Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

MFM Oral Boards 2020 > Biostats ABC > Flashcards

Biostats ABC Flashcards

1
Q

Power analysis explained

A

4 components

  • 3 are known and your solving for one that is not
    1) effect size
    2) significance level =type 1 error= alpha=probability of finding an effect that is not there = typical 0.05 (5%) (most similar to p-value)
    3) power= beta= type 2 error= probability of finding no effect that actually is there = failing to reject the null = typically 0.2 (80% chance of identifing)
    d) sample size (n) - what you are solving for
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

how do you calculate effect size

A

estimated from literature

clinical significance?

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

why use case control

A

rare outcomes

retrospective

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

why use cohort

A

start with exposure

can be prospective or retrospective

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

accurate

A

free of error or bias

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Precise

A

minimal effects from chance

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

types of bias

A

recall
reporting - subjects in one group more likely to report prior events
selection (food diary)
inter/intra observer

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

confounding variable

A

when a characteristic or variable is not distributed the same in the study vs the control (chance or bias)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

sensitivity

A

of everyone with the disease this % will test positive

True positive/all with disease
A/A+C
Disease on top
exposure/test on sides

Does not change with prevelance

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

specificity

A

of everyone without the disease this % will test negative

true neg/all negative
D/B+D

of all the patient’s without a disease x% had a negative test

does not change with prevelance

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

PPV

A

if the test is positive the chance the patient actually has the disease
- increases with prevalence
A/A+B

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

NPV

A

the probability that if the test is negative the subject actually does not have the disease

D/D+C
- decreases with prevalence

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

ROC curve

A

x axis- rate of false positive (1-specificity(true negative))
y- axis rate of true positive (sensitivity)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

two types of experimental studies

A

randomized/non-randomized

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

two type of observational studies

A

analytical vs descriptive
cohort
case control
cross section

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

cross-sectional study

A

looks a prevelance and not incidence

temporal relationship can be unclear

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

stats for cohort study

A

true incidence rate
attributable risk
relative risk

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

case control studies

A

careful of control group and recall bias
can only calculate odds ratio - when outcome is rare it is very similar to rr

consider more stringent inclusions to ensure less confounding (preeclampsia with severe features requiring delivery vs preelcampsia)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

major problem with non-randomized experimental studies

A

selection bias

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

radomized controlled trials plus and minus

A

avoid confounding and selection bias

external validity can be a concern -volunteers can be different from the population

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

ratio

A

numerator is not included in the denomator

MMR

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

proportion

A

numerator is included in the denomator
-prevalance (proportion)

dimensionless

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

Rate

A

numerator is included in the denomator and takes into consideration time
- incidence rate

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

relative risk

A

Frequency of the outcome in an exposed group / frequency of he outcome in the unexposed

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Q

odds ratio

A

case control- odds of exposure among the cases/ odds of exposure in controls

cohort/cross sectional/RCT
- ratio of the odds in favor of the disease in the exposed vs unexposed. indicate the RR when the prevelance of the outcome is <5-10%

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
26
Q

what is confidence interval

A

precision of study results.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
27
Q

discriptive- ecological correlational studies

A

look for associations - trend analysis, healthcare planning, hypothesis generation

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
28
Q

correlation studies

A

measure the association between exposure and outcome with the correltation coefficient r

  • can not address causation, can not control for confounding
29
Q

informational bias

A

incorrect determination of exposure outcome or both

-misclassification

30
Q

3 types of bias

A

selection - berkson-
(different management when expsoure is known, Neyman- selection inherently excludes pts, unmasking, nonrespondent )

informational - observation, classification, or measurement- ascertainment bias, recall,

confounding

31
Q

control for confounding

A

restriction- decreased external vaildity

matching- difficult recruitment, can not measure the effect of confounder

stratification- post hoc restriction - mantel haenszel - if it differs from the crude effect than confounding is present - multi variate logistic regression

32
Q

what does the p value measure

A

chance - type 1 error- false positive

33
Q

2 risks that an association is not causal when statistical significance is met

A

bogus- bias
indirect- confounding
or real!

