A researcher wants to determine whether a new drug prevents heart attacks. She identifies 5,000 healthy adults, randomly assigns half to receive the drug and half to receive placebo, and follows both groups for 10 years. What study design is this, and what is its primary advantage over an observational cohort study?
ACase-control study — it is more efficient because it starts with the outcome
BRandomized controlled trial — random assignment controls for both measured and unmeasured confounders
CProspective cohort study — it follows subjects forward in time
DCross-sectional study — it captures a snapshot of drug use and heart attack status
Random assignment is the defining feature of an RCT and its principal advantage: it ensures that treatment and control groups are, on average, balanced on all characteristics — including unmeasured confounders that could bias observational estimates. A prospective cohort also follows subjects forward in time, but without randomization, any observed association may reflect confounding rather than causation.
Question 2 Multiple Choice
In a case-control study of lung cancer, investigators identify 200 patients with lung cancer (cases) and 200 patients without lung cancer (controls), then look backward to compare their smoking histories. Which measure of association can be directly estimated from this design?
ARelative risk (risk ratio)
BIncidence rate
COdds ratio
DAttributable risk
Case-control studies sample on outcome status, not exposure, so the marginal totals for disease status are fixed by design. This means you cannot estimate the incidence of disease (and therefore cannot compute a risk ratio directly). However, you can estimate the odds ratio, which approximates the risk ratio when the outcome is rare (the rare-disease assumption). This is the fundamental statistical consequence of the case-control sampling scheme.
Question 3 True / False
A cross-sectional study finds that people who exercise regularly have lower rates of depression. This proves that exercise prevents depression.
TTrue
FFalse
Answer: False
Cross-sectional studies measure exposure and outcome at the same time point, making it impossible to establish temporal sequence. The association could reflect reverse causation (depression reduces motivation to exercise), confounding (a third factor like socioeconomic status drives both), or indeed a causal effect. Cross-sectional designs can generate hypotheses and estimate prevalence, but they cannot establish causation because they lack the temporal ordering that is a necessary (though not sufficient) condition for causal inference.
Question 4 Short Answer
Why does a cohort study provide stronger evidence for causation than a case-control study, even though both are observational?
Think about your answer, then reveal below.
Model answer: A cohort study establishes temporal sequence by enrolling participants based on exposure status before the outcome occurs, then following them forward. This confirms that exposure preceded outcome — a necessary condition for causation. Case-control studies identify people who already have the outcome and look backward at exposure, making temporal ordering ambiguous. Additionally, cohort studies can directly estimate incidence rates and risk ratios, while case-control studies can only estimate odds ratios.
Temporal sequence is one of the Bradford Hill criteria for causation. Cohort designs inherently satisfy it because they define exposure first and observe outcomes later. Case-control studies reconstruct exposure history retrospectively, introducing recall bias and making it harder to confirm that exposure truly preceded disease onset. However, cohort studies are more expensive and take longer, which is why case-control designs remain essential for studying rare diseases.