Epidemiology

Question 1

Formula for positive likelihood ratio

Answer 1

+LR=sensitivity / (1-specificity)

Question 2

Formula for negative likelihood ratio

Answer 2

-LR=(1- sensitivity) / specificity

Question 3

Formula for absolute risk reduction

Answer 3

(ARR)=CER - EER

Experimental event rate-EER
Control event rate-CER

Question 4

Formula for relative risk

Answer 4

RR=EER / CER
OR
a/a+b / (c/c+d)

Question 5

Formula for odds ratio

Answer 5

Odds ratio=a/c (disased) divided by b/d (nondiseased)=(a x d) / (b x c)

Question 6

Formula for relative risk reduction

Answer 6

RRR=(CER - EER) / CER 
OR 
ARR/CER
OR 
1-RR

Experimental event rate-EER
Control event rate-CER

Question 7

Formula for NNT

Answer 7

1/ARR

Question 8

What is type 1 error?

Answer 8

• The error of rejecting a null hypothesis when it is actually true.
• It occurs when we are observing a difference when in truth there is none
• Usually accepted as 0.05

Question 9

What is power?

Answer 9

Probability of not making a type II error (1-ß) is called the Power of the test

Type II error= the error of accepting a null hypothesis when the alternative hypothesis is the true state of nature

-Usually accepted at 0.10-0.20

Question 10

List 5 criteria for an effective screening test

Answer 10

1. Epidemiology and natural Hx of condition adequately understood
2. Detectable risk factor, disease marker, latent period or early symptomatic stage
3. If carriers of mutation are identified as result of screen, natural Hx of carriers should be understood
4. Test should be simple, safe and validated
5. Test should be sensitive
6. Test should be acceptable to population
7. Effective treatment or intervention available for patients identified, with evidence of early treatment leading to better outcomes than late treatment

Question 11

What is intention to treat?

Answer 11

Participants are analyzed as members of the treatment group to which they were randomized regardless of their adherence to, or whether they received, the intended treatment

Question 12

List 4 ways to avoid systematic bias in a trial

Answer 12

Randomization with allocation concealment
Standardize measurements
Blinding participants and personnel
Blinding outcome assessors
Intention to treat analysis

Question 13

What is selection bias?

Answer 13

Systematic differences between baseline characteristics of the groups

Fixed by randomization

Question 14

What is measurement bias?

Answer 14

Systematic difference in measurements between the groups (e.g. patients reporting symptoms differently to MD vs RN)

Fixed by standardizing measurements

Question 15

What is performance bias?

Answer 15

Systematic differences between groups in the care that is provided, or in exposure to factors other than the interventions of interest

Fix with blinding participants and study personnel to allocation

Question 16

What is detection bias?

Answer 16

Systematic differences between groups in how outcomes are determined

Fix with blinding assessors to outcome

Question 17

What is attrition/analysis bias?

Answer 17

Systematic differences between those who withdraw from a study or are lost to follow-up

Fix with ITT

Question 18

What is reporting bias?

Answer 18

Systematic differences between reported and unreported findings – statistically significant findings are more likely to be reported

Question 19

How does odds ratio relate to relative risk?

Answer 19

OR provides a reasonable estimate of the relative risk for uncommon conditions

Question 20

What does an odds ratio of 1.5 mean?

Answer 20

The risk of an outcome was increased by 50% in the exposed group

Question 21

If the confidence interval for an odds ratio is 0.1 to 2.2, what does that mean?

Answer 21

True risk lies between 0.1 and 2.2

Crosses 1 and therefore confidence interval is not statistically significant

Not very precise as the range is very high