Cohort studies

Cohort Studies - The Time Traveler's Log

An observational design that follows a group of people (a cohort) forward in time. It compares outcomes between individuals exposed to a factor and those unexposed. Think of it as watching a movie unfold from a specific starting point.

• Process: Start with a disease-free population → group by exposure status → follow over time → compare disease incidence.

⭐ Best observational design for establishing temporality (i.e., the exposure truly precedes the outcome).

Types of Cohorts - Future vs. Past

• Prospective Cohort Study

  • Starts in the present and follows a group of individuals (cohort) forward in time to observe who develops the outcome.
  • Pros: Better control over data collection, can establish incidence.
  • Cons: Expensive, time-consuming, and inefficient for rare diseases.

• Retrospective (Historical) Cohort Study

  • Uses existing data (e.g., medical records) to identify a cohort and trace them forward from a past exposure to a subsequent outcome.
  • Pros: Quick and inexpensive.
  • Cons: Less control over data quality; prone to confounding.

⭐ A key advantage of prospective studies is their ability to directly calculate incidence rates (new cases over a period), which is not possible with retrospective designs.

Risk Calculation - Crunching the Numbers

-   Formula: $RR = [a/(a+b)] / [c/(c+d)]$
-   Interpretation: RR > **1** (increased risk), RR < **1** (decreased risk), RR = **1** (no association).

• Attributable Risk (AR): The difference in risk between exposed and unexposed groups.
  • Formula: $AR = [a/(a+b)] - [c/(c+d)]$

⭐ Relative risk is the primary measure of association for cohort studies.

Pros & Cons - A Balancing Act

• Strengths:

  • Can establish temporality (exposure → outcome).
  • Allows calculation of incidence and relative risk ($RR$).
  • Excellent for studying rare exposures.
  • Can investigate multiple outcomes simultaneously.

• Weaknesses:

  • Inefficient and costly for rare diseases.
  • Long duration (especially prospective) leads to high costs and potential for loss to follow-up (attrition bias).
  • Susceptible to confounding variables.

⭐ The key strength of cohort studies is their ability to establish temporality, a crucial element for inferring causality.

Common Biases - The Study Spoilers

• Selection Bias: Groups differ systematically at baseline. Particularly in retrospective cohorts due to non-random selection of records (e.g., healthy worker effect).
• Loss to Follow-up (Attrition) Bias: Participants who drop out differ from those who remain. E.g., if sicker patients leave one group, the outcome is skewed.
• Confounding Bias: An external variable is associated with both the exposure and the outcome, distorting the true relationship.

⭐ Loss to follow-up >20% is considered high and seriously threatens the study's validity.

High‑Yield Points - ⚡ Biggest Takeaways

• Cohort studies can be prospective (forward-looking) or retrospective (using past records).
• They follow groups with and without an exposure over time to see who develops the disease.
• The key measure of association is Relative Risk (RR).
• Excellent for establishing temporality (exposure precedes outcome) and calculating incidence.
• Major limitations: expensive, time-consuming, and inefficient for rare diseases.
• Particularly susceptible to selection bias and loss to follow-up bias.