Observational studies

Observational studies are non-interventional research designs used to examine associations between exposures and outcomes in real-world settings. They are essential for studying exposures that are unethical or impractical to randomise (e.g. smoking, air pollution).

1. Case-control study

Definition: A retrospective, observational study that starts with individuals who already have a specific outcome (cases) and compares their previous exposures to individuals who do not have the outcome (controls).

Main outcome measure: Odds Ratio

Advantages::

• Efficient for rare outcomes or diseases
• Relatively quick and inexpensive
• Suitable when outcomes occur long after exposure
• Can investigate multiple exposures

Disadvantages:

• Prone to recall bias (retrospective self-reporting)
• Prone to selection bias and confounding
• Cannot calculate incidence or absolute risk
• Temporal relationship between exposure and outcome may be unclear
• Unsuitable for rare exposures

---

2. Cohort study

Definition: A longitudinal observational study that follows exposed and unexposed groups over time to assess the incidence of outcomes.

There are two main types of cohort studies:

• Prospective Cohort: Follows participants forward in time
• Retrospective Cohort: Uses historical records for both exposure and outcome

Main Outcome Measure: Relative Risk (Risk Ratio)

Advantages:

• Determines incidence and absolute/relative risk
• Clear temporal relationship (especially prospective)
• Suitable for studying rare exposures
• Can assess multiple outcomes from one exposure

Disadvantages:

• Time-consuming and expensive (especially prospective)
• Large sample sizes needed for rare outcomes
• Prone to loss to follow-up bias
• Prone to confounding
• Less efficient for rare diseases

---

3. Cross-sectional study

Definition: Observes a population or representative sample at a single point in time to assess exposure and outcome status.

Main Use: Estimate prevalence, define normal ranges, describe disease burden

Advantages:

• Quick, simple, and inexpensive
• Can study multiple outcomes
• Good for generating hypotheses
• Useful for establishing population health benchmarks

Disadvantages:

• Cannot determine causation or temporal relationships
• Not suitable for rare conditions
• Cannot assess changes over time (no trend data)
• Susceptible to recall and confounding bias

---

4. Ecological study

Definition: Uses group-level data (e.g., countries, regions) to examine associations between exposure and outcome.

Main Use: Hypothesis generation, trend analysis

Advantages:

• Quick and low-cost
• Useful for examining population-level effects
• Can detect large-scale associations

Disadvantages:

• Subject to ecological fallacy (associations at group level may not apply to individuals)
• Prone to confounding
• Cannot confirm causality

---

5. Case report / Case series

Definition: Descriptive studies that detail one (case report) or more (case series) instances of a clinical presentation or exposure.

Advantages:

• Fast and low cost
• Useful for recognising new or rare conditions
• Good for early signal detection or hypothesis generation

Disadvantages:

• No control group
• Cannot quantify risk or incidence
• Cannot infer causality
• High risk of selection and reporting bias

---

6. Interventional study (for comparison)

Definition: Participants are randomly allocated to intervention or control arms. Used to test the efficacy or safety of treatments.

Main outcome measures: Risk reduction, number needed to treat (NNT), relative risk, etc.

Advantages:

• Gold standard for determining causality
• Allows control of confounding through randomisation
• Enables meta-analyses and evidence synthesis

Disadvantages:

• Expensive and time-consuming
• May raise ethical concerns
• Results may not always generalise to real-world settings

---

Study comparison table