Epidemiology / Biostatistics Test 1

Front 1

RCT

Back 1

1) hypothesis
2) inclusion ↔ exclusion criteria
3) informed consent
4) DSMB
5) IRB
6) randomize accepted subject
7) asses outcomes

Front 2

DSMB

Back 2

Data Safety Monitoring Board

under what conditions will we stop a trial?

→ overwhelming differences or futility

Front 3

IRB

Back 3

Institutional Review Board

give Yes/No/Maybe

review ethical issues of study

Front 4

Internal Validity Measures

Back 4

1) subjects = entry criteria?
2) @ baseline, are study groups comparable?
3) loss to follow-up?
4) compliance assessed?
5) blinded?
6) agree w/ definition of outcome?
7) statistical power?

Front 5

External Validity Measures

Back 5

Cost
Availability
Believable?
Clinically beneficial?

Front 6

RCT +/-

Back 6

+ gold std.
+ equally distributes confounders (known and unknown)
+ incidence data & multiple outcomes
- unethical (in cases)?
- costly (time & money)
- poor for rare outcomes

Front 7

observation ↔ intervention

Back 7

obs: case report
case series
cross sectional
case control
cohort

int: RCT
cohort

Front 8

Confounding

Back 8

related to distribution of variables across study arms

Front 9

Bias

Back 9

related to errors in study design

Front 10

ENHANCE

Back 10

statins for ↓ cholesterol:
↑ clearance ↓ production

are drugs better in combo?

Chol. ↓ 27% w/ combo, but no change in atherosclerotic plaque

Front 11

Case Report

Back 11

singular report on a case

to alert medical community

no comparison group

Front 12

Case Series

Back 12

plural report on a series of observed cases

to alert medical community

no comparison group

Front 13

Cross Sectional Study

Back 13

Prevalence

data mining for association and observations

Front 14

Case Control Study

Back 14

subjects chosen by outcome

+ rare diseases
- rare exposures

+ quickly study multiple exposures

Front 15

Cohort Study

Back 15

subjects chosen by exposure status

prospective ↔ retrospective

+ rare exposures
- rare outcomes
+ generate incidence data

Front 16

Nested Case Control

Back 16

internal case control w/in context of a prospective study

Front 17

Prevalence

Back 17

total cases
---------------- = I x duration
total population

"snapshot in time," inc. old cases

Front 18

Incidence

Back 18

Cumulative I. (total cases over a set period of time)

Incidence Density -- cases*personyears^-1

*New Cases

Front 19

Mortality

Back 19

deaths
----------------
population

Front 20

Case Fatality

Back 20

deaths
--------------------
those affected

Front 21

RR

Back 21

Ie
----
Io

Front 22

AR

Back 22

Ie - Io

Front 23

OR

Back 23

ad
-----
bc

Front 24

z

Back 24

x - mean
-------------
SD

Front 25

variance

Back 25

spread

Front 26

SD

Back 26

√v

Front 27

Epidemic

Back 27

occurrence of a disease in an area with a greater than usual expectancy

Front 28

Endemic

Back 28

constant presence of a disease in an area

Front 29

Pandemic

Back 29

worldwide epidemic

Front 30

Mean

Back 30

Average

Front 31

Median

Back 31

Middle value

(if even # of values, average 2 in middle)

Front 32

Mode

Back 32

most common value

*poss. to be bi-modal

Front 33

Range

Back 33

highest - lowest

Front 34

Percentile

Back 34

% of variables at or below value

Front 35

Interquartile Range

Back 35

middle 50% {exclude <25th & >75th percentile}

Front 36

Coefficient of Variation

Back 36

SD
---------
mean

Front 37

Continuous Data

Back 37

like a wave on the ocean

data limited only be accuracy of measurement

Front 38

Discrete Data

Back 38

are OR are not

categorical, nominal, ordinal

Front 39

Z-score

Back 39

# of SD away from average

--> implies normal distribution

Front 40

Gaussian Distribution

Back 40

1 SD = 68%
2 SD = 95%
3 SD = 98%

if normal dist., mean=median=mode

Front 41

Problems of Internal Validity

Back 41

Chance

Bias

Confounding

Front 42

Chance

Back 42

design inherently flawed

--> erroneous conclusions

Front 43

Information Bias

Back 43

"garbage in --> garbage out"

