Printable list of all statistics SAQs
Critical appraisal of randomised controlled trials
Definitions of terms in statistics: prevalence, relative risk and attributable risk
Calculate sensitivity, specificity, PPV, NPV and accuracy
Allocation concealment , block randomisation, stratification, minimisation algorithm.
Anatomy of the forest plot; advantages and disadvantages of meta-analysis
Meta-analysis: advantages and disadvantages. Also, the funnel plot
Define: external validity, bias and selection bias.
Meta-analysis: advantages and disadvantages. Also, publication bias.
Analysis of clinical trials: overall assessment of quality
Define "EBM"; also, levels of evidence and intention to treat analysis
Definitions: Risk ratio, NNT, P-value and confidence interval.
Blinding and allocation concealment
Definitions Relative risk, absolute risk, NNT, and the power of a study.
Calculate sensitivity, specificity, PPV and NPV
Calculate: Relative risk, absolute risk, NNT
The receiver operating characteristic plot, i.e. ROC curve.
Meta-analysis: the forest plot. Its anatomy in some detail.
Definitions: Type 1 error, type 2 error, study power, sample size and effect size.
Definitions: Phase III trial, absolute and relative risk reduction, NNT.
Calculate sensitivity, specificity, PPV and NPV.
Meta-analysis: features suggestive of "high quality".
Definitions: sensitivity, specificity, PPV NPV and the positive likelihood ratio.
Define intention to treat analysis and randomisation.
Definitions: relative risk, absolute risk, NNT and study power.
The Kehoe principle and the Precautionary principle
Definitions: sensitivity, specificity, PPV and NPV.
Parametric and non-parametric analytical tests
Analysis of clinical trials: Methodological quality
Tests of statistical significance: Chi-squared test, Fisher’s Exact Test and logistic regression.
Explore the "relevance" of evidence based medicine in ICU
Analysis of clinical trials: criteria used to assess validity.
[Submit a comment or correction]
This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.