Study Types

Question 1

What are the levels in the pyramid of evidence from strong to weak?

Answer 1

Sys Reviews-Meta-analysis (not narrative review)

RCTs

Cohorts studies

Case-control

Cross-sectional studies

Case series, case reports

Ideas, opinions, editorials, anecdotal

Question 2

What are the two types of study designs & their subtypes?

Answer 2

Experimental- Randomised controlled trial

Observational- Cohort, case-control, cross-sectional

Question 3

What does equipoise mean and why is it important?

Answer 3

Means you are not sure which treatment is more effective

Important b/c if you are certain one tx is better than the other, there are moral implications

Question 4

What are the essential components of a RCT?

Answer 4

Control group (from same population)

Randomization

Fully blinded

Identical TX & outcome measurements

Question 5

What are the steps to conducting a RCT?

Answer 5

1. Conduct a literature search
2. Statement of justification (justify why conducting study)
3. Objective & hypothesis (specific & measurable)
4. Identify outcome variables (what you are measuring)
5. Define the experimental unit (ind. animal, groups, parts of animal)
6. Define study population
7. Define the inclusion & exclusion criteria (want cases & controls similar in all respects possible)
8. Consult statistician (sample size calculations)
9. Recruit & enroll subjects

10 Random allocation

11 Identify other sources of bias

12 data collection

13 data grooming & analysis

14 Publish the study

Question 6

When a disease you want to study in a RCT is common, what type of enrollment can be used?

Answer 6

simultaneous enrollment

Question 7

When a disease you want to study in a RCT is uncommon, what type of enrollment can be used?

Answer 7

sequential enrollment

Question 8

What is the main reason for random allocation?

Answer 8

REDUCE BIAS

Question 9

What are the ways sources of bias can be reduced?

Answer 9

Blinding of all study participants

Standardization

Random allocation of treatment groups

Question 10

What is the purpose of performing a baseline comparison of the treatment groups?

Answer 10

confirm comparison groups are statistically the same

Question 11

In which type of study (experimental vs. observational) are the treatment groups not assigned?

Answer 11

Observational

Question 12

In which observational study are the study subjects healthy at the start?

Answer 12

cohort study

Question 13

Cohort study

Answer 13

starting with healthy animals, we prospectively compare them to animals with different exposure status

Question 14

What are the measures for different factors that can be used in a cohort study design?

Answer 14

compare incidence in exposed vs. non-exposed in the future

BEST measure of RR (strength of association)

Question 15

What are the advantages of a cohort study?

Answer 15

directly measures incidence

(good for looking at rare exposures or risk factors)

look at multiple diseases (outcomes) for a selected risk factor

stronger measure of RR (risk factor precedes disease)

Question 16

What are the disadvantages of a cohort study?

Answer 16

Very $$$

long time

Follow-up difficult

large # needed for rare dz

Ethical considerations

Question 17

What is the % of increase in risk for a factor if the RR=4.2?

Answer 17

320%

% increase= (RR-1) x 100

Question 18

If a risk factor is protective RR=0.57, then what is the % decrease in risk?

Answer 18

43%

% decrease= (1-RR) x 100

Question 19

Given a group of diseases animals, we retrospectively compared diseased animals to a similar group to healthy animals.

Answer 19

Case-control study

Question 20

What measure of the strength of the association can be made in a case-control study?

Answer 20

Odds ratio

Question 21

What are the advantages of a case-control study?

Answer 21

less $$

quicker

fewer subjects needed

no risk to the animals

Question 22

Which is better for a rare disease (case-control or cohort)?

Answer 22

case-control

Question 23

What are the disadvantages of a case-control study?

Answer 23

careful selection of a control population (BIAS)

historical risk factors - difficult to verify & must be remembered

incidence cannot be determined***