Dr Priya Nair
MBBS, MRCEM, MSc EBHC
Emergency physician and evidence-based healthcare researcher who makes statistics genuinely approachable for clinicians.
Evidence-Based Medicine & Statistics9 years
Course overview
Critical appraisal is an essential part of the FRCEM examination and everyday emergency medicine practice. EMHub’s study materials explain each topic in detail, helping you understand research methods and statistics, interpret published evidence, and answer critical appraisal questions in both the FRCEM SBA and OSCE.
The course begins with evidence-based medicine, clinical questions and study designs. It then explores how research is conducted, including randomisation, sampling, outcome measures, missing data, research ethics and governance. Clear explanations cover probability, diagnostic test performance, measures of risk and treatment effect, confidence intervals, hypothesis testing and choosing statistical tests.
You will learn to recognise bias and confounding, assess validity and reliability, and interpret survival analysis. Dedicated sections guide you through systematic reviews and meta-analyses, including forest plots, funnel plots and heterogeneity. Clinical decision rules and screening are examined with a focus on their usefulness in practice.
The materials bring these concepts together in a step-by-step appraisal of a research paper. You will work through its question, methods, results, strengths and limitations before deciding whether the conclusions are supported and applicable to UK emergency medicine. The aim is to help you approach exam questions with confidence and make sound, evidence-based clinical decisions.
How it works
This course is not a question bank. Each topic explains a critical appraisal concept in detail, from evidence-based medicine and study design to statistics, bias and the step-by-step appraisal of a research paper.
30
Topic areas
What’s covered
30 topic areas, 351 topics
- Evidence-based medicine principles
- Hierarchy of evidence
- Formulating a clinical question
- PICO framework
- Literature searching
- Sources of evidence
- Primary and secondary research
- Applying evidence to Emergency Medicine
- Clinical versus statistical significance
Observational Studies
- Case report
- Case series
- Cross-sectional study
- Cohort study
- Prospective cohort study
- Retrospective cohort study
- Case-control study
- Ecological study
Experimental Studies
- Randomised controlled trial
- Controlled clinical trial
- Crossover trial
- Cluster randomised trial
- Factorial trial
- Non-inferiority trial
- Equivalence trial
- Superiority trial
Diagnostic Studies
- Diagnostic accuracy studies
- Reference/gold standard
- Index test
- Spectrum of disease
- Verification bias
Evidence Synthesis
- Systematic review
- Meta-analysis
- Scoping review
- Randomisation
- Random sequence generation
- Allocation concealment
- Blinding
- Placebo control
- Control groups
- Intention-to-treat analysis
- Per-protocol analysis
- Loss to follow-up
- Protocol deviations
- Crossover
- Contamination
- Stopping trials early
- Interim analysis
- Target population
- Study population
- Sample population
- Inclusion criteria
- Exclusion criteria
- Sampling methods
- Random sampling
- Convenience sampling
- Consecutive sampling
- Selection bias
- Representativeness
- External validity
- Generalisability
Types of Data
- Categorical data
- Numerical data
- Nominal data
- Ordinal data
- Discrete data
- Continuous data
- Dichotomous data
Variables
- Independent variable
- Dependent variable
- Exposure
- Outcome
- Confounding variable
Measures of Central Tendency
- Mean
- Median
- Mode
Measures of Dispersion
- Range
- Interquartile range
- Variance
- Standard deviation
- Standard error
Distribution
- Normal distribution
- Skewed distribution
- Outliers
- Percentiles
- Quartiles
- Basic probability
- Conditional probability
- Pre-test probability
- Post-test probability
- Odds
- Probability versus odds
- Bayes' theorem
- Bayesian reasoning
- 2 × 2 tables
- True positive
- False positive
- True negative
- False negative
- Sensitivity
- Specificity
- Positive predictive value
- Negative predictive value
- Prevalence
- Accuracy
- Positive likelihood ratio
- Negative likelihood ratio
- Pre-test probability
- Post-test probability
- Diagnostic odds ratio
- ROC curves
- Area under the ROC curve
- Diagnostic thresholds
- Risk
- Incidence
- Prevalence
- Relative risk
- Absolute risk
- Odds
- Odds ratio
- Hazard ratio
- Risk difference
- Attributable risk
- Experimental event rate
- Control event rate
- Absolute risk reduction
- Relative risk reduction
- Relative risk increase
- Absolute risk increase
- Number needed to treat
- Number needed to harm
- Relative risk
- Odds ratio
- Null hypothesis
- Alternative hypothesis
- Statistical hypothesis testing
- P-value
- Statistical significance
- Type I error
- Type II error
- Alpha
- Beta
- Statistical power
- One-tailed tests
- Two-tailed tests
- Confidence intervals
- 95% confidence interval
- Precision
- Confidence intervals around means
