EBD

Question 1

What is the formula for risk?

Answer 1

no. of events of interest/total number of observations

Question 2

What is the formula for odds?

Answer 2

no. of events of interest/no. without the event

Question 3

What is the formula for odds?

Answer 3

no. of events of interest/no. without the event

Question 4

Difference between relative risk vs absolute risk reduction?

Answer 4

relative risk - reduction between 2 events in percentage

absolute risk is the difference in percents by subtraction

Question 5

How do you figure out the relative/absolute risk?

Answer 5

using a 2x2 contingency table

Question 6

What is the value of no difference for the absolute risk difference?

Answer 6

0

Question 7

What is the value of no difference for the absolute risk difference?

Answer 7

0

Question 8

What is the number needed to treat? How do you work it out?

Answer 8

Question 9

What is the value of no difference for risk ratio (relative risk)?

Answer 9

1

Question 10

Why might you have a wide confidence interval?

Answer 10

small population size

Question 11

Pyramid of evidence

Answer 11

Question 12

What are cohort studies good for?

Answer 12

Incidences of disease

Measures exposires and follows people over time

Question 13

Design elements of RCTs?

Answer 13

-inclusion/exclusion criteria
-control/comparison groups
-randomisation
-blinding/masking

Question 14

Which part of randomised control trial design minimises bias and how?

Answer 14

-randomisation of participants to groups = prevents researchers from influencing which participants go into each group (might put certain people in groups so that the drug/intervention looks better)
-blinding/masking of participant nad researcher (prevents maniupulation of results and assessment of the groups)

Question 15

Which part of randomised control trial design minimises bias and how?

Answer 15

-randomisation of participants to groups = prevents researchers from influencing which participants go into each group (might put certain people in groups so that the drug/intervention looks better)
-blinding/masking of participant nad researcher (prevents maniupulation of results and assessment of the groups)

Question 16

How can you check trial quality?

Answer 16

Using the Consolidation of the Standards of Reporting Trials publication (CONSORT)

Can also use CASP tool (critical appraisal skills programme)

Question 17

How do you assess if a RCT results are valid?

Answer 17

-Did the trial address a clearly focused issue (PICO)
-was assignment randomised?
-were all of the patients who entered the trial accounted for at its conclusion?
-was it blinded/masked
-were the groups similar at the start of the trial
-were the groups treated equally apart from the intervention

Question 18

Benefits of systematic revies over single studues?

Answer 18

-saves readers time
-provide reliable evidence (unbiased comprehensive picture of body of evidence)
-resolve inconsistencies
-identify gaps (where good studies are not available)
-identify when Q’s have not been fully answered
-explore differencies between studies

Question 19

What are the key characteristics of a systematic review?

Answer 19

-well formulated Q (PICO)
-comprehensive data search
-unbiased selection and abstraction process
-assessment of papers
-synthesis of data

Question 20

What are the key characteristics of a systematic review?

Answer 20

-well formulated Q (PICO)
-comprehensive data search
-unbiased selection and abstraction process
-assessment of papers
-synthesis of data

Question 21

What is reporting bias?

Answer 21

-positive results are more likely to be pubished, published more rapidly, in English and more likely to be cited as this is what sells

When doing systematic reviews need to look at unpublished data too

Question 22

Why does the risk of bias need to be calculated?

Answer 22

because it determines the extent to which results of a study can be believed

Question 23

Why does the risk of bias need to be calculated?

Answer 23

because it determines the extent to which results of a study can be believed

bias can overestimate or underestimate the effect