Biostats Flashcards Preview

Boards > Biostats > Flashcards

Flashcards in Biostats Deck (34):
0

What is the equation for sensitivity?

TP/(TP+FN)

1

What does 100% sensitivity indicate?

Low false negative rate

2

What is sensitivity used for?

Used for screening in diseases with low prevalence

3

What is the equation for specificity?

TN/(TN+FP)

4

What does 100 percent specificity indicate?

A low false positive rate

5

What is the equation for positive predictive value?

TP/(TP+FP)

6

What is the positive predictive value?

Probability that person actually has the disease given a positive test result

7

What is negative predictive value?

The probability that a person is actually drug free given a negative test result

8

What is the equation of negative PPV?

TN/(FN+TN)

9

What is the true positive rate?

Sensitivity

10

What is the true negative rate?

Specificity

11

What is the equation for calculating incidence?

Number of new cases/number of people in population at risk (don't already have the disease)

12

What is the equation for prevalence?

Number of cases in a population/total number of people in the population

13

What is the equation for odds ratio?

Ad/bc
Used in case-control studies (outcome --> exposure)

14

What is the equation for relative risk?

A/(a+b)/c/(c+d)

15

What is relative risk?

Risk among exposed/risk among unexposed

16

What is the equation for absolute risk?

(A/a+b) - (c/c+d)
(RR-1)/RR
#of adverse events in tx/# of adverse events in placebo

17

What is the number needed to treat?

The number of patients who need to be treated for 1 patient to benefit

18

What is the equation for the number needed to treat?

1/absolute risk
1/ (event rate -control) - (event rate - tx)

19

What is the number needed to harm?

Number of patients who need to be exposed to a risk factor for 1 patient to be harmed

20

What is the equation for the number needed to harm?

1/attributable risk = 1/(a/a+b) - (c/c+d)= 1/((RR-1)/RR)

21

What is the equation for event rate?

#of subjects per tx/total # of subjects

22

What is precision?

The consistency, reproduciblity and reliability of a test

23

What lowers precision?

Random error

24

What is accuracy of a test?

Validity
Absence of systematic error

25

What is the equation for SEM?

SD / square root sample size
So as sample size increases, standard error of mean decreases

26

What is the null hypothesis?

Says there is no difference btween two groups
P = high

27

What is the p-value?

The likelihood of observing what we did if the null is true.
The lower the p-value the more evidence for the alternative hypo

28

What is a type I error?

False positive error
Stating there s a difference when there really isn't.
P-value is judged against this
Ex: convicting an innocent man

29

What is a type II error?

Stating there isn't an effect or different when there is one
False negative error
Ex: Setting a guilty man free

30

What is power?

The probablility of seeing a difference when one is there
Ex: convicting a guilty man
= 1- B

31

What does power increase with?

Sample size
Expect effect size
Precision of measurement

32

What is the equation for confidence interval?

Mean - Z(SEM) to mean + Z(SEM)
Z= 1.96 for 95% CI
Z = 2.58 for 99% CI

33

What is the equation for likelihood ratio?

Sensitivity/1-specifity
The higher the better likelihood the patient has a disease they are being treated for