- Confidence intervals around proportions
- Confidence intervals around relative risk
- Confidence intervals around odds ratios
- Confidence intervals around risk differences
- Relationship between CI and statistical significance
Parametric Tests
- Student's t-test
- Paired t-test
- Unpaired t-test
- ANOVA
- Pearson correlation
Non-parametric Tests
- Mann-Whitney U test
- Wilcoxon signed-rank test
- Kruskal-Wallis test
- Spearman correlation
Categorical Data
- Chi-squared test
- Fisher's exact test
Choosing the Appropriate Statistical Test
- Continuous versus categorical data
- Paired versus unpaired data
- Normal versus non-normal distribution
- Two groups versus multiple groups
- Correlation
- Correlation coefficient
- Pearson correlation
- Spearman correlation
- Linear regression
- Multiple regression
- Logistic regression
- Confounder adjustment
- Regression coefficients
- Time-to-event data
- Censoring
- Kaplan-Meier curves
- Log-rank test
- Hazard ratio
- Cox proportional-hazards regression
- Median survival
- Sample-size calculation
- Statistical power
- Effect size
- Alpha
- Beta
- Event rate
- Clinically important difference
- Underpowered studies
- Overpowered studies
Selection Bias
- Sampling bias
- Admission bias
- Non-response bias
- Loss-to-follow-up/attrition bias
Information Bias
- Measurement bias
- Observer bias
- Interviewer bias
- Recall bias
- Reporting bias
Study-related Bias
- Performance bias
- Detection bias
- Attrition bias
- Publication bias
- Verification bias
- Spectrum bias
- Definition of confounding
- Recognising confounders
- Confounding versus bias
- Restriction
- Matching
- Stratification
- Randomisation
- Multivariable analysis
- Adjusted versus unadjusted results
Validity
- Internal validity
- External validity
- Construct validity
- Criterion validity
- Face validity
Reliability
- Repeatability
- Reproducibility
- Inter-observer reliability
- Intra-observer reliability
- Cohen's kappa
- Research question
- Search strategy
- Inclusion/exclusion criteria
- Study selection
- Data extraction
- Quality assessment
- Risk-of-bias assessment
- PRISMA
- Publication bias
- Evidence synthesis
- Pooled effect estimate
- Forest plots
- Fixed-effect model
- Random-effects model
- Weighting of studies
- Heterogeneity
- Cochran's Q
- I² statistic
- Publication bias
- Funnel plots
- Sensitivity analysis
- Subgroup analysis
- Line of no effect
- Point estimate
- Confidence interval
- Study weighting
- Pooled estimate
- Diamond
- Relative risk forest plots
- Odds ratio forest plots
- Mean-difference forest plots
- Interpretation
- Funnel-plot interpretation
- Publication bias
- Small-study effects
- Asymmetry
- Limitations of funnel plots
- Development of clinical decision rules
- Derivation studies
- Validation studies
- External validation
- Sensitivity and specificity
- Calibration
- Discrimination
- Clinical utility
- Impact analysis
- Screening principles
- Sensitivity
- Specificity
- Predictive values
- Lead-time bias
- Length-time bias
- Overdiagnosis
- Screening thresholds
- Primary outcomes
- Secondary outcomes
- Composite outcomes
- Surrogate outcomes
- Patient-reported outcomes
- Hard versus soft outcomes
- Objective versus subjective outcomes
- Clinically meaningful outcomes
- Loss to follow-up
- Missing data
- Missing completely at random
- Missing at random
- Missing not at random
- Complete-case analysis
- Imputation
- Multiple imputation
- Sensitivity analysis
- Informed consent
- Capacity to consent
- Emergency research
- Deferred consent
- Research Ethics Committee
- Confidentiality
- Data protection
- Vulnerable participants
- Clinical equipoise
- Declaration of Helsinki
- Good Clinical Practice
- Research protocols
- Trial registration
- Protocol deviations
- Adverse-event reporting
- Serious adverse events
- Data monitoring committees
- Research misconduct
- Data integrity
- Sponsorship
Introduction
- Research question
- Study rationale
- Existing evidence
- Hypothesis
Methods
- Study design
- Setting
- Population
- Inclusion/exclusion criteria
- Intervention
- Comparator
- Outcomes
- Randomisation
- Blinding
- Sample-size calculation
Results
- Participant flow
- Baseline characteristics
- Primary outcome
- Secondary outcomes
- Effect estimates
- Confidence intervals
- P-values
- Adverse events
Discussion
- Principal findings
- Strengths
- Limitations
- Bias
- Confounding
- Generalisability
- Comparison with existing evidence
Conclusion
- Does the evidence support the conclusion?
- Clinical significance
- Applicability to UK Emergency Medicine
- Would this change practice?
Subscription plans
One-off payment for a fixed duration. Nothing renews automatically. Prices in US dollars (USD), VAT included where applicable.
Every plan includes
- Full course access
- All topic explanations
- Progress tracking
- Mobile and tablet access
- Faculty Q&A
- Deferral protection
Faculty on this course
The consultants who wrote the content and answer your questions.
Create your free account
Register now and subscribe when you are ready. No card details required.