Biostatistics Flashcards by pardis hejazi

Health promotion prevention category

Primary prevention

Enables people to improve their health

How well did you know this?

Not at all

Perfectly

Health risk assessment method

Questionnaires collecting demographic, medical, lifestyle, family history information to calculate a patient’s risk age.

How well did you know this?

Not at all

Perfectly

Percent of data between +/- 1 SD

68%

How well did you know this?

Not at all

Perfectly

Percent of data between +/-2 SD

95%

How well did you know this?

Not at all

Perfectly

Percent of data between +/- 3 SD

99.7%

How well did you know this?

Not at all

Perfectly

The parameter of diagnostic tests most important for screening

High sensitivity

How well did you know this?

Not at all

Perfectly

Decreasing the cutoff point will result in:

Sensitivity: increased
True positive: increased
False positive: larger increase
PPV: decreased
False negative: decreased

How well did you know this?

Not at all

Perfectly

Hawthorne effect (observer effect)

The tendency of study subjects to change their behavior as a result of their awareness that they are being studied

Affects validity

How well did you know this?

Not at all

Perfectly

Berkson’s bias

Selection bias created by choosing hospitalized pts as the control group

How well did you know this?

Not at all

Perfectly

Pygmalion effect

A researcher’s beliefs in the efficacy of treatment can potentially affect the outcome

How well did you know this?

Not at all

Perfectly

Chi-square test for independence

To test the association between two categorical variables

How well did you know this?

Not at all

Perfectly

Two-sample z test

Two/sample t test

Used to compare 2 group means(numerical values), not categorical variables.

How well did you know this?

Not at all

Perfectly

ANOVA

Used to compare the means of 2 or more groups

How well did you know this?

Not at all

Perfectly

Test parameter influenced by disease prevalence

PPV

NPV

How well did you know this?

Not at all

Perfectly

Test parameter influenced by pretest probability

NPV

How well did you know this?

Not at all

Perfectly

Validity of a test

The test measures what it is supposed to measure

How well did you know this?

Not at all

Perfectly

Positive likelihood ratio

the probability of a person who has the disease testing positive divided by the probability of a person who does not have the disease testing positive.

=sens/ (1-spec)

How well did you know this?

Not at all

Perfectly

Negative likelihood ratio

the probability of a person who has the disease testing negative divided by the probability of a person who does not have the disease testing negative

=(1-sens)/ spec

How well did you know this?

Not at all

Perfectly

In a smoker, diabetic, alcoholic, sedentary, hypertensive individual, what is the intervention most likely to reduce overall mortality rate?

Smoking cessation

How well did you know this?

Not at all

Perfectly

Negative skewed distribution curve

Peak (mode) to the right
Mean to the left
Median in between

How well did you know this?

Not at all

Perfectly

Positively skewed curve

Mode to the left
Mean to the right
Median in between

NNT calculation

1/ARR

ARR

Control rate - treatment rate

RRR

ARR/Control rate

Relative risk

Treatment rate/control rate Or Risk of outcome in exposed / risk of outcome in non-exposed

Type II error (beta) definition

The probability of concluding there is no difference when such a difference truly exists

Power of study

1- beta The ability of a study to detect a difference when such a difference truly exists

Type one error (Alpha)

The probability of seeing a difference when there is no difference in reality

Definition of a reliable test (precision)

It gives similar or very close results on repeat measurements

Validity or accuracy

A test’s ability to measure what it is supposed to measure

How to quantify the reliability of a test

In terms of the coefficient of variation ( standard deviation divided by mean of repeated measurements) generally expressed as a percentage

The odds ratio measurement

Is a measure of association between an exposure and an outcome All in favor of association/all against an association =ad/bc

The significance of confidence of interval in a meta-analysis study

If the range of confidence of interval does not include a null value, the result is statistically significant. If the CI crosses the null value then there is no statistically significant difference between the groups.

Matching Method

Used in design stage of case-control studies to control confounding

Stratified analysis

Can be used to negate confounding effect

Confounding effect vs effect modification in stratified analysis

Stratified analysis disappears the observed difference in the study due to “confounding effect” but potentiates the observed difference due to “effect modification”

Case fatality rate measurement

Calculated by dividing the number of fatal cases by the total number of people with the disease

Atributable risk percentage

100 x (Risk in exposed- Risk in unexposed) / risk in exposed

Attrition bias

In prospective studies if loss to follow-up occurs disproportionately between the exposed and unexposed groups attrition bias can result if the lost subjects differ in the risk of developing the outcome compared to the remaining subjects. It is a form of selection bias.

Latency definition

The concepts of latency. It can be used for infectious diseases, risk factors and risk reducers. The time between exposure and clinically apparent disease.

The rare disease assumption

Refers to the fact that the odds ratio approximates RR When disease prevalence is low

The relative risk null value

1 For a result to be considered statistically significant, its corresponding CI must not contain the null value

The relationship between CI and P-value

A statistically significant 95% CI corresponds to a P-value <0.05. A statistically significant 99% CI corresponds to a P-value <0.01.

Mean +/- 1.96 SD would cover

95% of the observations

The mean +/- 2.58 SD would cover

99% of the observations

SE= standard error

SD/√n A way to account for variability due to sampling 95% CI= mean +/- 1.96 SE 99% CI= mean +/- 2.58 SE

Relative risk, relative rate are for which type of study?

Cohort Because these are incidence measures, they cannot be calculated from case-control studies

Exposure odds ratio is for which type of study?

Case-control

Prevalence odds ratio is for which type of study?

Cross-sectional

Generalizability or external validity

The applicability of the obtained results beyond the cohort that Was studied