Biostatistics

Question 1

Cross-Sectional Study

Answer 1

Snapshot — measures prevalence, not causality

Question 2

Case-Control Study

Answer 2

Start with disease → look back at exposure.
Measures Odds Ratio

Question 3

Cohort study

Answer 3

Start with exposure → follow for disease.
Measures Relative Risk

Question 4

Randomized Controlled Trial (RCT)

Answer 4

Best for determining causality

Question 5

Meta-Analysis

Answer 5

Pools data → ↑ power, ↑ generalizability

Question 6

Double-Blind Study

Answer 6

Prevents observer and subject bias

Question 7

Crossover Study

Answer 7

Each subject serves as own control.

Question 8

Ecological Study

Answer 8

Data on populations, not individuals

Question 9

Prevalence

Answer 9

All cases / total population.
Tip: Prevalence = Incidence × Duration

Question 10

Relative Risk (RR)

Answer 10

Cohort studies → [a/(a+b)] / [c/(c+d)]

Question 11

Odds Ratio (OR)

Answer 11

Case-control → (a/c) / (b/d) = ad/bc

Question 12

Attributable Risk (AR)

Answer 12

Risk difference: [a/(a+b)] - [c/(c+d)]

Question 13

Relative Risk Reduction (RRR)

Answer 13

(1 - RR)

Question 14

Absolute Risk Reduction (ARR)

Answer 14

Control rate − Treatment rate

Question 15

Number Needed to Treat (NNT)

Answer 15

1 / ARR

Question 16

Number Needed to Harm (NNH)

Answer 16

1 / Attributable Risk

Question 17

Sensitivity

Answer 17

“True Positives”
Rule out disease – SNOut

Question 18

Specificity

Answer 18

“True Negatives”
Rule in disease – SPIn

Question 19

Positive Predictive Value (PPV

Answer 19

↑ with ↑ prevalence

Question 20

Negative Predictive Value (NPV)

Answer 20

↑ with ↓ prevalence

Question 21

False Positive Rate

Answer 21

1 - Specificity

Question 22

False Negative Rate

Answer 22

1 - Sensitivity

Question 23

Lead-Time Bias

Answer 23

Early detection ≠ longer survival

Question 24

Length-Time Bias

Answer 24

Screening more likely detects slow-growing diseases

Question 25

Selection Bias

Answer 25

Sample isn't representative (e.g., volunteer bias

Question 26

Berkson Bias

Answer 26

Hospital patients ≠ general population

Question 27

Recall Bias

Answer 27

Common in case-control; patients misremember exposure

Question 28

Observer Bias

Answer 28

Researcher expectations influence outcome

Question 29

Confounding Bias

Answer 29

Third variable distorts true relationship

Question 30

Effect Modification

Answer 30

A third variable changes the strength of the association

Question 31

p-value < 0.05

Answer 31

Statistically significant — reject H₀

Question 32

Confidence Interval (CI)

Answer 32

If 95% CI for RR or OR crosses 1 → not significant

Question 33

Type I Error (α)

Answer 33

False Positive — reject H₀ when it's true

Question 34

Type II Error (β)

Answer 34

False Negative — fail to reject H₀ when it's false

Question 35

Power = 1 - β

Answer 35

Probability of finding a true effect.

Question 36

Standard Error (SE)

Answer 36

SE = SD / √n — ↓ with ↑ sample size

Question 37

Standard Deviation (SD)

Answer 37

Measures spread — 68-95-99.7 rule

Question 38

Z-score

Answer 38

of SDs from the mean

Question 39

Correlation (r)

Answer 39

Strength of linear relationship (r² = % of variation explained)