Study Designs 1: Observational Studies

Question 1

What are observational studies?

Answer 1

Studies that observe distributions and determinants of health: they do NOT involve interventions

Question 2

What are some examples of questions observational studies can ask?

Answer 2

How much disease is there?
What causes a disease?
What puts people at greater or less risk for a disease?

Question 3

What can observational studies inform?

Answer 3

Health policy Planning + provision

Future research

Question 4

What are the 2 types of observational studies?

Answer 4

1. Descriptive studies that examine DISTRIBUTIONS e.g. how much measles is there in different UK regions?
2. Analytical studies that examine DETERMINANTS e.g. is vaccination related to measles incidence?

Question 5

What designs would be relevant for descriptive studies?

Answer 5

Ecological

Cross-sectional

Question 6

What designs would be relevant for analytical studies?

Answer 6

Cross-sectional
Case-control
Cohort

Question 7

What are the general outcomes of observational studies?

Answer 7

1. Measure of effect size: How big is effect? How strong is association? - assess CLINICAL significance
2. 95% CI: how precise is our measure of effect? - depends on error factor
3. P-value: how likely is it that size of effect is due to chance? - assess STATISTICAL significance

Question 8

What is the rule of thumb of P-values?

Answer 8

If it is less than 0.05, we reject our hypotheses, however if it is more than 0.05 we dont reject the hypothesis but not necessarily accept it (might need more data)

Null hypothesis P-value = 0 (no significance between variables)

Question 9

In order of increasing reliability and validity, list the types of study.

Answer 9

Expert opinion
Case reports
Observational: 
- Ecological
- Cross-sectional
- Case-control
- Cohort
RCT
SR/MA

Question 10

What is an ecological study?

Answer 10

Studies where the unit of analysis is NOT the individual but a group e.g. school, area, region or country generally using available administrative data or population-level data

Question 11

What are the advantages and disadvantages of ecological studies?

Answer 11

Adv:

• Good for hypothesis generation
• Cheap and simple to do
• Low or no ethics involved

Disadv:

• Can only establish association, not causation
• Data may be unreliable

Question 12

What is the common measure of effect in ecological studies?

Answer 12

Correlation coefficient (R) using methods such as Pearson (ND), Spearman (skewed) + Kendall’s tau (ordered) - possible range -1 to +1 and there is a 95% CI for R value

Null hypothesis value = 0

Question 13

What can R^2 be used for?

Answer 13

To evaluate the proportion of variation in the outcome (y-axis) explained by variance in the predictor (x-axis)

Question 14

What are the potential biases of ecological studies?

Answer 14

Ecological fallacy
Confounding variables
Quality of data (timing)
Selective reporting

Question 15

What is ecological fallacy?

Answer 15

Assuming that the population values are the same as the individual values

Question 16

What is a cross-sectional study?

Answer 16

Studies where data are collected from a sample at a point in time

Can be repeated (usually with a different sample)

DESCRIPTIVE or ANALYICAL

Question 17

What are the advantages and disadvantages of cross-sectional studies?

Answer 17

Adv: good for hypothesis generation

Disadv: can only establish association but not causation

Question 18

What is the common measure of effect of cross-sectional studies?

Answer 18

Standardised difference (absolute difference if data is not ND’d) in means - 95% CI for difference in means + null hypothesis value = 0

Cohen’s D = (Mean group 1 - Mean group 2)/pooled SD

Small ES = 0.2
Med ES = 0.5
Large ES = 0.8

Question 19

What are the methods for statistical testing in cross-sectional studies?

Answer 19

2 independent groups:

1. Unpaired T-test (ND)
2. Mann Whitney U test (skewed)

Question 20

What are the potential biases of cross-sectional studies?

Answer 20

Sample selection bias
Response bias
Recall bias
Responder/social desirability bias
Confounding variables
Direction of causation

Question 21

What are case-control studies?

Answer 21

Analytic studies comparing exposure for a group with a condition (cases) and a group w/o the condition (controls) so look backwards - can have matched (1+ per case) or unmatched controls

Question 22

What are the advantages and disadvantages of case-control studies?

Answer 22

Adv:

• Useful for rare diseases or diseases with long latent (undetectable/undetected) periods
• Fast + cheap
• Loss to follow-up not an issue
• Good to examine multiple exposures/risk factors

Disadv:
- Cannot measure incidence or prevalence

Question 23

What is the common measure of effect of case-control studies?

Answer 23

OR = odds of being a case if exposed/odds of being a case if not exposed (may use adjusted OR to control for potential confounders)

95% CI + null hypothesis value = 1

Question 24

How do you calculate an Odds Ratio (OR)?

Answer 24

Odds (case if exposed) = A/B
Odds (case if not exposed) = C/D

OR = A/B / C/D

Question 25

How can you calculate a 95% Confidence Interval (CI) for case-control studies?

Answer 25

OR/EF

OR

RR/EF

(calculation for EF will be given)

Question 26

What are the potential biases of case-control studies?

Answer 26

Response + recall bias (particularly for controls)

Confounding

Question 27

What is a cohort study?

Answer 27

An analytical study in which a group of people e.g. a population sample (UK) or occupational sample (US Nurses Health Study) is followed up over time to compare incidence of an outcome in exposed and unexposed groups - usually prospective, but can also by retrospective or historical using administrative data

Question 28

What are the advantages and disadvantages of cohort studies?

Answer 28

Adv:

• Useful for rare exposures (not rare diseases)
• Can establish incidence
• Can estimate dose-response relationships
• Evidence of causality when cannot do an RCT

Disadv:

• Can be lengthy + expensive
• Complex ethics

Question 29

What are common measure of effect of cohort studies?

Answer 29

1. OR
2. RR
3. Incidence Rate Ratio (IRR) per unit of person time/person years
4. Hazard Ratio (HR) - rates at any moment in time

95% CI + null hypothesis value = 1 for all

Question 30

What are the potential biases of cohort studies?

Answer 30

Selection bias
Response bias
Loss to follow-up
Confounding

Question 31

What 2 types of study are the opposite of one another?

Answer 31

Case-control and cohort as case-control looks back at cases/controls to see if they was exposed or not whereas cohort find people exposed/unexposed and follow them up into the future to see if they get the disease or not

Question 32

What does the Risk Ratio/Relative Risk (RR) tell you?

Answer 32

How much more likely is it?

Question 33

What study design would be used for the research question: Do scores on the UKCAT predict performance in finals?

Answer 33

Cohort

Question 34

What study design would be used for the research question: Are doctors sanctioned by the GMC more likely to have had FtP Declarations on Registration compared to those not sanctioned?

Answer 34

Case-control

Question 35

What study design would be used for the research question: Do students on UG medical courses have different opinions about the proposed UKMLA compared to those on PG medical courses?

Answer 35

Cross-sectional

Question 36

What study design would be used for the research question: Do medical schools with high mean Prescribing Safety Assessment scores also have high mean Situational Judgement Test scores?

Answer 36

Ecological