Front 44

Interviewer Bias

Back 44

unique to case-control

*blind interviewers and use standard questions

Front 45

Selection Bias

Back 45

way subjects are selected associated with disease

(ex. story of lupus alcohol link)

only in case control / retrospective cohort

*varied definitions

Front 46

Reporting Bias

Back 46

social concerns → misreporting by subject

Front 47

Recall Bias

Back 47

notion that cases think harder about exposure status than controls and thu overestimate their status

unique to case control

Front 48

Surveillence / Detection Bias

Back 48

avoid by blinding treating physicians

notion that cases may be looked at in more detail, and thus more incidences of disease found

Front 49

Loss-to-Follow Up Bias

Back 49

10% ++

>20% -- Red Flag

is it differential or even?

Front 50

Misclasification Bias

Back 50

entries ↔ outcomes

*sometimes difficult to conclude if condition present, may put things in the wrong bucket

Front 51

Random / Non-differential

Back 51

equal in both study arms

truth → null hyp.

avoid w/ strict definitions of outcomes

Front 52

Differential / Non-random

Back 52

can push in both directions

avoid by blinding everyone

*more important w/ subjective outcomes

Front 53

Confounding

Back 53

one side stacked for potential confounders

*association between result & exp.
*independent risk factor

cannot be from intermediate
failure to fix → bias

Front 54

Interaction

Back 54

:-)
external validity issue

Are risks / benefits equal for all populations?

→ a good thing to know for treating various populations clinically

Front 55

Health Determinants

Back 55

Fixed <----------------------> Modifiable
(genetics, time)

env., lifestyle
care, society

Front 56

Genomic Medicine

Back 56

NOT Genetics

look at whole population distribution

Front 57

Pharmacogenomics

Back 57

goal: to tailor drug therapies based on a patient's genetics

Front 58

Aging

Back 58

Lose adaptive capacity with age

theories:
genetic -- pre-programmed on cellular level
stochastic -- chance, accum. of env. factors.

Front 59

Health Determinants

Back 59

Environment
Lifestyle
Medical Care
+Society

**all hopelessly intertwined

Front 60

Social Capital

Back 60

Ability to take advantage of people who care about you

-->getting involved is a strong + influence

Front 61

Stress

Back 61

a normal physiological response

chronic response has long term (-) effects

What do you do about it {clinically}?

Front 62

Common Health Behaviors

Back 62

diet
exercise
smoke
Alcohol / drugs
Sex
Recklessness

**Our ability to change them is ABYSMALL!

Front 63

Our impact as physicians?

Back 63

↑ prolonging life
↑ reducing suffering

↓curing disease / healing
↓ prevention

Front 64

Limitations of Statistical Significance

Back 64

Statistical Significance ≠ Patients' Benefit

Front 65

P-value

Back 65

Probability of getting a difference of x% or larger due to chance alone

Front 66

α

Back 66

Level of Significance

arbitrary "line in the sand" to help us make a decision

freq = 0.05

Front 67

P < or = α

Back 67

chance is unlikely {but NOT impossible}

∴ reject null on statistical significance
∴ likely hood of getting result or better due to chance alone <5%

Front 68

P > α

Back 68

fail to reject null

∴ conclude result is due to chance alone (not statistically significant)

Front 69

Power

Back 69

1 - P(type II error)

ability to detect a difference of a certain magnitude or more if one exists

Front 70

Power (cont.)

Back 70

opp. P value, calc. assuming alt. hyp. true

unlikely to find true differences w/ small sample

↑ power α ↑ sample size

Front 71

1 Tail <-----> 2 Tail

Back 71

1) existing vs. {better} treatment
***not worried about negative outcomes

2) discerning differences btw A & B (regardless of direction)

Front 72

Type I Error

Back 72

false +

conc. diff., but no diff.

p < α, but still due to chance

Front 73

Type II Error

Back 73

false -

conc. no diff., but truth is difference

p > α

Front 74

α adjustments for multiple subgroups

Back 74

"data dredging" will find something where p < α

∴ α'= α / # of analyses