Stats Flashcards

1
Q

How is SEM calculated?

A

Standard Deviation /
square root of ‘n’

n = sample size

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

Calculate relative risk reduction

100 working women and 110 women who are house wives are followed up through pregnancy. 40 working women had LBW babies and 21 house wives had LBW babies.

A

Draw table
Top: Behaviour: working/housewife
Left side: Outcome: LBW Y/N

(40/100) / (21/110) = 2.1

2.1

CONTRAST odds ratio of LBW
(40/21) / (60/89) = 2.8

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

What non-parametric tests can be used to test the induction of labour interval between normal and pre eclamptic women? Name 3

A

Non parametric is for Names or Not normally distributed

Mann Whitney
Wilcox Rank
Spearman

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

How is the perinatal mortality rate defined?

A

The number of stillbirths and the number of neonatal deaths within 7 days of life, per 1000 births

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

ABO blood grouping is?
Parametric/nonparametric
nominal/ordinal

A

Non-parametric nominal

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

What is the risk of uterine rupture during a VBAC after 1 CS?

A

0.5%
or
1 in 200

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

The diamond shape Forest graph, lateral angles represents:

A

95% CI

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

What is the failure rate of vasectomy

A

0.5/1000

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

Null hypothesis is ACCEPTED if
i.e. say the variable is without effect

A

p > 0.05

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

Statistical analysis used for data with normal distribution. Name 3

A

Student t
Pearson (P for parametric)
ANOVA

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

What is the data type of
‘moderate / mild / severe’

A

Ordinal

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

What is an alpha or type 1 error
(with regards to the null hypothesis)

A

False positive
Incorrectly rejecting the null hypothesis

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

What gestation is defined as stillbirth rather than miscarriage?

A

24 weeks

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

Statistical heterogenicity means

A

Difference between studies

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

Calculate maternal mortality ratio

Total number of live births 250,000, early neonatal death 1500, Maternal death 750.

A

Maternal mortality ratio =
(Maternal deaths /live births) X 100,000

(750/250,000) X 100,000 =
0.003 X 100,000 = 300

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

Calculate SEM
For a group of 10,000
If the Mean 245 and SD 50.

A

SEM = SD/√n = 50/√10,000 = 50/100 = 0.5

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

Which study design would describe the relationship between shoulder dystocia and neonatal birth weight ?

A

retrospective case-controlled study

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

Calculation for negative predictive value

A

TN / TN + FN

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

What is the mode?

A

The most frequently occurring number (fashionable!)

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

Grade B recommendation of RCOG means

A

well controlled studies but non randomized

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

Incidence of “RARE” disease when translated into mathematics is?

A

1:1000 or rarer

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

Calculate BMI
150cm height and 90 kgs weight

A

BMI = weight / height in metres squared
90kg / 1.5 x 1.5
90/ 2.25 = 40

<18.5 underweight range.
<25, healthy weight range.
<30, overweight range.
Class 1 Obesity <35
Class 2 Obesity <40
*Class 3 Obesity 40 or higher

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

Calculate perinatal mortality

A

stillbirths (>=24) and early neonatal (<7/7) deaths per thousand

(BUT question on paper from 2018 and WHO define perinatal period as from 22 weeks?)

24
Q

What represent the 2 axis of ROC curve ?

A

Sensitivity (vertical) over
1-specificity (horizontal)

25
Q

What do the bottom and top represent on the box-and-whisker plot?

A

Lower and upper quartile

26
Q

Odds ratio calculation

Del <34/40 (total 209)
Smokers 31
Non-smokers 178

Del >34/40 total 926
Smokers 58
Non-smokers 659

A

Behaviour: Smoking
Outcome/risk: Del at <34/40/>34/40

(31/178) / (58 / 659)
17.42/8.8 = 1.98

So odds doesn’t ‘use’ the total it’s e.g.
1 in 9 (out of 10 total)

27
Q

NNT calculation

A

1 / ARR

Absolute risk reduction
ARR is % risk decrease from one group to another
e.g. drug reduces chance of bad outcome from 50% to 40% then ARR = 10% or 0.1

1 / 0.1 = 10

28
Q

Pregnancy rate with COCP if perfectly compliant and non perfectly compliant

A

0.1 / 5-6

29
Q

VTE with COCP

A

6 / 10,000

30
Q

IUD Pregnancy. What is the risk of Ectopic Pregnancy

A

1/20

31
Q

The Probability of sarcomatous change occurring in fibroid is

A

0.1%

32
Q

Grade A RCOG recommendation means

A

At least one randomised controlled trial

33
Q

What type of variable is BMI ?
ordinal / nominal / continuous / categorical / discrete

A

Continuous

34
Q

Incidence of uterine rupture in IUCD insertion

A

1/5000

35
Q

What is represented by the equation Sensitivity/(1-specificity)?

A

Positive likelihood ratio

36
Q

What is the correct formula to calculate specificity?

A

TN / TN + FP

37
Q

Calculate PPV

A

TP / TP+FP

38
Q

Calculate sensitivity

A

TP / TP + FN

39
Q

A study between soya milk and improvement of menopause symptom. the null hypothesis H0 should state that…

A

There is no relation/difference between soya milk intake and menopause symptoms

40
Q

WHO defines perinatal mortality as

A

number of still births after 22 wks & early neonatal deaths per 1000 births

41
Q

What kind of variable is ‘gestational age’

A

interval
named, ordered, with proportionate intervals between variables

42
Q

How do you calculate the type 2 error?

A

1 - power
e.g. power is 90%
1 - 0.9 = 0.1 or 10%

43
Q

Risk of expulsion of IUD

A

~5% (google)

44
Q

Risk of ectopic pregnancy with IUD

A

<1 per 1000 years of use

45
Q

Risk of malignant transformation of leiomyoma into leiomyosarco

A

0.01%

46
Q

When a pregnant person is on carbamazepine and lamotrigine, what percent will have foetal congenital anomalies

A

> 10%

47
Q

What is the leading cause of maternal death according to MBBRACE

Overall

direct
indirect

A

Overall - Cardiac

Thrombosis and thromboembolism is the leading direct cause of death during or up to six weeks after end of pregnancy.
You clot because you are pregnant.

Cardiac disease leading indirect cause of death
You have cardiovascular disease before you are pregnant.

48
Q

What is the recurrence rate of ectopic pregnancy with a history of salpingectomy

A

15 - 20%

49
Q

What % of people with GDM go on to have T2DM

A

30-40% or more

50
Q

Calculate the interquartile range
Number top 25%
minus
Number bottom 25%
2,3,4,5,6,6,6,7,7,8

A

7 - 4 = 3

51
Q

Calculation for the false negative rate

A

FN/FN+TP

Or if sensitivity is 90% and specificity is 90%
FNR is 10%

52
Q

Calculate the absolute risk reduction
Placebo 50% risk
New drug 40% risk

A

ARR absolute risk reduction of 10%
or 0.1

53
Q

What is the % of BP injury in cases of shoulder dystocia

A

16%

54
Q

Rhesus negative incidence in uk

A

16%

55
Q

When does the Odds ratio approximate risk ratio?

A

When the disease studied is rare

56
Q

What kind of data does a histogram present

A

Continuous

57
Q

What is the % risk of recurrent preeclampsia?

A

20%