Skip to Main Content

Public Health: Study Types

Observational & Experimental Studies

Observational Studies:

  • The researchers "allow nature to takes its course" by measuring but not intervening. 
  • They can be descriptive or analytical.
    • A descriptive observational study describes the disease in a population.
    • An analytical observational study goes a step further by analyzing the relationship between health and other variables. 

Experimental Studies:

  • Experimental Studies, also called intervention studies, intervene in an active attempt to change an exposure, behavior, or disease. 

Case Control (Observational)

  • Starts with disease and goes backwards in time looking for the exposure.
  • Compares a group with the disease and without the disease.
  • The main outcome is odds ratio.
  • Better for rare diseases.

Advantages:

  • They are fairly quick and cheap.
  • Usually are the most important way to investigate rare diseases.
  • They don't require a huge number of subjects.
  • You investigate multiple causes of a disease.

Disadvantages:

  • Often relies on a person remembering things which can lead to recall bias.
  • Only investigate cases that have been identified or diagnosed.
  • Selecting controls relies on the researcher's judgment.
  • Cannot determine the rate or risk of a disease in the exposed and nonexposed.
  • Cannot prove a cause and effect.

Case Study/Case Series (Observational)

  • Purely descriptive.
  • Often describe a new disease with an unclear cause.
  • No control group
  • Considered the lowest level of evidence, but also the first line of evidence
  • Do not address causality
  • Do not provide prevalence

Cohort (Observational)

  • Starts with exposure
  • Looks at a group over time.
  • Grouped together because there is a common experience or exposure.
  • Looks for relative risk.
  • Looks for exposure and then disease.
  • Does not work with rare diseases.

Advantages:

  • They are the only study type that can determine the absolute risk of contracting a disease. 

Cross-Sectional (Observational)

  • Survey of a population at a single point in time (usually a day or less than month) -- a "snapshot" in time.
  • Used for population- based surveys to determine the prevalence of disease.
  • Measures the outcome and the exposure(s) in the population at the same time.
  • Frequency of disease and risk factors are identified.
  • The main outcome is prevalence.
  • Can't determine relative risk or odds ratio.*
  • Do not address causality.

* You can estimate OR.

OR v RR

Ranganathan, Priya, Rakesh Aggarwal, and C. S. Pramesh. “Common Pitfalls in Statistical Analysis: Odds versus Risk.” Perspectives in Clinical Research 6, no. 4 (2015): 222–24. https://doi.org/10.4103/2229-3485.167092.

Abstract:

"In biomedical research, we are often interested in quantifying the relationship between an exposure and an outcome. “Odds” and “Risk” are the most common terms which are used as measures of association between variables. In this article, which is the fourth in the series of common pitfalls in statistical analysis, we explain the meaning of risk and odds and the difference between the two."

Ask yourself: Do we know the total number of people at risk? If we don't have the total number of people exposed, like in retrospective studies, you can't calculate RR. 

Epidemiological Studies - made easy!

Crossover (Experimental)

  • Patients in a crossover study will "crossover" to another treatment arm during course of the trial.
  • This means that even if they are initially put into a placebo arm, they will also eventually receive the study drug or standard of care during the trial.
  • Data integrity is ensured by instituting a washout period—a gap in between arms where patients don't receive any medication—in order to reduce carryover effects from the previous treatments.
  • Crossover studies are often used when researchers feel it would be difficult to recruit participants willing to risk going without a promising new treatment.

Meta-Analysis (Experimental)

A meta-analysis is a statistical technique that allows researchers to combine the findings of multiple independent studies. It is the type of study that is most frequently used to assess the clinical effectiveness of a healthcare intervention and provides the highest level of evidence of the choices given.

Meta-analysis would be used for the following purposes:

  • To establish statistical significance with studies that have conflicting results. 
  • To develop a more correct estimate of effect magnitude To provide a more complex analysis of harms, safety data, and benefits.
  • To examine subgroups with individual numbers that are not statistically significant.
  • Provides the highest level of evidence for efficacy of a drug.

Randomized Controlled Trial (Experimental)

  • A study design that randomly assigns participants into an experimental group or a control group.
  • As the study is conducted, the only expected difference between the control and experimental groups in a randomized controlled trial (RCT) is the outcome variable being studied.
  • You must have an intervention and a comparison group.

* Randomized double blind placebo control studies are the "Gold Standard" research design in intervention based studies.

What to Look For

Hierarchy of Evidence