BioStats

Question 1

Incidence

Answer 1

NEW cases that develop in a population over certain period of time.
Denominator only takes into account the population at risk for acquiring the disease

Question 2

Prevalence

Answer 2

Measure of the TOTAL number of cases (new and old) measured at a particular point in time

Question 3

Standardized mortality ratio

Answer 3

Divide the observed number of deaths by the expected number of deaths.
Ex: SMR of 2.0 means that the observed mortality in a particular group was twice as high as that in the general population

Question 4

Attack rate

Answer 4

Divide the number of patients with the disease by the total population at risk

Question 5

Relative risk definition

Answer 5

Compares the probability of DEVELOPING an outcome between two groups over a certain period of time
Implies PROSPECTIVE study

Question 6

Odds ratio definition

Answer 6

Compares the chance of exposure to a risk factor between cases and controls
Implies a RETROSPECTIVE

Question 7

Relative risk math

Answer 7

(Exposed and have disease now/Total exposed)/(Unexposed that have disease now/Total unexposed)

Question 8

Odds ratio math

Answer 8

(Have risk factor and have the disease/Do not have risk factor and have the disease)/(Risk factor and do not have disease/Do not have risk factor and do not have the disease)

Question 9

How to determine the variability percentage with correlation coefficient

Answer 9

Square the correlation coefficient, for example if =-0.8 then the variability would be 0.64

Question 10

Attributable risk

Answer 10

AR = (Disease in people with risk factor/total people with risk factor) - (Disease in people without risk factor/total people without risk factor)

Question 11

Number needed to treat

Answer 11

NNT = 1/risk difference between the two groups

Question 12

Null hypothesis

Answer 12

States that there is NO association between the exposure of interest and the outcome

Question 13

Sensitivity

Answer 13

TP/(TP+FN)

Question 14

Specificity

Answer 14

TN/(TN+FP)

Question 15

Positive likelihood ratio

Answer 15

Sensitivity/(1-Specificity)

Question 16

Positive predictive value

Answer 16

PPV = TP/(TP+FP)

Question 17

Negative predictive value

Answer 17

NPV = TN/(TN+FN)

Question 18

Z score

Answer 18

How many standard deviations a given value is from the mean

Question 19

Two sample t-test

Answer 19

Compare means of two independent groups (example, depression scores in patient who are taking beta blockers and those who are not)
Gives you a p value

Question 20

Paired t-test

Answer 20

Compare two means, but unlike two sample t-test, in paired t-test you use in situations where the two means are different (example, testing weight loss drug and having to compare the weight of patients before and after drug use)
Think of as values that are paired to the same test subject
Gives you a p value

Question 21

Analysis of variance (ANOVA)

Answer 21

Compare means of three or more variables

Question 22

Chi-squared

Answer 22

Use when outcomes are yes/no, high/low, etc. Values in these questions will be in a 2x2 table.

Question 23

Fisher’s exact test

Answer 23

Use in situations where values in a 2x2 (like chi-squares) are <10.

Question 24

Wrongfully concluding that there is an association between exposure and disease when in fact there is none

Answer 24

Type I error

Alpha

Question 25

Wrongfully concluding that there is no association between exposure and outcome, when in fact there is one.

Answer 25

Type II error | Beta

Question 26

Power of the study

Answer 26

Probability of detecting an association if there is one | 1 - beta