Active comparator and placebo trials: Why the comparison group shapes the result

By MedClinRes.org Editorial Team

A placebo-controlled trial asks whether a treatment does more than an inactive look-alike. An active-comparator trial asks how a treatment performs next to one already known to work. The comparison group decides which of those questions a result can answer, so the same improvement figure can mean different things in different trials. Before weighing a result, find out who the treatment was compared with.

This article is general education, not medical advice. It has not been peer-reviewed or clinically reviewed. It does not evaluate any product and cannot tell you which treatment is right for you; that decision belongs with you and your own clinician. See our Medical Disclaimer.

What Each Comparison Group Is

Clinical trials divide participants into groups called arms. A placebo comparator arm receives a placebo, which the NIH describes as an inactive product that resembles the test product but has no treatment value. The NIH adds that comparing a new product with a placebo can be the fastest and most reliable way to show whether it is effective.

An active comparator arm receives a treatment considered effective, most often one already used in clinical care. In the ClinicalTrials.gov glossary definition, that arm's outcomes are compared with the experimental arm's to judge whether the studied therapy works better or is safer than the current one.

The ICH E10 guidance on control groups, adopted by the FDA, explains why a control group is used at all: it helps separate the effects of the test treatment from disease progression, expectations of patients and observers, and other treatment. It names randomization and blinding as the main tools for keeping the groups comparable and limiting bias.

Where the Two Designs Diverge

What the control group shows. According to ICH E10, a placebo arm accounts for more than the placebo response. It also accounts for the natural course of the condition, regression to the mean and other therapy. An active-comparator arm shows how an established treatment performed in that trial, but it does not show what would have happened with nothing active.
What a successful result proves. ICH E10 says a significant difference from placebo can demonstrate efficacy without relying on outside information. A trial against an active comparator often aims for non-inferiority, meaning the new treatment is not worse than the comparator by more than a margin set in advance. The guidance warns that a non-inferiority result may not prove efficacy on its own, because if the trial was unable to tell effective treatments from ineffective ones (what the guidance calls assay sensitivity), an ineffective drug could still look “not worse” than the comparator.
When a placebo can be used. The NIH states that placebos are not used when participants would be put at risk by going without effective therapy, particularly in serious illness, and participants are told if a placebo is used. ICH E10 lists the same ethical limit, and where effective treatment exists, active-comparator designs are the usual alternative.
Participant expectations. A review by Enck and colleagues discusses meta-analytic evidence that comparator trials, where everyone expects to receive an active treatment, may produce higher response rates than placebo-controlled trials. That evidence centers on antidepressant and irritable bowel syndrome trials, so it should not be assumed for every condition.

A Worked Comparison: Two Headlines About the Same Hypothetical Tablet

The numbers below are invented for illustration. They do not come from any real study, product or condition.

Imagine a reader sees two headlines about “Compound Q,” a made-up tablet for a made-up symptom. Each trial enrolled 100 people per arm and counted how many reported meaningful symptom improvement.

Trial 1: Compound Q versus placebo

Placebo arm: 40 of 100 improved.
Compound Q arm: 55 of 100 improved.

The placebo arm shows that 40 of 100 people improved with no active ingredient. The 15-person gap is the part of the improvement this trial links to Compound Q. Whether that gap is large and certain enough to matter depends on the statistical analysis and on how the trial defined “improved.”

Trial 2: Compound Q versus an established treatment

Established treatment arm: 52 of 100 improved.
Compound Q arm: 54 of 100 improved.

These numbers suggest Compound Q performed about as well as the established treatment. But with no inactive arm, the trial cannot say how much of that improvement came from either treatment and how much would have happened anyway. If the trial was unable to separate working treatments from non-working ones in that group of participants, a treatment that did little could still produce numbers like these. Per ICH E10, whether a non-inferiority result can be trusted rests on historical evidence that the comparator beat placebo in similar trials, plus careful trial conduct. A reader would need the full trial report and that background research to judge it.

The comparison trap

It is tempting to set 55 beside 54 and conclude the two trials agree. They do not support that. The participants, definitions of improvement, follow-up lengths and expectations may all differ, so results from separate trials with different comparison groups should not be lined up as if measured on the same scale.

How to Read a Trial Report With the Comparison Group in Mind

Name the arms. Look in the abstract or registry record for arm labels such as experimental, placebo comparator or active comparator. Our guide to matching a study with its registry record explains how to find that record.
Identify the stated goal. Is the trial testing whether the treatment is better, or whether it is not worse by a set margin? The primary outcome and analysis plan should say. Our explainer on primary and secondary outcomes covers what a trial was built to measure.
Check what the comparator was. If an active comparator was used, note which treatment, at what dose and for how long. The comparison is only as informative as the comparator was.
Look at who was enrolled. A comparison describes only the people in the trial. See inclusion and exclusion criteria for why that matters.
Follow the participants. A CONSORT participant-flow diagram shows how many people entered, left and were analyzed in each arm.
Check whether the plan changed. Arms, outcomes and analyses can be amended. See how to check what changed before you rely on a result.
Read the safety data for each arm. An adverse event table records what happened to one specific group of participants, not a verdict on safety.

Limits of This Comparison

These are not the only designs. ClinicalTrials.gov also lists sham comparator arms, among other arm types, and each answers a different question.
A placebo arm does not make a trial good, and an active comparator does not make it weak. Quality also depends on randomization, blinding, sample size, outcome choice, follow-up and reporting.
The worked example is invented. It shows how to read a design, not how any real treatment performs.
The evidence on participant expectations across designs comes from a review focused on a few conditions, so it is not a rule for every trial.
This article draws on three sources and did not review any specific trial report. Nothing here tells you whether a treatment is safe or effective for you. Discuss treatment decisions with a physician or pharmacist.

The Takeaway

The comparison group defines what a result is a result of. Against placebo, a gap points to an effect beyond an inactive product. Against an active comparator, similar numbers suggest comparable performance, but they leave open how much either treatment adds beyond doing nothing, unless the comparator has already been shown to beat placebo in similar trials. When you read a trial, ask who the treatment was compared with before asking how big the number was.

References

National Institutes of Health. The Basics. In: NIH Clinical Research Trials and You. Page last reviewed April 24, 2025.
ClinicalTrials.gov Glossary of Common Site Terms, definitions of “Active comparator arm,” “Placebo comparator arm” and “Arm,” as reproduced in the NCATS Toolkit for Patient-Focused Therapy Development glossary, National Center for Advancing Translational Sciences, NIH.
International Council for Harmonisation. E10 Choice of Control Group and Related Issues in Clinical Trials. Guidance for Industry, U.S. Food and Drug Administration, May 2001.
Enck P, Klosterhalfen S, Weimer K, Horing B, Zipfel S. The placebo response in clinical trials: more questions than answers. Philosophical Transactions of the Royal Society B. 2011;366(1572):1889-1895. doi:10.1098/rstb.2010.0384.

Sources were read on October 9, 2026. ClinicalTrials.gov definitions were read through the NCATS glossary. Corrections are handled under our Corrections Policy. For more on reading studies, see How to Read Health Research.