EBM Day 2

Question 1

Statistical Hypot

Answer 1

is there difference between groups,

Question 2

Ho

Answer 2

no association between x and y

Question 3

Fail to reject means

Reject null

Answer 3

fail to reject, never prove hypo because may be due to the 5% chance

consider possibility of type 1 error (unless p

Question 4

statistically sig means

Answer 4

result where reject Ho at whatever alpha level we set

Question 5

type 1 vs type 2 errors

Answer 5

t1-fail to reject null when true

t2- reject null when false

Question 6

T1 and T2 in releation to power, alpha, and beta

and what is alpha, beta, power

Answer 6

alpha-willingness to be wrong (reject null when we shouldn’t)
beta-=t2 error=willingess to fail to reject a false Ho (typically .1 or .2)
Power-1-beta=power to correctly reject false null (80%)-ability to detect or verify difference is real

Question 7

Sensitivity and Specificity

Answer 7

sensitivity=true positives/(true positives and false negs)

specificity=true negatives/(true negatives and false positives)

Question 8

power determination

Answer 8

alpha (more stringent, less power), beta (too lax, less power), prevalence of condition, magnitude of effect, sample size (more subjects make more power)

Question 9

Effect size

Answer 9

how big of a difference we look for

smaller effect size=larger the sample size needed

Question 10

Bonferroni adjustments

Answer 10

adjusting for multiple comparisons-increases type 2 error and decreases power

Question 11

T test

Answer 11

-compares difference between two means divided by variability in sample
assumes equal variances

Question 12

Mann Whitney U

Answer 12

Non parametric-operates on ranks

ignores mean and median

Question 13

Minimize T1 and T2 at same time?

Answer 13

there is always a tradeoff
Tolerate type 1 if false positive okay
Tolerate type 2 if if procedure may be serious danger to patient

Question 14

Increase power

Answer 14

lower beta, raising alpha, raising sample size, testing a large difference

Question 15

2x2 table

Answer 15

Draw=THERE ARE QUESTIONS AT END OF HIS SLIDES

Question 16

Risk and how offset

Answer 16

Probabilty of an outcome

Offset by intervention, treatment, prevention

Question 17

Primary, secondary, tertiary prevention

Answer 17

before disease
Catching early
treatment

Question 18

Pathogenic Triangle

Answer 18

host, environment, agents

Question 19

Risk factors

Answer 19

Anything which increase likelihood of disease

ex. other disease, environmental, genetic

Question 20

Environmental risk factors (5) and examples

Answer 20

chemical (oxidation substances), physical (radioactivity), biological (pathogens), psychosocial (stress), mechanical (heavy lifting)

Question 21

2x2 of Karaoke job Decision latitude

Answer 21

low demand, low decision-passive
low demand, high decision-low -strain (scientist)
high demand, low decision-high- strain
high demand, high decision-active (doctor)

Question 22

Latency

Answer 22

Period between exposure and disease

Question 23

Cohort study main question

Answer 23

Asking if incidence of an outcome in a group who were exposed different (greater/less) compared to incidence among similar group who were unexposed?

Question 24

Cohort study result

Answer 24

Get incidence rate of exposed

Get risk ratio, relative risk/difference when compare

Question 25

Picking a cohort

Answer 25

Should not have outcome when picked All should be at risk for outcome Should be observed over natural history of disease Observe over entire time of disease

Question 26

How to do cohort study

Answer 26

Find population where everyone at risk for something | Divide people into two groups depending on exposure to risk factor

Question 27

Cohort study other names

Answer 27

``` incidence study longitundinal study prospective study retrospective historical ```

Question 28

Retrospective vs Prospective Cohort

Answer 28

Prospective takes much longer time, assemble cohorts in present, choose which risk factors/confounder to measure, chose how to measure Retrospective may not have all data you need, cohorts assembled in past (by medical records), outcome accessed at later date

Question 29

Relative Risk

Answer 29

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

Question 30

Type 1/2 and false ... error

Answer 30

1. false postive | 2. false negative

Question 31

Causes of error

Answer 31

Chance-nondifferential-random error-type 2 error Bias-can be differential or non differential-type 1 error Confouding

Question 32

Differential vs Nondifferntial bias regarding direction

Answer 32

Differential-towards one direction or another | Nondifferential-towards norm

Question 33

Cross sectional study

Answer 33

Measure disease and time at same time

Question 34

Confounding Variable

Answer 34

Associated with DV and IV but not in pathway | Can cause t1 or t2 error

Question 35

Confounding by indication (and error associated with it)

Answer 35

Sicker patients are more likely to be treated and to have worse outcomes Increase T1 error ex. use of drug for really sick people associated with increased mortality because people are really sick (even if it helps)

Question 36

Decrease confoundng

Answer 36

Randomization-distribute potential confounders between groups Restriction-restrict a confounding variable during study duration (lower sample size and power though) Matching-match with people of similar characteristics Stratifcation-data separated by potential confounder-if confoudner present-risk ratios lower than in unstratified data (risk due to confounding no difference between strata) MV adjustment-control effects of many variables simulataneously

Question 37

Effect modification

Answer 37

effect mods-variables that change effect of exposure of interest on risk of disease AKA- interaction One exposure effects other exposure

Question 38

Selection Bias

Answer 38

Selective differences between comparison groups that impact relationship between exposure and outcome

Question 39

Selection bias examples (4)

Answer 39

healthy worker effect Self selection bias Withdrawal bias (primarily cohort studies) Information bias-Investigators who know exposure status may be more or less likely to ascertain the outcome (diagnostic bias)

Question 40

Fix for info bias

Answer 40

BLINDING of all personnel, investigators, and subjects to the exposure status of the subjects CAN ONLY FIX BIAS IN DESIGN STAGE or after with mv but then must measure bias

Question 41

Cohort Studies Adv (4)

Answer 41

good when exposure is rare, can minimize selection and measurement bias, can directly determine incidence rate and risk, can look at multiple outcomes from single exposure

Question 42

Cohort studies Disadv (5)

Answer 42

``` inefficient for rare outcomes Needs large sample long time to complete Loses to follow up Expensive potential ethical issues ```

Question 43

What test should i do for prevalence?

Answer 43

Cross sectional

Question 44

What test should i do for risk of harm

Answer 44

cohort, cross sectional

Question 45

tWhat test should i do for treatment or prevention

Answer 45

RCT, Cohort, case-control

Question 46

prognosis

Answer 46

cohort

Question 47

screening

Answer 47

rct, cohort, case control