Skip to content

Case Cohort Study

  • Observational design that analyzes a subset of a larger cohort: all cases plus a random sample of non-cases.
  • Less costly and faster than a full cohort study while aiming to detect exposure–outcome associations.
  • Can be subject to selection bias and is not well suited for rare outcomes.

A case-cohort study is a type of observational study in which a subset of individuals from a larger population, known as the cohort, are selected for analysis. The subset consists of individuals who have developed a specific outcome of interest (cases) and a random sample of individuals from the cohort who have not developed the outcome (non-cases or controls).

In a case-cohort design the starting point is a defined cohort. From that cohort, researchers select all individuals who developed the outcome of interest (cases) and a randomly chosen sample of cohort members who did not develop the outcome (non-cases). Data on exposures or risk factors are collected for both groups and analyzed to assess whether an association exists between the exposure and the outcome. Compared with a full cohort study, a case-cohort study requires data collection on fewer individuals, reducing cost and time. Compared with a traditional case-control study, it can be more efficient at detecting associations because it is anchored in the original cohort.

Cohort: all individuals who have ever smoked cigarettes.
Selected: a sample of individuals who have developed lung cancer (cases) and a random sample of individuals who have not developed lung cancer (non-cases or controls).
Data collected: cigarette smoking habits, such as the number of cigarettes smoked per day and the duration of smoking, for both cases and non-cases.
Analysis: determine whether there is a statistically significant association between cigarette smoking and the development of lung cancer.

Cohort: all individuals who are obese.
Selected: a sample of individuals who have developed type 2 diabetes (cases) and a random sample of individuals who have not developed type 2 diabetes (non-cases or controls).
Data collected: obesity-related factors, such as body mass index (BMI) and waist circumference, for both cases and non-cases.
Analysis: determine whether there is a statistically significant association between obesity and the development of type 2 diabetes.

  • Examining associations between exposures and outcomes within a defined population while reducing the resources required compared with a full cohort study.
  • Selection bias: the cases and non-cases may not be randomly selected from the entire cohort, which can lead to overestimation or underestimation of associations.
  • Not suitable for rare outcomes: there may not be enough cases to adequately power the study.
  • Cohort study
  • Case-control study
  • Case (outcome)
  • Control (non-case)