2. Types of study design

Question 1

What are the two types of study design?

Answer 1

Experimental and observational

Question 2

Give two examples of an experimental study

Answer 2

Clinical trial for a new antibiotic

Testing a new diet

Question 3

What occurs in a Randomised Control Trial?

Answer 3

Observational units are randomly assigned to one of two levels of an exposure
All other exposures are identical between the two groups

Question 4

Name 4 characteristics of an experimental study

Answer 4

1. They are very controlled
2. Maximise ability to observe one effect independent of other variables
3. Subjects are as similar as possible or stratified
4. Analysis is simple

Question 5

What are the two types of observational study and what do they do?

Answer 5

descriptive - describe what is occurring

analytical - statistical comparisons are made

Question 6

Name 4 characteristics of an observational study

Answer 6

1. No intervention during research
2. Can be difficult to tease out individual effects
3. Analysis can be very complicated
4. Few ethical constraints

Question 7

What is the sequence of events for determining causality?

Answer 7

1. Case studies
2. Descriptive studies
3. Analytical studies
4. Experimental studies

Question 8

What are the three main analytical studies?

Answer 8

cross sectional study
case control study
cohort study (survival study)

Question 9

What is a cross sectional study?

Answer 9

outcome variable and associated exposure variables measured at one point in time

Question 10

what are the 4 properties of a cross sectional study?

Answer 10

1. bad at identifying causes
2. good at generating hypotheses
3. quick
4. relatively cheap

Question 11

What is a case control study?

Answer 11

These take ‘cases’ and compare them with ‘controls’
Case = an individual with disease
Control = an individual from the same population without the disease that could have been included as a case if they had the disease

Question 12

What are the 4 properties of case control studies?

Answer 12

1. Better at identifying causes than cross-sectional studies
2. Identify associations more clearly
3. Are cheap and quick to do
4. Rely heavily on recall or past records, both of which may be inaccurate or biased

Question 13

What are cohort studies?

Answer 13

Take a cohort which contains individuals with different exposures and follow prospectively through time until the outcome happens in a defined proportion of the sample

Question 14

What are the 4 properties of cohort studies?

Answer 14

1. Good at identifying causes because the change occurs after the observations
2. Data on exposures are gathered as they happen (no recall bias)
3. Expensive and time consuming (“loss to follow-up”)
4. Difficult to analyse (repeated observations)

Question 15

What are the 7 tests for causality?

Answer 15

1. Strength of association e.g. high RR or OR
2. Dose-related i.e. the disease increases as the exposure increases
3. In the correct time sequence. i.e. the exposure occurs before the disease.
4. Independent of recognised confounders
5. Consistent finding in more than one study
6. Biologically plausible
7. Reversible i.e. Remove the risk factor and there is a change in the outcome.