Study Designs

Question 1

What is causality? What are 2 approaches to it?

Answer 1

Differentiating association from causation
Deterministic approach (systematic observations certainly predict future events)
Stochastic approach (systematic observations give the risk of future events)

Question 2

What’s a confounding variable?

Answer 2

Something that’s associated with both exposure and outcome

Question 3

What does Brandon Hill’s criteria for inferring causality 1965 set out to do?

Answer 3

9 criteria that seek to provide epidemiological evidence of a causal relationship between a presumed cause and an observed effect

Question 4

What are the 9 Brandon Hill criteria?

Answer 4

Strength of association
Specificity of association
Consistency of association
Temporal sequence (temporality)
Dose response 
Reversibility
Coherence of theory
Biological plausibility
Analogy

Question 5

What are observational studies and why are they used?

Answer 5

Observe distributions and determinants of health

Can inform health policy, planning, provision and future research

Question 6

What are 4 main types of observational study? What category do they come under (descriptive vs analytical) and which are most reliable?

Answer 6

Cohort studies > case control studies > cross-sectional studies > ecological studies

Ecological and cross-sectional studies are descriptive
Cohort and case control studies are analytical
Cross-sectional can be both

Question 7

What’s the long term observational study? The other 3 are short term

Answer 7

Cohort studies - analytical study where a group is followed over time

Question 8

Which type of observational study is good for rare diseases?

Answer 8

Case-control studies - analytical study comparing a group with a condition (case) to a group without a condition (control)

Question 9

What are ecological and cross-sectional studies good for?

Answer 9

Hypothesis generation

Question 10

What are randomised control trials?

Answer 10

Controlled clinical trials where the therapies are allocated by a chance mechanism

Question 11

What are the advantages of RCTs?

Answer 11

Eliminate selection bias
Balance prognostic factors
Validity of statistical tests

Question 12

Outline random allocation

Answer 12

Gives equal chance of receiving each treatment - reduces selection bias if patients enter groups before randomisation

Question 13

Outline blind trials

Answer 13

Single or double (both clinical and patient unaware of treatment given)
Treatment must appear identical in all ways
Compare active drug with inert placebo
Use designated pharmacy to label identical containers and code numbers

Question 14

Outline the conduct of an RCT

Answer 14

Eligible patients present -> consent -> randomised to groups -> assessment

Question 15

What should informed consent at clinical trials include information on?

Answer 15

That the patient is invited to be in the trial
What all the alternative treatments are, including side effects
Explain that treatments are allocated at random
Explain that patients may withdraw at any time
Info should be given verbally and in writing with a ‘cooling off’ period

Question 16

What’s the concept of clinical equipoise?

Answer 16

Reasonable uncertainty about which treatment is genuinely better, including non-treatment

Question 17

What’re systematic reviews?

Answer 17

A review of a clearly formulated question that uses specific and explicit methods to identify, select and critically appraise relevant research and to collect and analyse data from studies included in a review

Question 18

When are systematic reviews used?

Answer 18

To identify best forms of healthcare - evidence-based

Question 19

What are different types of bias?

Answer 19

Selection bias
Performance bias
Detection bias
Attrition bias
Reporting bias

Question 20

What does PICOS stand for?

Answer 20

Translates research question into a review
Population
Intervention
Comparator
Outcomes
Study of Design