Biostatistics

Question 1

What are the Types of Study Data?

Answer 1

1- Continuous: Data is provided by some type of measurement which has unlimited options of continuous values.

2- Discrete (Categorical): Data fits into a limited number of categories. It has categories!

Question 2

What are the types of Continuous Data?

Answer 2

1- Ratio :Has meaningful zero (Zero = none) e.g. age, height, weight, time, blood pressure

2- Interval : No meaningful zero (Zero NOT EQUAL none) e.g. temperature scales

Question 4

What are the types of Discrete Data?

Answer 4

1- Nominal : Order does NOT matter e.g. gender, ethnicity

2- Ordinal : Order matters e.g. pain scale

Question 5

What is Range?

Answer 5

This is the difference between the highest and lowest value.

Question 6

Characteristics of a Gaussian (Normal) Distribution?

Answer 6

1- The mean, mode and median are the same value, and are at the center point of the curve.

2- 68% of data falls within 1SD of the mean

3- 95% of data falls within 2SD of the mean

4- 99.7% of the data falls within 3SD of the mean

Question 7

What is the Null Hypothesis?

Answer 7

A null hypothesis (H0) states that there is NO statistically significant difference between groups.

Drug efficacy = Placebo efficacy.

The null hypothesis is what the researcher tries to disprove/reject.

Question 8

What is the Alternative Hypothesis?

Answer 8

The alternative hypothesis states that there is a statistically significant difference between groups.

Question 9

What is the Alpha Value and P-Value?

Answer 9

Alpha is the maximum permissible error margin that is selected by investigators. Alpha is the threshold for rejecting the null hypothesis.

The p-value is compared to the alpha value.

If the alpha is set at 0.05 and the p-value is < 0.05, the null hypothesis is rejected, and the result is statistically significant.

If the p-value >= 0.05, the study has failed to reject the null hypothesis, and the result is not statistically significant.

Question 10

What is the Confidence Interval?

Answer 10

CI = 1 - alpha

(a is 0.05, p-value >= 0.05): Not statistically significant

(a is 0.05, p-value < 0.05): 95% probability (confidence) that the conclusion is correct; less than 5% chance it’s not.

(a is 0.05, p-value < 0.01): 99% probability (confidence) that the conclusion is correct; less than 1% chance it’s not.

(a is 0.05, p-value < 0.001): 99.9% probability (confidence) that the conclusion is correct; less than 0.1% chance it’s not.

A narrow range implies high precision, and a wide range implies poor precision.

Question 11

How can we determine statistical significance with CI?

Answer 11

The result is statistically significant if the CI range does NOT include ZERO.

IF we are comparing Ratio Data (Relative Risk, Odds Ratio, Hazard Ratio), the result is statistically significant if the CI range does NOT include ONE. e.g. The 95% CI for the RR of moderate exacerbations (0.72-0.99) does not include 1, therefore the result is statistically significant.

Question 12

What is a Type 1 Error?

Answer 12

False-Positives

Here the alternative hypothesis was accepted, and the null hypothesis was rejected IN ERROR.

Question 13

What is a Type 2 Error?

Answer 13

False-Negatives

Here the null hypothesis is accepted when it should have been rejected.

The risk of a type 2 error increases if the sample size is too small. To decrease this risk, a power analysis is performed to determine the sample size needed to detect a true difference between groups.

Question 14

What is the Study Power?

Answer 14

Power is the probability that a test will reject the null hypothesis correctly.

Power = 1 - beta

If Beta is set at 0.2, the study has 80% power (there is a 20% chance of missing a true difference and making a type 2 error)

Question 15

Relative Risk (RR)

Answer 15

This is the ratio of risk in the exposed group (treatment) divided by the risk in the control group.

Risk = [# of subjects in group with unfavorable events] / [Total subjects in group]

RR = [Risk in treatment group] / [Risk in Control group]

RR Interpretation:
RR=1 (or 100%): No difference in risk of the outcome between the groups [Intervention had no effect]
RR>1: There is a greater risk of the outcome in the treatment group
RR<1: There is a lower risk (Reduced Risk) of the outcome in the treatment group

Question 16

Relative Risk Reduction (RRR)

Answer 16

RRR is calculated after the RR and indicates how much the risk is reduced in the treatment group compared to the control group.

