO N' T-M IS S FA C TS

  1. A statistically flawed study is an ethically flawed study

  2. 95% CI = effect size ± 1.96 × SE; SE = SD/√n (uncertainty in estimate, not spread of data)

  3. P<0.05 conventional threshold – P-value is the probability of the observed (or more extreme) result assuming H0 true, NOT the probability H0 is true

  4. Type I error = α (reject true H0); Type II error = β (accept false H0); Power = 1−β, usually target 80%

  5. Statistical significance ≠ clinical importance – always ask about effect size, not just P-value

  6. OR ≈ RR only when event prevalence <10%; OR overestimates RR at higher prevalence

  7. Fisher's exact test for small samples/single-digit cell counts; chi-squared otherwise

  8. Worked example: smoking & low birthweight – RR 1.61 (95% CI 1.06–2.44), OR 2.02 (95% CI 1.08–3.77)

PARAMETRIC VS NON-PARAMETRIC (2 INDEPENDENT GROUPS)

Parametric: t test (ratio/interval, normal) → effect size + CI + P

Non-parametric: Mann–Whitney U/Wilcoxon rank-sum (ordinal/non-normal) → P-value only

ABBREVIATION KEY

SD — standard deviation ANCOVA — analysis of covariance

SE/SEM — standard error (of the mean) MWU/WRS — Mann–Whitney U / Wilcoxon rank-sum test

CI — confidence interval IQR — interquartile range

RCT — randomised controlled trial α — probability of a type I error

OR — odds ratio β — probability of a type II error

RR — risk ratio (relative risk)

Source: The Obstetrician & Gynaecologist (TOG), RCOG journal — Scally AJ. A practical introduction to medical statistics. TOG 2014;16:121–8. doi:10.1111/tog.12081

Clinically reviewed by NHS obstetricians and gynaecologists

MRCOG AI is an independent educational tool. It is not affiliated with, endorsed by, or connected to the Royal College of Obstetricians and Gynaecologists (RCOG).

Included with a Pass Package

Start free