BIOSTATS

Question 1

wider CI is most likely due to what

Answer 1

smaller sample size

Question 2

RR 2.0, CI (1.2-2.8) - is this statistically significant

Answer 2

yes - b/c does not include 1

Question 3

95% of study data will lie b/w how many standard deviations from the mean

Answer 3

2 SD from the mean

Question 4

99.7% of study data will fall b/w how many standard deviation form mean

Answer 4

3 SD from the mean

Question 5

68% of study dad will fall b/w how many SD of the mean

Answer 5

1 SD of mean

Question 6

if the test is negative - what is the probability that the patient will have the disease

Answer 6

1 - negative predictive value

Question 7

cofounding bias

Answer 7

bias that results when the exposure disease relationship is mixed with the effect of extraneous factors

Question 8

what study can have recall bias

Answer 8

retrospective study

- case control studies

Question 9

chi-squared - when to use

Answer 9

when comparing the proportions of a categorized outcome

2x2 table ( ex: HRT vs. no HRT)

Question 10

when to use two sample x test

Answer 10

two MEANS not proportions

Question 11

when to use two sample T test

Answer 11

To compare two means NOT proportions

Question 12

when to use ANOVA

Answer 12

compare MEANS of 3 or more variables

Question 13

Meta-analysis

Answer 13

spidemiologic method of pollen the data from several studies to do an analysis having a relatively big statical power

Question 14

what does it mean to say randomization was successful

Answer 14

means that the baseline characteristics of the patients were the same b/w treatment and placebo gourds

Question 15

NNT

Answer 15

1-ARR

Question 16

hawthorne effect

Answer 16

tendency of the study population to affect the outcome since they are aware that they are being studied

Question 17

incidence

Answer 17

number of new cases of disease that develops in a specific population over a specified time period

Question 18

prevalence

Answer 18

existing case of disease in population at a particular point in time

Prevalence = incidence x duration of disease

Question 19

increase in prevalence but same incidence - what can cause this?

Answer 19

• attributed to factors that prolong duration of disease

ex> improved quality of care and disease management

Question 20

case control studies help to control what type of bias

Answer 20

confounding

Question 21

what is an efficient method to control confounding

Answer 21

matching - patients selected with same age, race and are used as control

• used in CASE CONTROL STUDIES

Question 22

what study is the best to pick up incidence of disease

Answer 22

COHORT

Question 23

retrospective cohort

Answer 23

population with disease

Compares disease incidence with and without a risk factor

Question 24

what study type looks at prevalence

Answer 24

Cross sectional

Question 25

What study looks at risk factor frequency

Answer 25

Case control Diseased cases and non disease controls - looks at risk factor frequency

Question 26

methods to control cofounding

Answer 26

Design Stage: - MATCHING - RESTRICTION (limiting study participation to individuals with specific characteristics i.e.: only men - RANDOMIZATION Analysis Stage - stratified analysis Statistical modeling

Question 27

post hoc analysis

Answer 27

performing unplanned statistical test on patterns that were identified after the fact in the data after completed study

Question 28

effect modification

Answer 28

results when an external variable positively or negatively impacts the effect of risk factor on the disease Ex: estrogen treatment increase risk of VTE and this effect is augmented by smoking

Question 29

factorial design studies

Answer 29

randomization to different interventions with additional study of 2 or more variables

Question 30

cross over study

Answer 30

one group patients - get treatment A Other group of patients get treatment B Then they cross over

Question 31

Attributable risk percent or etiologic fraction

Answer 31

ARP = risk in exposed - risk in unexposed / risk in exposed

Question 32

rr 4.0 95% CI 1.5 - 6.5 The risk of getting disease 4Xmore with alcohol than in people without using alcohol what is the precent of disease in people who drink alcohol

Answer 32

They want you to calculate ARP (attributable risk percent) ARP = 4-1/4 = 0.75%

Question 33

generalizability or external validity

Answer 33

applicability of the study results to other populations

Question 34

median

Answer 34

blue located in the middle of dataset

Question 35

mean

Answer 35

all numbers added together an then divided by the number of numbers

Question 36

mode

Answer 36

the most frequent number in data set

Question 37

smoking and alcohol as cofounders

Answer 37

smoking should be related to alchol comsumption

Question 38

tighter CI - means

Answer 38

the more precise the results

Question 39

mode of inheritance for hemophilia A

Answer 39

X linked recessive disorder

Question 40

two sample t test vs. two sample x test

Answer 40

Two sample t test - compares two means Two sample z test - compares two means but population ( not sample) - since population variance is not usually known - test has limited applicability