RRR = 1 - RR
*Must use the decimal version of RR

RRR Interpretation:
If the RRR is 43%, the metoprolol-treated patient were 43% less likely to have HF progression than placebo-treated patients.

Question 17

Absolute Risk Reduction (ARR)

Answer 17

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

ARR Interpretation:
ARR = 12% or 0.12
12 out of every 100 patients benefit from the treatment.

Question 18

Number Needed to Treat (NNT)

Answer 18

NNT is the number of patients who need to be treated for a certain period of time in order for one patient to benefit,

NNT = [1] / [(Risk in Control group) - (Risk in Treatment group)]
OR
NNT = 1 / ARR

NNT Interpretation:
NNT = 9
For every 9 patients being treated with metoprolol for 1 year, HF progression is prevented for 1 patient.

-For NNT, anything greater than a whole number, round up to the nearest whole number. NNT of 52.1 –> 53

Question 18

Number Needed to Harm (NNH)

Answer 18

This is the number of patients who need to be treated for a certain period of time in order for one patient to experience harm.
*Formula is same with NNT. Rounding is different

-For NNH, anything greater than a whole number, round down to the nearest whole number. NNH of 41.9 –> 41

NNH Interpretation:
NNH = 90.9 = 90
One additional case of major bleeding is expected to occur for every 90 patients taking clopidogrel instead of placebo.

Question 19

Odds Ratio (OR)

Answer 19

In case control studies, the OR is used to estimate the risk of unfavorable events associated with a treatment or intervention.

OR = AD / BC

Interpretation:
OR = 1.23
Antidepressants are associated with a 23% increased risk of falls with fracture.

Question 20

Hazard Ratio (HR)

Answer 20

In a survival analysis (e.g. analysis of death or disease progression), instead of using ‘risk’, a ‘hazard rate’ is used. A hazard rate is the rate at which an unfavorable event occurs within a short period of time.
This is similar to RR

HR = [Hazard rate in the treatment group] / [Hazard rate in the control group]

Question 21

HR and OR Interpretation

Answer 21

OR and HR are interpreted in a similar way to RR:

-OR or HR = 1: the event rate is the same in both groups. There is no advantage to the treatment.

-OR or HR >1: the event rate in the treatment group is higher than the event rate in the control group. e.g. a HR of 2 indicates that there are twice as many deaths in the treatment group.

-OR or HR <1: the event rate in the treatment group is lower than the event rate in the control group e.g. a HR of 0.5 indicates that there is half as many deaths in the treatment group.

Question 22

What type of statistical test should be used to analyze Continuous Data?

Answer 22

1- T-tests (use if normally distributed)
2- ANOVA (use if there is 3 or more groups)

Question 23

What type of statistical test should be used to analyze Discrete (Categorical) Data?

Answer 23

1- Chi Square Test

Question 24

Sensitivity and Specificity

Answer 24

Sensitivity [The True Positive] - describes how effectively a test identifies patients with the condition.

Sensitivity =[ A / [A + C] ] x 100

Specificity [The True Negative] - describes how effectively a test identifies patients without the condition.

Specificity =[ D / [B + D] ] x 100

Question 25

What is Intention-to-Treat?

Answer 25

Intention-to-Treat analysis includes data for all patients originally allocated to each treatment group, even if the patient did not complete the trail according to the study protocol. This gives a real-world estimate of the treatment effect.

Question 26

What is Per-Protocol?

Answer 26

Per Protocol analysis is conducted for the subset of the trial population who completed the study according to the protocol.

Question 27

Interpreting a Forest Plot

Answer 27

1- The small boxes show the effect estimate. Diamonds represent pooled results from multiple studies.

2- The horizontal lines through the boxes illustrate the length of the confidence interval for that particular endpoint. The longer the line, the wider the interval, and the less reliable the study results.

3- The vertical solid line is the line of no effect. A significant benefit is reached when data falls to the left of the line; significant harm is indicated when data falls to the right. For ratio data, the vertical line is at 1 instead of 0.

Question 28

Most Reliable –> Least Reliable Medical Studies

Answer 28

1- Systematic Reviews and Meta Analysis
2- Randomized Control Trials
3- Cohort
4- Case-Controlled
5- Case Series and Case Reports
6- Expert Opinion (least reliable)

RCT - prospective comparison

Cohort - retrospective or prospective comparison of patients with an exposure to those without an exposure.

Case-Controlled - retrospective comparison of cases (with disease) and controls (without disease)