Statistics Flashcards

1
Q

Attributable risk (AR)
Or
Risk difference (RD)

A

[a/(a+b)] - [c/(c+d)]

Draw box
X-axis disease: + or -

Y-axis exposure:

+

or

-

+ + = a; e+ d- = b; e- d+ = c; - - = d

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

Attributable risk %

A

ARx100%/[a/(a+b)]

Draw box

X-axis disease: + or -

Y-axis exposure:

+

or

-

+ + = a; e+ d- = b; e- d+ = c; - - = d

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

False negative

A

1- sensitivity

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

False positive

A

1 - specificity

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

Cumulative incidence

A

new cases in a given time period/

Total population at risk

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

Incidence rate

A

new cases of disease in a given time period/

Person-time years of observation

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

Likelihood ratios (LR)

A

LR+ (positive test) =
Sensitivity/(1-specificity)

LR- (negative test) =
(1- sensitivity)/specificity

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

Mortality rate, BW specific

A

deaths in specific BW categoryx1000/

# births in same BW category

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

Mortality rate, fetal

A

Fetal deaths x 1000/
(Live birth + fetal deaths)

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

Mortality rate, infant

A

Infant death (≤1 yo)x1000/
Live births

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

Mortality rate, neonatal

A

Neonatal deaths <28d x 1000/
Live births

*Can further divide to <7d and >7d but <28d

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

Mortality rate, perinatal

A

(Fetal deaths ≥ 28wk GA + neonatal deaths<7d)x1000/
(Live births+fetal deaths ≥28d GA)

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

Mortality rate, postnatal

A

Postneonatal deaths (28-364d)x1000/
Live births

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

NPV

A

d/(c+d) = TN/(FN+TN)

Draw box

X-axis disease: + or -

Y-axis test+:

+

or

-

+ + = a; e+ d- = b; e- d+ = c; - - = d

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

NNT

A

1/(A+B)

If A = events in the control/non exposed group = c/(c+d)
B = events in the treated/exposed group = a/(a+b)

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

OR aka cross product ratio

A

(axd)/(bxc)

Draw box

X-axis disease: + or -

Y-axis exposure:

+

or

-

+ + = a; e+ d- = b; e- d+ = c; - - = d

17
Q

PPV

A

a/(a+b) = TP/(TP+FP)

Draw box

X-axis disease: + or -

Y-axis test+:

+

or

-

+ + = a; e+ d- = b; e- d+ = c; - - = d

18
Q

Power

A

1- type II error

19
Q

Type I error

A

Reject null hypothesis when it is true

False positive

20
Q

Type II error

A

Do not reject null hypothesis when it is false

False negative

21
Q

Prevalance

A

= #new cases at point in time x 1000/ total population at risk

= (TP + FN)/(TP + FN + FP + TN)

22
Q

Risk ratio or relative risk

A

= AR in exposed/AR in non exposed
= [a/(a+b)]/[c/(c+d)]

Draw box

X-axis disease: + or -

Y-axis exposure:

+

or

-

+ + = a; e+ d- = b; e- d+ = c; - - = d

23
Q

Sensitivity

A

a/(a+c) = TP/(TP+FN)

Draw box

X-axis disease: + or -

Y-axis test+:

+

or

-

+ + = a; e+ d- = b; e- d+ = c; - - = d

24
Q

Specificity

A

d/(b+d) = TN/(FP+TN)

Draw box

X-axis disease: + or -

Y-axis test+:

+

or

-

+ + = a; e+ d- = b; e- d+ = c; - - = d