Free lesson · Epidemiology
Free · No sign-upStudy designs and bias, decoded
Free for everyone, with no account. Covers: Choosing between cross-sectional, case-control, cohort, and randomized designs · Relative risk, odds ratio, ARR, RRR, NNT, and NNH · Incidence vs prevalence · Lead-time, recall, selection bias, and confounding · Levels of prevention.
Study designs, risk, and bias
Pick the right study design, calculate relative risk, odds ratios, and the number needed to treat, and name the bias hiding in a question stem. Watch the video (it pauses to quiz you), review the flashcards, then run the drill.
Study designs, measures of association, and bias
Cross-sectional, case-control, cohort, and randomized designs; relative risk, odds ratio, ARR, RRR, and NNT; incidence vs prevalence; the biases boards love; and levels of prevention.
Preview lesson with automated narration. A version recorded by Dr. Dave will replace it.
Lesson flashcards
18 cards
Practice questions
12 questions
A cohort of 1,000 smokers and 1,000 nonsmokers is followed for 10 years. 60 smokers and 20 nonsmokers develop bladder cancer.
Question 1 of 12
What is the relative risk of bladder cancer for smokers compared with nonsmokers?
Make it stick
Epidemiology mnemonics
Each one is also in the flashcard deck above (tagged Mnemonic), so spaced repetition brings it back right before you would forget it.
Case-control → Odds · Cohort → Risk
Case-control studies report odds ratios; cohort studies report relative risk.
The bathtub
Incidence flows in, prevalence is the water in the tub, cure and death drain it. Prevalence ≈ incidence × duration.
NNT = 1 / ARR
Convert percents to decimals first: ARR 4% → 1 / 0.04 = 25.
P-S-T
Prevent (primary), Screen (secondary), Treat complications (tertiary).
Lead time looks like longer life
Earlier diagnosis, same date of death: compare mortality, not survival from diagnosis.
Quick reference
Formula sheet
| Measure | Formula |
|---|---|
| Relative risk | [a / (a + b)] ÷ [c / (c + d)] |
| Odds ratio | ad / bc |
| Attributable risk | risk in exposed − risk in unexposed |
| Absolute risk reduction | control event rate − treatment event rate |
| Relative risk reduction | ARR ÷ control event rate |
| Number needed to treat | 1 / ARR |
| Number needed to harm | 1 / absolute risk increase |
| Prevalence (steady state) | ≈ incidence × average duration |
Keep going
More free board resources
Want a tutor for the rest of the exam?
Board Exam Tutoring pairs you with a tutor matched to your exam, test date, and goals for one-on-one USMLE, COMLEX, and shelf prep. These lessons stay free either way.