Brainscape's Knowledge GenomeTM

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


See full index

Evidence Based Medicine > ECBM Lecture 1--stats review > Flashcards

ECBM Lecture 1--stats review Flashcards

1
Q

scientific method

A

prove or disprove hypothesis

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

precision

A

Target: clustered in same spot

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

accuracy

A

Target: hit bullseye

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

necessary characteristics of research

A 
Objectivity
precision 
verification
economic 
reasonable
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Research: Observational

A

Exploratory –> often when thing under observation would be unethical

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

RCT

A

randomized control trial

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

Comparitive analysis

A

research on drug A vs placibo
research on drug B vs placibo
comparison

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

Strongest evidence in clinical research?

A

Randomized control trial RCT

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

Observational study divisions

A

Cohort– groups with/ without exposure –>outcome

Case Control– grps exposure with/without disease–> outcome

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

N

A

population–compare pt to population

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

n

A

sample–characterized by statistics

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

n statistics

A

draw conclusions and broadly apply

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

variable being manipulated (the intervention)

A

Independent variable

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

any variable being measured (the outcome)

A

dependent variable

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

whole units of data Qualitative

A

discrete –

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

data with range

A

continuous

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

ranked data (1st, 2nd, 3rd)

A

ordinal scale

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

ordered data w/out a meaningful zero (water temp)

A

interval scale–0 temp is an actual temp

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

bar graph with no spaces–for trend

A

histogram

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

most frequent #

A

mode

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

value in middle

A

median

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

standard deviation

A

68% of data to either side of mean on line graph = confidence interval

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

confidence interval most often used

A

95% confidence – related to P needing to be less than 0.05

24
Q

trustworthy graph

A

tall and narrow–wide curve not precise

25
Q

when asked “what is the confidence interval?”

A

compare upper and lower limits–not 95%

26
Q

empirical rule

A

when you have a normalized bell curve–two standard deviations from mean will give 95% confidence interval

27
Q

null hypothesis is rejected when true

A

Type I error–ex. assume you have enough $ for groceries, but you don’t

28
Q

the null hypothesis is kept when it is fake

A

Type II error– you assume you don’t have enough $, but you don’t

29
Q

Better error to make in medicine?

A

Type II error–underestimate drugs efficacy

30
Q

Regression

A

predict one factor from existing graph/data

31
Q

how benefitial is drug

A

treatment effect

32
Q

which numbers for 2X2 square?

A

think of scenario

read abstract

33
Q

2X2 – the scenario

A

always deal with PEOPLE–may need to add/subtract groups

34
Q

2X2 – the abstract

A

avoid scores–look for PEOPLE–may be in percentages

35
Q

For 2X2– what is the % of people who something happened to

A

Event Rate (EER)–incidence of some thing–may be in abstract

36
Q

First step

A

calculate
EER (experimental event rate) and
CER (control event rate)

37
Q

second step

A

compare EER and CER (risk ratio, relative risk)

38
Q

see ratio think…

A

divide RR = EER/CER

39
Q

100 minus risk ratio– 100-X= Y

A

relative risk reduction RRR

40
Q

how many people would I need to treat with drugX to save one life

A

number needed to treat NNT

41
Q

NNT

A

100%/ARR (in %)

42
Q

apply NNT math to negative outcomes

A

number needed to harm NNH

43
Q

odds for treatment/ odds for control

A

odds ratio (OR) – Odds vs risk

44
Q

odds

A

“yes” column divided by “no”

45
Q

positive test has diseas

A

true positive TP

46
Q

negative test has disease

A

False negative FN

47
Q

positive test doesn’t have disease

A

False positive FP

48
Q

negative test doesn’t have disease

A

true negative

49
Q

sensitivity=TP/(TP+FN)

FN will impact sensitivity most–only on one side of equation

A

true positive rate (% positive test results in pt who have the disease

50
Q

SnNOUT

A

snesitivity–“SeNsitivity means Negative rules ““it”” OUT”

51
Q

SpPin

A

“Specificity means Positive rules ““it”” in”

52
Q

SpPin

A

low false positivity–

true negative/true negative + false positive

53
Q

SnNout

A

low false negativity=

true positive/true positive + false negative

54
Q

sensitivity and specificity equation key

A

look for the data point on only one side of equation–this factor will play in more to results

55
Q

sensitivity and specificity combined

A

Likelihood Ratios

56
Q

TPR + FNR =

TNR + FPR=

A

100% – always

57
Q

default hypothesis that A and B have no correlation i.e. drug has no effect

A

null hypothesis

Evidence Based Medicine (1 decks)

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