BIOSTATISTICS Flashcards

0
Q

What’s lead time bias?

A

Is the erroneous finding of prolongation of survival when just the illness was detected earlier .

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

What’s latent period factor in the effect of drugs or change in the risk?

A

Period needed to show differences

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

What is Observer bias ?

A

Is when the observer misclassifies data according its interpretation or its expectations

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

What’s selection bias ?

A

Inappropriate non-random selection

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

When use odds ratio?

A

CASE-CONTROL STUDIES !!!
RETROSPECTIVE!!!!

OR=ad/bc

“Patients with COPD had higher odds of a
history of smoking than those without COPD.”

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

When use Relative Risk?

A

COHORT STUDIES!!!!
PROSPECTIVE!!!!!

RR=[a/(a+b)] / [c/(c+d)]
“Smokers had a higher risk of developing COPD
than nonsmokers.”

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

What’s selection bias ?

A

Selection of unrepresentative sample ( all healthy ,all sick)

ATTRITION BIAS : a very different group of data is loss

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

What’s missclassification bias?

A

Use the wrong method to classify the groups

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

Big difference between case-control and cross-sectional study?

A

CASE -CONTROL : Make sick and non-sick subjects

CROSS-SECTIONAL: take random subjects

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

Big characteristic of randomized clinical trial?

A

COMPARES >2 TREATMENTS

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

What’s type II(beta) error?

A

Falsely conclude there is NO difference

FALSELY NOT REJECT

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

What’s type I (alpha) error?

A

Falsely conclude there is a difference

Falsely REJECT

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

The best answer to a college asking about a patient of your is …

A

No, I cannot say whether she is my patient .

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

How confounding bias can be controlled?

A

MATCHING
used in case control studies
Matching the subjects by age , race , gender etc…

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

What is statistical power?

A

1-beta

SAMPLE SIZE

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

Cause of mortality in TCAs overdoses?

A

Cardiac arrhythmia say inhibit fast Na channels

Refractory hypotension by alpha agonist

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

What is the center of study of ECOLOGICAL STUDY?

A

COMPARE POPULATIONS!!!!!!!

used to make HYPOTHESIS NOT CONCLUSIONS!!!!

17
Q

What’s nested case-control?

A

Begins whit cohort studies

The patients with the outcome go to case - control studies

18
Q

How the protest probability affects the NEGATIVE PREDICTIVE VALUE?

A

⬇️Pre-test probability ⬆️ NPV

⬆️Pre-test probability ⬇️ NPV

19
Q

What is near miss?

A

Error recognized before any harm

20
Q

What’s sentinels event?

A

Unexpected event involving death or serious physical or psychological injury that’s REQUIERS IMMEDIATE INVESTIGATION

21
Q

Which calculation can be used to comparete different tests whit different normal values?

A

Coefficient variance

22
Q

What’s a reliable test?

A

A test whit very similar results in the same patient is REPRODUCIBLE

23
Q

What’s a valid test ?

A

Is the ability to measure what is suppose to measure

Very similar whit GOLD STANDARD

24
Q

What’s the attributable risk percent in the exposed?

A

Percentage of disease attributable to the Exposition

=100x[(RR-1)/RR]

25
Q

What’s referral bias ?

A

You cannot compared a big specialized hospital whit a small health center even if there is the same illness

26
Q

What’s detection bias?

A

The risk factor increase the probability of diagnosis because they perform a lot of tests

27
Q

What’s sampling bias?

A

Very sick people cannot be compared whit sick people not as Ill

28
Q

How calculate the probability of a group of outcomes?

A

MULTIPLY THE INDEPENDENT EVENTS

29
Q

What is the confidence interval formula?

A

Mean➕➖Z(S/√N)

Z 95%=2. 99%=2.5
S standard deviation
N population

30
Q

What’s late-Look bias?

A

Miss severe cases because the patients dye before the study

31
Q

What happen Whit confidence intervals which contain de number 1?

A

IS NOT STATISTICALLY SIGNIFICAT!!!!!

32
Q

Alpha error is possible ONLY if the p value is….

A

p<0.05!!!!!

33
Q

When do you use ANOVA, Chi-square test, Meta-analysis , multiplier linear regression, Pearson correlation coefficient, t-test?

A

ANOVA compare means between 2 or more independent groups

Chi-square test association between 2 categorical variables

Meta-analysis compare a lot of good studies

multiplier linear regression linear relation btw 1 dependent & 2 inde

Pearson correlation coefficient Strength & dirección of linear relation between 2 variables

t-test. Two groups compared

34
Q

What is the equation for number need to treat?

Is equal to number needed to harm NNH…

A

NNT=1/ARR

Absolute risk reduction = ARR= control %-treatment %

35
Q

What is the Hawthorne and Pygmalion effect?

A

Hawthorne : observer effect

Pygmalion: all of mine is better

36
Q

Explain which measure two-sample t test, Two-sample z test, Chi-square, Analysis of variance.

A

two-sample t test 2 groups means
Two-sample z test 2 groups means
Chi-square. 2 categorical variables
Analysis of variance. Means of >2 groups

37
Q

How VPP and VPN are affected by prevalence?

A

⬆️prevalence ⬆️VPP

⬆️prevalence ⬇️VPN

38
Q

What’s effect modification?

A

When the outcome is modified by another variable

Ex. Thrombosis in women who take estrogens Vs Thrombosis In SMOKER women who take estrogens

39
Q

Types of distribution?(3)

A

NORMAL DISTRIBUTION 🔼 mean=median=mode
POSITIVE SKEWED 📉 mean>median>mode
NEGATIVE SKEWED📈 mean<median<mode