Biostatistics Naplex 2025

Define biostatistics

When statistics are used to understand the effects of a drug or medical procedure on people and animals

What are the steps to journal publication? (FYI)

1. Begin with a research question

2. Design study

3. Enroll subjects

4. Collect the Data

5. Analyze the data

6. Publish

What are the two types of data?

Continuous and Discrete/Categorical

Define Continuous data

Data is provided by some type of measure which has unlimited options (theoretically) of continuous values

Define Discrete/Categorical Data

Data fits into a limited number of categories

What are the 2 types of continuous data?

Ratio and Interval

What are the 2 types of discrete data?

nominal and ordinal

Define ratio data

A type of continuous data where 0 means none

Define interval data

A type of continuous data where 0 does not mean none

Define nominal data

A type of categorical data where the order of the categories does not matter

Define ordinal data

A type of categorical data where the order of the categories does matter

What are examples of ratio data?

age, height, weight, time, blood pressure, HR (0 HR is cardiac arrest)

What are examples of interval data?

temperature scales (F and C)

Ex. 0 degrees is a temperature

What are examples of nominal data?

gender, ethnicity, marital status, mortality

What are examples of ordinal data?

NYHA Functional Class I-IV

0-10 pain scale

What is mean?

Average

Add up sum and divide by total

What is median?

Middle number when arranged from lowest to highest

What is mode?

MOst frequently occuring number

What is range?

the difference between the highest and lowest scores in a distribution

What is standard deviation?

Indicates how spread out the data is

What is the normal distribution?

Gaussian/bell curve

What % falls within one standard deviation?

68%

What % falls within two standard deviations?

95%

What are outliers?

Extreme values

What are independent variables?

the factors being manipulated or changed in the experiment by the experimenter

What are dependent variables?

the factors being measured, the results we are examining affected by the independent variables

What is needed to show significance?

Show the null hypothesis is NOT true and should be rejected

Define null hypothesis

NO statistical significant difference between groups

- we are trying to disprove this or reject it

Define alternative hypothesis

Significant difference between the groups

- we are trying to prove this

What is the alpha level?

Maximum permissable error margin

- threshold for rejecting null hypothesis

What is alpha normally set at?

0.05 or 5%

What is p value?

P = Probability

Probability that the result obtained was due to chance. Generally a P-value < 0.05 indicated statistical significance.

- If P < 0.05 that means there is a < 5% probability that the result occurred by chance.

If p value is higher than alpha value, what does this mean?

Accept null hypothesis (we dont want to do this)

If p value is lower than alpha value, what does this mean?

Reject null hypothesis (significant)

Define confidence interval

the interval in which you are confident that the mean of the entire population is within a certain range (e.g. 95%)

- significance and precision

CI equation

CI = 1 - alpha

How can we tell if something is statistically significant from confidence interval range?

If it DOES NOT cross zero

How can we tell if something is statistically significant from confidence interval of relative risk?

If it DOES NOT cross zero

What are type 1 errors?

False positives ("saying the pt has cancer when he actually doesn't")

- alternative hypothesis accepted when it is wrong and null was rejected in error

What are type 2 errors? (Beta)

False negative ("saying the pt doesn't have cancer when actually does")

- accepting a false null hypothesis

What is study power?

Probability that a test will reject the null hypothesis correctly. Used to avoid type 2 error.

Power formula

Power = 1 - beta

What is power determined by?

- outcome values

- difference in outcome rates between groups

- alpha level (significance)

If beta is 0.1, what is power?

90% power (larger sample size is needed to increase power)

Define risk

Probability of an event when an intervention such as a drug is given

Define relative risk

Ratio of risk in the exposed group (treatment) divided by the risk in the control group

- AS likely

Risk formula

R = Number of subjects in a group with an unfavorable event / total number of subjects in the group

Relative risk formula

RR = risk in treatment group / risk in control group

What does RR = 1 indicate?

NO DIFFERENCE in risk of outcome between groups

What does RR > 1 indicate?

GREATER risk in the treatment group

What does RR < 1 indicate?

LOWER risk in treatment group

Define relative risk reduction

Indicates how much the risk is reduced in the treatment group compared to the control group

- Less likely

Relative risk reduction formula

RRR = (% risk in the control group - % risk in the treatment group) / (% risk in the control group)

OR

RRR = 1 - RR

RR + RRR = ?