34
Q

causal criteria

A 
cause is before effect 
strong associations ( RR >3, OR >4) 
consistency
dose response 
specificity of association (only one outcome)
biologic plausability 
experimental evidence 
analogy (similar to other associations)
35
Q

nested case- control

A

within a cohort

36
Q

what type of study is a before after study

A

most like cohort

37
Q

stats in cohort studies

A

RR
hazard ratio (cox proportional hazard-dicotomous results )
survival curves (Kaplan Meier-log rank compares curves)
incidence rate

38
Q

bias in cohort

A

exposure status can change( may want to quantify exposure)
loss to follow-up
likely selection bias

39
Q

Interval / ordinal /nominal

A

Interval - scale
ordinal- descrite numerals
nominal- yes/no

40
Q

unpaired t test

A

interval
normal distrobution
2 groups
independent

41
Q

paired t test

A

interval
normal
2 groups
dependent

42
Q

ANOVA

A

Interval
normal
> 2 groups
independent

43
Q

repeated measure ANOVA

A

interval
normal
>2 groups
dependent

44
Q

wilcoxon signed rank test/ sign test

A

ordinal (or non-parametric (non-normal))
2 groups
dependent

45
Q

mann whitney/ wilcoxon rank sum

A

ordinal (or non-parametric (non-normal))
2 groups
independent

compares medians

46
Q

kruskal- wallis test

A

ordinal (or non-parametric (non-normal))
>2 groups
independent

47
Q

Friedman two way anova

A

ordinal (or non-parametric (non-normal))
> 2 groups
related

48
Q

Chi-square/ fishers exact (small numbers)

A

nominal / categorical
independent
2 or more groups

with any proportion

49
Q

chocharn Q

A

nominal /categorical
dependent
> 2 groups

50
Q

McNemar Chi Square

A

nominal /categorical
dependent
2 groups

51
Q

shapero wilks

A

tests for normalicy in data

52
Q

prevelance

A

how many people have something

53
Q

incidence

A

how many people got something that were at risk for it ( didn’t have it before)

54
Q

will odd ratio over estimate or under estimate RR

A

over estimate
odd of exposure in cases/odds of exposure in control

A/C odds of those with disease/
B/D odds of those without disease

Odd ratio of 1 - same risk
Odd ratio <1- protective
odd ratio >1- increases risk

55
Q

relative risk (incidence)

A

incidence of diease in those exposed A/A+B
divide by
incidence of diease in those not exposed C/ C+D

RR 1- no association
RR 2- double risk
RR .5- half the risk

56
Q

Number needed to treat

A

1/attributable relative risk - Number needed to treat

57
Q

Positive likelihood ratio

A

sensitivity/1-specificity

True positive/false positive

58
Q

negative likelihood ratio

A

1- sensitivity/specificity

false negative /true negative

59
Q

STOBE guidelines are used for

A

cross sectional

case control

60
Q

CONSORT is used for what studies

A

RCT

61
Q

PRISMA is used for what studies

A

metaanalysis

62
Q

important stat for metaanalysis

A

measures of consistance

63
Q

power is

A

1-beta

64
Q

population attributable risk

A

Incidence in exposed- incidence in unexposed

65
Q

statistical test for association between 2 continous variables

A

linear regression

66
Q

statistical test for association between 2 categorical variables

A

pearson’s correlation coefficient

67
Q

statistical test for association between 2 non-parametric variables

A

spearman’s rank correlation

68
Q 
models for analysis 
continous
binary (categorical) 
count
survival time
A

continous- linear regression
binary (categorical) - logistic regression
count- poisson regression
survival time- cox proportional hazards regression

MFM Oral Boards 2020 (25 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions