Epidemiology · Reference
What is a cohort study?
A cohort study is an observational design that follows a group of people, classified by their exposure, over time to see who develops an outcome. Because it observes new cases as they arise, it can measure incidence and estimate relative risk.
How a cohort study works
A cohort study begins by identifying a group of people who do not yet have the outcome of interest and classifying them according to their exposure — for example, exposed versus unexposed to a particular factor. The cohort is then followed over time and the occurrence of the outcome is recorded in each group. Because the design moves from exposure to outcome and observes incidence directly, it can calculate the risk in each group and divide them to obtain a relative risk. The defining feature is following people forward through the natural history of the outcome.
Prospective and retrospective cohorts
Cohort studies come in two temporal forms. In a prospective cohort, exposure is assessed at the outset and participants are then followed forward in time as outcomes accumulate. In a retrospective (or historical) cohort, the investigator uses existing records to define a cohort and its exposures in the past and traces outcomes that have already occurred up to the present.
Both share the same logic — grouping by exposure and comparing outcome frequency — and differ only in whether the outcomes have happened yet when the study begins. Retrospective cohorts are quicker and cheaper but depend on the quality of pre-existing records.
Strengths and limitations
Cohort studies have notable strengths: they establish that exposure preceded outcome, can measure incidence and absolute risk, and can examine several outcomes of a single exposure. They are well suited to studying relatively common outcomes and to exposures that are rare or difficult to assign. Their main limitations are practical: prospective cohorts can require long follow-up and large samples, making them expensive and vulnerable to participants dropping out (loss to follow-up). As observational studies, they remain susceptible to confounding and to selection bias.
Place in the evidence hierarchy
Among observational designs, well-conducted cohort studies are generally regarded as providing stronger evidence about causes than case-control or cross-sectional studies, because the temporal sequence from exposure to outcome is clearer. They sit below randomised controlled trials in conventional hierarchies, since exposure is observed rather than assigned, but they are often the most feasible and ethical option for studying harmful or unmodifiable exposures. Transparent reporting following the STROBE guidelines helps readers judge a cohort study’s validity.
Key facts
At a glance
- Type: Observational analytic study
- Direction: Exposure → outcome (followed over time)
- Measures: Incidence and relative risk
- Two forms: Prospective and retrospective (historical)
- Key limit: Time, cost and loss to follow-up; confounding
Common questions
FAQ
What is the difference between a prospective and a retrospective cohort study?+
In a prospective cohort, exposure is measured at the start and participants are followed forward as outcomes occur. In a retrospective cohort, the investigator defines the cohort and its past exposures from existing records and traces outcomes that have already happened. Both group people by exposure and compare outcome frequency.
What can a cohort study measure that a case-control study cannot?+
Because it follows people from exposure to outcome and observes new cases, a cohort study can measure incidence and calculate absolute risk and relative risk directly. A case-control study starts from outcome and generally yields an odds ratio instead.
What are the main limitations of cohort studies?+
Prospective cohort studies can require long follow-up and large samples, making them costly and prone to loss of participants over time. As observational designs they are also vulnerable to confounding and selection bias, so associations they find are not automatically causal.
The step most authors miss
Doing CRediT right? Don’t stop at the statement.
A CRediT statement credits you inside one paper. The recognition CRediT was built for happens when those roles are tied to you, persistently. Sign in with your ORCID — free — and claim your CRediT contributions on casrai.org, the home of the standard. They become a verified, portable part of your identity, not a line that disappears into one PDF.
Free: claim your contributions, then export a journal-ready CRediT statement, schema.org structured data, JATS XML, CSV or BibTeX — and preview your public profile. A membership publishes that profile publicly and verifies the journals you serve.