100%

What is absolute risk reduction?

Includes the reduction of risk AND the incidence rate of the outcome

- If ARR is not given, it could mean that the RRR is minimal but we need to compare it to general population

What is an example of absolute risk reduction?

If the risk of nausea is reduced but the risk was small to begin with, the large risk reduction has no real benefit

Absolute risk reduction formula

ARR = (% risk in control group) - (% risk in treatment group)

How is ARR interpreted?

For ARR x: for every 100 patients treated with the intervention, x fewer patients would experience the negative event

What is number needed to treat?

Number of patients who need to be treated for a certain period of time in order for ONE patient to benefit

1/ARR

What is NNT formula?

1/ (risk in control group - risk in treatment group)

OR

1/ARR

How is NNT interpreted?

If NNT is 9, for every 9 patients who receive a drug for a year, disease progression is prevented in one patient

What is number needed to harm?

Number of patients who need to be treated for a certain period of time in order for one patient to experience harm

How is NNT rounded?

UP

How is NNH rounded?

DOWN

NNH formula

NNH = 1/ARR (same as NNT)

How is NNH interpreted?

If NNH is 90, one additional event is expected to occur for every 90 patients taking drug instead of placebo

What is odds ratio?

the probability that an event will occur, versus the probability that it will not occur

Example of odds ratio

Ex: Estimates the risk of unfavorable events associated with treatment or with intervention

Odds ratio formula

OR = AD/BC

Odds ratio formula meaning

A= # that will have outcome with exposure

B = # without the outcome with exposure

C = # with outcome without exposure

D = # without outcome without exposure

What is hazard ratio?

The rate at which an unfavorable event occurs within a short period of time

Hazard ratio formula

HR = (Hazard rate in treatment group) / (Hazard rate in control group)

OR or HR =1 means

The event rate is the SAME in treatment and control, no advantage to treatment

OR or HR >1 means

The event rate in the TREATMENT group is HIGHER than control, do not use treatment

OR or HR < 1 means

The event rate in the TREATMENT group is LOWER than the control group, use treatment group

Define primary endpoint

Main result that is measured to see if the treatment had a significant benefit

Define composite endpoint

Combines multiple, individual endpoints into one measurement. Must use the composite endpoint value when assessing the composite endpoint because just adding the individual endpoints will NOT equal the same thing

When are T tests used?

When an endpoint has continuous data and is normally distributed

What is analysis of variance (ANOVA)?

F test, tests for statistical significance when using continuous data with 3 or more samples/groups

What tests are used with continuous data?

For parametric measurements

- t tests

- f tests

What tests are used with continuous data for NONparametric measurements?

- sign test

- wilcoxon signed rank test

- mann whitney

- kruskal wallis test

What tests are used for discrete/categorical data?

- Chi square

- Wilcoxon square test

- Fishers exact test

- kruskal wallis test

Describe Chi square tests

Used to determine statistical significance between treatment groups

When is a one sample t test used?

1 group

When is a dependent/paired t test used?

1 group with before and after measures

When is an independent/unpaired student t test used?

2 groups (treatment and control)

When is an ANOVA test / f test used?

3 or more groups

When is a sign test used?

1 group

Nonparametric data

When is a wilcoxon signed rank test used?

1 group (before and after measures)

Nonparametric and Discrete

When is a mann whitney test used?

2 groups (Treatment and control)

Nonparametric

When is a kruskal wallis test used?

3 or more groups

Nonparametric or discrete

Define correlation

A statistical technique that is used to determine if one variable changes, or is related to another variable

What is positive correlation?

as one variable increases, so does the other

What is negative correlation?

one variable decreases, the other does too

Define regression

Used to describe the relationship between a dependent variable and one or more independent variable, or how much the value of the dependent variable changes when the independent variables change

What are the 3 types of regressions?

1. Linear (continuous)

2. Logistic (Categorical)

3. Cox (categorical in survival analysis)

What is sensitivity?

The true positive

Describes how efficiently a test identifies patients with the condition

What is specificity?

The true negative

Describes how effectively a test identifies patients without the condition

Sensitivity formula

Sensitivity = A/ (A+C) x 100

Specificity formula

Specificity = D/ (B+D) x 100