Evidence Based Medicine

Question 1

Criticisms of EBM?

Answer 1

Impossible for clinicians to appraise articles all the time.

Randomised controlled trial evidence comes from teaching hospitals-not representative.

An academic exercise for students, not relevant to clinical practice.

Been used to justify health cuts and restructuring by governments.

Question 2

What can EBM impact?

Answer 2

Clinical Findings- how to gather findings from history and examination.

Aetiology

Clinical manifestations- how often a disease causes it’s manifestations

Differential diagnosis- select serious causes of disease.

Prognosis, diagnosis, and therapy selection.

Question 3

What is an association and give 4 different explanations for it.

Answer 3

Chance: Chance assessed by statistical significance. Get around chance by increasing sample size.

Bias: inherent study design problem. Selection bias or measurement bias. Account for bias with randomization.

Confounding: Causal link with risk of disease. Common confounders are age, sex, geography. Get around confounding with matching, randomization, multivariate analysis, stratification.

Causation: Rule out the above. Look at all evidence.

Confounding:

Question 4

What is a confidence interval?

Answer 4

The range within which the true value is expected to lie with a given degree of certainty.

Question 5

6 types of descriptive bias?

Answer 5

Selection
Measurement
Observer bias- Researchers cognitive bias subconsciously influences participants.

Response bias
Voluntary response bias
Non response bias

Question 6

Bradford-hill Causation factors?

SSPECCTRA

Answer 6

TPCSDSECA
1. Temporal Relationship- Exposure must precede outcome.

2. Plausibility
3. Consistency with other investigations
4. Strength of the association
5. Dose response relationship
6. Specificity- Pleural mesothelioma
7. Experimental Evidence
8. Coherence
9. Analogy