Lecture 13: Evidence Evaluation: Observational Studies II

Question 1

What kind of study is case-control?

Answer 1

Observational, analytic study

Question 2

What is the starting point of a case-control study

Answer 2

Outcome (ex.disease)

Question 3

Steps of a case-control study

Answer 3

1a. Identify cases (individuals with outcome or disease)

1b. Identify controls (individuals without the outcome or disease of interest)

2. Look at exposure histories

Question 4

Case-control study flow

Answer 4

Look at canvas

Question 5

Why can’t we measure risk of an outcome when analyzing case-control studies?

Answer 5

Because we don’t know when they occur (study design does not involve a follow up)

Question 6

Does relative risk apply to a case control study?

Answer 6

No

Question 7

What is the relevant concept when analyzing case-control studies

Answer 7

Odds

Question 8

What are odds?

Answer 8

The odds of an even can be defined as the ratio of the number of ways the event can occur to the number of ways the even cannot occur

Question 9

Interpretation is there a relationship? OR > 1

Answer 9

Association: Exposed group have higher odds of outcome
compared to unexposed group

Question 10

Interpretation is there a relationship? OR = 1

Answer 10

No association

Question 11

Interpretation is there a relationship? OR < 1

Answer 11

Inverse association: Exposed group have lower odds of
outcome compared to unexposed group

Question 12

Interpretation: What is the strength of the relationship: OR > 1

Answer 12

% increase = (OR - 1) x 100

Question 13

Interpretation: What is the strength of the relationship: OR < 1

Answer 13

% decrease = (1 - OR) x 100

Question 14

What is the problem with mis-labelling?

Answer 14

“Case-control” which refers to the design of the study is often
confused with “cases” and “controls” which refer to participants in a
study

• Cohort study is often mislabelled as case control study

Question 15

What are the advantages of a case-control study?

Answer 15

Advantages
* Good for less common or rare
outcomes
* Versus a cohort study where one
would have to study many
participants over a long follow-up
for outcomes to occur
* Quick(er) and less expensive
(compared to cohort study)
* Can study multiple exposures

Question 16

What are disadvantages to case-control studies?

Answer 16

• Cannot establish incidence
• Cannot study multiple outcomes

Question 17

What is a cross-sectional study? Analytic or descriptive?

Answer 17

An observational design that surveys exposures and/or outcomes at a single point in time (‘snapshot’)

• Can be analytic (i.e. measure association between exposure and disease)
• Can be descriptive (i.e. focus on exposure or disease)

Question 18

Describe a cross-sectional study as an analytic study

Answer 18

Temporal sequence of exposure and outcome impossible to determine (don’t know which occurred first) à main limitation (compared to cohort and case-
control studies)

Question 19

Describe a cross-sectional study as a descriptive study

Answer 19

Examines patterns (of disease, health behaviours)
* Measures prevalence not incidence

Question 20

What measure is used for an analytic cross-sectional study?

Answer 20

Odds Ratio

Question 21

What do you study in a descriptive cross-sectional study?

Answer 21

Studying:
* Frequency of disease
* Frequency of health behaviour(s)

Question 22

What do descriptive studies describe/measure?

Answer 22

Occurrence of an outcome (i.e. disease)

Question 23

Do descriptive studies evaluate intervention?

Answer 23

No

Question 24

What is the start and end of a descriptive study?

Answer 24

Start: no hypothesis
End: possible hypothesis

Question 25

What kind of studies are case report and case series?

Answer 25

Observational, descriptive study

Question 26

What are case reports and case series?

Answer 26

Detailed presentation
* Case report – 1 or 2 cases
* Case series - > 2 cases

Question 27

What do case reports and case series do?

Answer 27

• Report a new or unique condition
• Describe previously undescribed disease
• Show unexpected link between diseases
• Show unexpected new therapeutic effect
• Report adverse events

Question 28

What don’t case reports and case series do?

Answer 28

Measure disease incidence
* Case series assess prevalent
(existing) disease
* Identify risk factors
* Identify cause of disease

Question 29

When outcome is rare what study do we use?

Answer 29

Case-control

Question 30

When there are multiple exposures what study do we use?

Answer 30

Case-control

Question 31

When investigating an outbreak what study do we use?

Answer 31

Case-control