Epidemiology 7/8 Flashcards

1
Q

what does the P value mean

A

probability of getting a result as extreme or more extreme if the null hypothesis is true (i.e.. no difference)

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

when do we reject null hypothesis

A

P less than or equal to 0.05

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

T/F: P values measure how likely study hypothesis is true

A

FALSE

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

What is a confidence interval

A

a range of values that could be plausibly generated by the sample

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

When you want to be more “confident” what occurs with the confidence interval?

A

It gets wider so you can capture the true value more often

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

define sensitivity

A

how well does a test correctly identify those with a disease (TP/all with disease)

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

specificity

A

how well a test correctly identifies those WITHOUT a disease (TN/all without disease)

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

PPV

A

of those that test positive, what proportion actually have disease

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

NPV

A

of those who test negative, what proportion actually don’t have the disease

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

What happens to sensitivity/specificity/PPV/NPV when prevalence changes

A

Sensitivity/specificity don’t change

  • PPV increases with increasing prevalence, decreases with decreasing prevalence
  • NPV decreases with increasing prevalence and increases with decreasing prevalence
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Receiver Operating characteristic

A

describes intrinsic accuracy of a test considering various threshold changes

  • larger area under curve indicates greater discriminating ability of test
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

number needed to treat/screen (NNT/NNS)

A

what is the impact of the treatment

- number of ppl you need to treat/screen to prevent 1 bad outcome

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

What does a higher NNT mean

A

increased number needed to treat = less effective test

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

Case-control study

A

select ppl with/without disease or outcome. Compared with controls regarding past exposures

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

Cohort study

A

select ppl with exposure/not but are initially free of the outcome. Followed and assess incidence of outcome in both exposure groups

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

ODDS Ration

A

Odds of disease in exposed/odds of dz in non-exposed

ad/bc

17
Q

main difference between cohort and case-control study

A

start with exposure in cohort and outcome in case-control