Comparator

The "C" in PICO represents the comparison intervention or control group

With PICO, the "C" denotes an important part of designing a study to test treatments. A central utility of research design is to test the efficacy of an intervention, exposure, test, or agent against a control group to see if it has any clinically meaningful effect above other treatments or nothing at all.  

When using PICO in diagnostic studies, the comparator or control is the "gold standard" method by which a disease state is detected in the current clinical environment.

Comparison or control groups are also necessary when trying to understand the magnitude, variance, and direction of a treatment effect.

Baseline or preliminary measures of an outcome can also be considered the "C" in PICO when used in within-subjects designs.

How to choose control participants

  • If at all possible, use propensity score matching algorithms to effectively match controls with cases.
  • Choose controls that have similar demographic, prognostic, and clinical characteristics to the treatment group.
  • Explicitly define the comparison group, intervention, exposure, test, or agent and how it is applied on the research design
  • Collect twice as many (2:1), three times as many (3:1), or four times as many (4:1) controls versus cases to improve upon the precision and accuracy of research findings.
  • Clearly explain the methodology undertaken to select controls and the control intervention itself (active or inactive).
Click on the Outcome button to continue.
Outcome
Back to Intervention
Research Engineer Home Page

Hire A Statistician - Statistical Consulting for Students

$75.00
Buy Now

DO YOU NEED TO HIRE A STATISTICIAN?

Eric Heidel, Ph.D. will provide the following statistical consulting services for undergraduate and graduate students at $75/hour. Secure checkout is available with Stripe, Venmo, Zelle, or PayPal.


  • Statistical Analysis 
  • Research Design
  • Sample Size Calculations
  • Diagnostic Testing and Epidemiological Calculations
  • Survey Design and Psychometrics
When writing a research question, the comparator should have an explicit definition in terms of being active or inactive. Propensity score matching should be used if possible and participants should have similar demographic, prognostic, and clinical characteristics. A control-to-case ratio of 2:1, 3:1, or 4:1 can be used to increase statistical power.