Epidemiology

Question 1

Sensitivity

Answer 1

Percent of animals with a disease who test positive

Question 2

Specificity

Answer 2

Percent of animals without a disease who test negative

Question 3

What does a negative result on a sensitive test tell you?

Answer 3

Rules disease out

Question 4

What does a positive result on a specific test tell you?

Answer 4

Rules disease in

Question 5

Positive predictive value

Answer 5

Proportion of patients that test positive that are actually positive

Question 6

Negative predictive value

Answer 6

Proportion of patients that test negative that are truly negative

Question 7

How can you improve positive predictive value (2)

Answer 7

Increase disease prevalence
Increase test specificity

Question 8

How can you improve negative predictive value (2)

Answer 8

Decrease disease prevalence
Increase test sensitivity

Question 9

What is a cross-sectional study?

Answer 9

All observations on a given group of individuals (outcomes and exposures) are made at one point in time

Question 10

List one advantage and one limitation of cross-sectional studies

Answer 10

A: Determines prevalence of disease
L: Cannot determine incidence or temporality

Question 11

What is a cohort study?

Answer 11

Selects a group based on exposure then follows them through time and compares to non-exposed group

Question 12

List one advantage and one limitation of cohort studies

Answer 12

A: can establish incidence and temporality
I: prone to confounding variables, time-consuming, information bias (retrospective)

Question 13

What is a case-control study?

Answer 13

Selects individuals based on outcome and follows them through time to detect the frequency of an exposure

Compares to similar group without the outcome/disease

Question 14

List one advantage and one limitation of case-control studies

Answer 14

A: good for rare disease, evaluate for multiple risk factors
I: recall bias, cannot estimate incidence or prevalence

Question 15

Limitation of uncontrolled clinical trials

Answer 15

Lack of control group makes assessment of treatment affect difficult

Question 16

What is a Chi-squared test used for

Answer 16

To determine whether a difference between two categorical variables is due to chance or a relationship between them

Question 17

Limitations of Chi-squared test (3)

Answer 17

Requires minimum sample size (50)
Assumes random sampling
Does not provide info about strength of a relationship

Question 18

What is a T-test used for

Answer 18

To determine if there is a significant difference between the means of two groups

Question 19

Advantage of a T-test

Answer 19

Can be used with small sample sizes (<30)

Question 20

Limitations of T-tests (4)

Answer 20

Requires normality assumption
Low power with high sample sizes
Not useful for categorical data
Sensitive to outliers

Question 21

What is an ANOVA used for?

Answer 21

To compare the means of >2 groups

Question 22

Limitations of an ANOVA (3)

Answer 22

Normality assumption
Doesn’t specifiy which groups are different from each other
Requires equal group size

Question 23

What is type I (alpha) error?

Answer 23

Concluding that a treatment is better than the control when it is not (false positive)

Question 24

What is the typical accepted alpha level?

Answer 24

p < 0.05

Question 25

What is type II (beta) error?

Answer 25

Concluding that there was no treatment effect when there was (false negative)

Question 26

What beta level is typically accepted?

Answer 26

0.2

Question 27

How is power calculated?

Answer 27

1- beta level

Question 28

Performance bias

Answer 28

Prior knowledge of which group patients belong to, resulting in differences in care level

Question 29

Investigator bias

Answer 29

Expectation of a response that may influence the subjective interpretation of results from either a placebo or treatment

Question 30

Misclassification bias

Answer 30

Assignment of subjects to groups is made incorrectly

Question 31

Measurement bias

Answer 31

Occurs when uniform standards of measurement are not maintained

Question 32

Name two ways to assess correlation

Answer 32

Pearson correlation coefficient Spearman correlation coefficient Both -1 to +1 with either -1 or +1 indicating perfect association, while 0 indicates no association

Question 33

Odds ratio

Answer 33

Ratio of the odds of the outcome in an exposed group compared to the odds of an outcome occurring in an unexposed group

Question 34

Test accuracy

Answer 34

(# true positives + # true negatives)/ total # in test population

Question 35

Relative risk

Answer 35

Probability of an occurrence in exposed group/ probability of event occurring in non-exposed group

Question 36

Standard deviations?

Answer 36

68.3%, 95.5%, 99.7% (1, 2, 3 SD, respectively)